STATEMENT OF WORK

Title: Daisy Drive Systems for the Precise Alteration of Local Populations

1.0 SCOPE

The Recipient will develop self-limiting “daisy drive” technology into a platform capable of safely, efficiently, and reversibly editing local subpopulations of target organisms. They will construct modular daisy drive systems that are easily adapted in the future for alteration, suppression, or reversal in three key target species relevant to public health and conservation: Aedes aegyptiCulex quinquefasciatus, and (optionally) Anopheles gambiae. Working primarily in nematodes inside contained facilities, they will build and test different configurations of daisy drive systems, examine their evolutionary stability and dynamics in large populations, and observe their spread through linked meta-populations configured to match those of mosquito species of interest. During laboratory experiments, the Recipient will demonstrate daisy drives that can overwrite every copy of a “rogue” gene drive, and then restore the population to wild type. The Recipient will also develop centralized, high-throughput transgenesis systems. Although the work pursued as part of this effort will not include any form of field trial or open release, risk assessments and technology development will be conducted in collaboration with local communities and regulators, and scrutinized by ethicists, to improve the relevance, effectiveness, and safety of the technologies developed and the data collected.

1. BACKGROUND

Among existing options for large-scale gene editing, standard CRISPR-based drive systems are the easiest to construct and most likely to function across species. However, because they are anticipated to spread to every population of a target species, and not just within a desired subpopulation, they are challenging to build, test, and deploy in a safe and responsible manner. Yet there is no existing way to exclusively alter a local subpopulation of wild organisms save by near inundating the environment with engineered organisms. Use of this technique requires ~100-fold more organisms than proposed daisy drive systems would require. Other strategies that have been discussed by scientists are theoretically less effective and have not been demonstrated.

The recipient proposes to develop daisy drives, which function by harnessing Mendelian inheritance to programmably eliminate components of the daisy drive system until the drive is no longer active. Daisy drives have the potential to exhibit controlled geographic and spatial spread due to the serial loss of daisy elements over generations due to inheritance factors and natural selection. For example, in a C→B→A drive, B and A can drive but C does not. Loss of C causes B to cease driving and its subsequent loss prevents the cargo element A from driving and eventually be lost.This strategy has never before been used and has no known natural counterpart. Because daisy drives would rely on CRISPR genome editing, theoretically they could mimic any effect achievable with a standard RNA-guided gene drive system, but at a local and temporary scale, affording advantages for safety testing, local community decision-making, and regulation by governing bodies. Daisy drives may be capable of precise and complete genetic removal of rogue genetic elements from populations. Studying gene drive evolutionary dynamics in huge populations of nematodes will allow empirical testing of drive system stability and safety over time, a likely prerequisite for future regulatory approval. Additionally, early engagement of local communities that might in the future benefit from daisy drive deployment will allow experiments to be shaped to collect data that address stakeholder concerns.

2.0 APPLICABLE Documents

2.1 DARPA-BAA-16-59.

3.0 PROJECT WORK DESCRIPTION AND REQUIREMENTS

Animal use is not anticipated in this effort.

The effort may not create any gene drive system anticipated to spread indefinitely in a wild population. All experiments will employ at least two forms of stringent confinement as recommended by the most stringent published laboratory safety guidelines.

The Recipient shall support data sharing with other performers including sharing of raw and processed experimental data, processing methods used, algorithms used to process the data, artifact information, research reports, and software including source code and executables. The Recipient shall deliver to DARPA, in the form identified by the DARPA Program Manager (PM), all the experimental data collected in this effort within four months of completion of data collection (or at a date agreed upon in the statement of work), as agreed by the PM. Experimental data should be accompanied with sufficient metadata to capture laboratory conditions, equipment, supplies/reagents, procedures, and methodology, and will be presented in a manner that provides enough description for a third party to conduct appropriate analysis and interpretation of data. DARPA, for the purposes of scientific research, may choose to share this data with other performers in the program, with a third party, and possibly with a larger research community in the future. Data sharing will be included in the project milestones and reported regularly to DARPA.

There exists one major exception: technical specifications of the high-throughput transgenesis systems to be developed by Hive Biosystems are proprietary and will be protected as such. The details of these systems will therefore be disclosed to DARPA only and kept private to the extent allowable under DARPA contracts with private commercial entities.

Activities to address ethical, legal, and social issues (ELSI) are anticipated in this effort. The Recipient shall support such activities with DARPA in an expanded format hereafter referred to as LEEDR (Legal, Ethical, Environmental, Dual-use, and Responsible innovation) including semi-annual teleconference calls with the LEEDR Panel and regular feedback from the Panel regarding the recipient’s activities. LEEDR activities will be reported regularly to DARPA.

The Recipient shall engage with relevant regulatory bodies and local communities to identify and mitigate challenges to transition and, as appropriate, eventual deployment of the resulting technology.

3.1 BASE PERIOD (PHASE 1)

Technical Area 1 - Control of genome editing activity

3.1.1 Mathematical Modelling (TA1 Task 1)

The Recipient shall:

3.1.1.1 Thoroughly analyze baseline, deterministic well-mixed (DW) daisy drive model.

3.1.1.1.1 Understand sensitivities of CRISPR/repair parameters. (Subtask 1.1.1)

Develop a model that enhances the- understanding of the effects of key CRISPR parameters (cutting efficiency and number of sgRNAs) and cellular repair parameters (homologous recombination and nonhomologous end-joining rates) on daisy drive system fitness.

3.1.1.1.2 Understand sensitivities of fitness parameters. (Subtask 1.1.2)

Study the sensitivities of daisy element fitness costs as well as effects that are recessive, co-dominant, and dominant.

3.1.1.1.3 Understand sensitivities of population-level parameters.

Study sensitivity of initial release frequency of daisy drive organisms.

Performance Metrics and Assessment: In sub-sub tasks 3.1.1.1.1 through 3.1.1.1.3, sensitivities of daisy drive parameters will be calculated with respect to three model outputs: (i) maximum cargo spread, (ii) time to 0.99 frequency, and (iii) time above 0.99 frequency, with respect to the parameters named in each sub-sub task.

Deliverables: Precise analysis of daisy drive parameter space, including an understanding of which parameters are most critical for daisy drive success and quantification of targets for later experimental optimization.

3.1.1.2 Analyze the importance of off-target cutting.

3.1.1.2.1 Understand sensitivities of drives to off-target cutting.

Explicitly model fitness penalties resulting from off-target cuts as a function of cutting frequency and fitness cost distribution for dominant and recessive phenotypes. Measure robustness of maximum cargo frequency to off-target cutting parameters (off-target cutting rate, fitness cost distribution, dominance).

3.1.1.2.2 Understand the effects of driven off-target changes.

Determine the parameter regime in which daisy drive results in spread of driven off-target mutations to an appreciable frequency (which is defined as greater than the typical standing genetic variation for a target species in the absence of drive, roughly 10-3). This will depend on the rates of off-target cutting and subsequent mutation and the corresponding fitness cost distribution.

Deliverables: Determination of whether off-target cutting at low frequencies is ever problematic enough to require experimental measurement for a given drive system (where problematic is defined as leading to spread of off-target changes to a frequency greater than the typical standing genetic variation for a target species in the absence of drive).

3.1.1.3 Evaluate the importance of existing or created drive-resistant alleles.

3.1.1.3.1 Extend basic daisy drive model (DW) to include resistant alleles (DW+R).

Extend the deterministic, well-mixed model to include resistant alleles generated by standing variation and nonhomologous end-joining (NHEJ). Measure the robustness of maximum cargo frequency, time to 0.99 frequency, and time above 0.99 frequency with respect to standing variation and NHEJ-mediated resistance with respect to design parameters, including number of elements, fitness effects of resistance, and number of sgRNAs per element.

Deliverables: An extension of the DW model that includes resistance (DW+R), along with characterization of the robustness of cargo spread to resistance. If resistance leads to a qualitative change in dynamics (>0.2 cargo frequency difference after 20 generations between DW and DW+R models) for biologically reasonable values of cellular parameters (cutting efficiency, HR and NHEJ rates), determine targets for experimental optimization of parameters (number of elements, fitness cost of resistance, number of sgRNAs per element, NHEJ rates).

3.1.1.4 Study daisyfield and underdominance architectures.

3.1.1.4.1 Extend the DW model to include an arbitrary number of parallel daisy elements.

3.1.1.4.2 Extend the DW model to include underdominance at the cargo element.

3.1.1.4.3 Iteratively refine the DW model based on experimental results.

*Deliverables:*Model extensions including daisyfield (DW+F) and quorum/underdominance (DW+Q) architectures.

3.1.1.5 Construct models for multiple-population nematode experiments.

3.1.1.5.1 Metapopulation extensions of previous models.

Extend previous models (DW, DW+R, DW+F, DR+Q, DW+S, DW+IR) to include Ndistinct populations connected by gene flow to mimic and inform nematode experiments by allowing arbitrary connectivity between the populations to be modelled, such as adjacent flasks with liquid transfer or grids of populations under control of a liquid-handling robot. Each metapopulation model will be completed shortly after the corresponding well-mixed model (DW: 12mo, DW+R: 15mo, DW+F: 18mo, DW+Q: 20mo, DW+S: 21mo, DW+IR, 24mo).

3.1.1.5.2 Determine spatial parameter regimes for experimentation*.*

Use metapopulation models to determine the effect of gene flow rates on daisy element containment. Understand and predict the robustness of containment with respect to variations in gene flow. Determine gene flow regimes for experimental testing (specifically, three regimes: (i) low mixing, where strong containment is anticipated, (ii) medium mixing, where the cargo is anticipated to surpass its initial population release frequency in adjacent populations, and (iii) high mixing, where all populations are anticipated to behave similarly).

3.1.1.5.3 Iteratively refine metapopulation models based on experimental results.

Deliverables: Models that directly connect earlier insights to nematode work, thereby informing experiments and refining models.

Technical Area 2 – Population suppression and genetic remediation.

3.1.2 Mathematical Modelling (TA2 Task 1)

The Recipient shall:

3.1.2.1 Construct genetic remediation models in well-mixed populations.

3.1.2.1.1 Extend the DW model to encompass population suppression (DW+S).

Extend the DW model to include two distinct sexes. Evaluate the efficacy of genetic load (e.g., targeting a female fertility gene for disruption) for different parameter sets.

3.1.2.1.2 Extend the DW model to include immunizing reversal (DW+IR).

Extend the DW model to include a rogue global drive and amend the daisy drive dynamics such that the daisy drive acts as a global reversal agent against the rogue drive and a daisy immunizing agent in the wild-type population.

*Deliverables: *Model extensions (DW+S, DW+IR) which assess genetic remediation strategies in well-mixed populations.

3.1.2.2 Assess genetic remediation strategies in linked populations with gene flow.

3.1.2.2.1 Assess targeted daisy suppression in linked populations.

Employing the DW+S metapopulation model from sub-sub task 3.1.1.5.1, determine effectiveness of daisy suppression for remediation of a rogue gene or a rogue global drive system at different release rates in linked metapopulations.

3.1.2.2.2 Assess daisy immunizing reversal in linked populations.

Employing the DW+IR metapopulation model from sub-sub task 3.1.1.5.1, determine the effectiveness of a daisy drive acting as a reversal agent against a rogue global drive, as well as a daisy immunizing agent in the wild-type population.

Performance Metrics and Assessment: In sub-sub tasks 3.1.2.2.1 and 3.1.2.2.2 performance of remediation strategies will be quantified by absolute frequency of rogue drive 20 generations after release of the remediation drive. Furthermore, robustness of the absolute frequency measure with respect to design and release parameters (drive efficiency, fitness costs, initial release frequency of remediation drive and frequency attained by rogue drive before release of remediation drive).

Deliverables:** **Assessment of genetic remediation strategies in linked, well-mixed populations. Spatial dynamics will be important, so these will serve as important baselines for the subsequent spatially explicit models.

Technical Area 1 - Control of genome editing activity.

3.1.3 Nematodes (TA1 Task 2)

The Recipient shall:

3.1.3.1 Procure whole-genome sequence and assembly of C. brenneri lab strain.

Sequencing and assembly using Chicago HiSeq method (Dovetail Genomics). The genome sequence will be required for targeting, off-target avoidance, and ensuring that daisy drive elements are not genetically linked to one another.

3.1.3.1.1 Sequenced C. brenneri genome with

50x coverage.

Deliverable:** *Assembled genome of C. brenneri with >50x coverage. *

3.1.3.2 Characterize the necessary components for daisy drive.

3.1.3.2.1: Make a marker & docking strain that tests germline transgene expression.

Insert docking sites for recombinase-mediated cassette exchange next to a mCherry-2A-GFP-2A-BFP cassette. Evaluate strains for germline fluorescence or its absence by microscopy. If germline-specific silencing occurs, attempt to “license” the fluorescent genes for expression in germline cells by fusing to genes that are transcriptionally active in germline, inserting genes with native polyA/T-rich regions, or both. Redesign and repeat until robust germline expression of markers is observed. If required, the same technique will be used to license CRISPR nuclease for germline expression.

3.1.3.2.2: Optimize multiplexing via tRNA-processing for SpCas9 sgRNAs.

Microinject nematodes with an SpCas9 expression plasmid and arrays driven by U6 promoter consisting of sgRNAs separated by tRNAs. The sgRNAs shall target 3 different fluorescent markers, and optionally rol-6 and dpy-5 if they can be made to produce dominant phenotypic signals. Quantify efficiency of disruption by measuring the proportion of progeny that are non-fluorescent or those exhibiting visible phenotypes. As a control for guide RNA activity, compare the results to those produced by injecting the nuclease plasmid along with each individual sgRNA driven by its own U6 promoter. Use this method to identify tRNAs suitable for efficient multiplexing of guide RNAs. Goal: >90% cutting of targeted genes per two sgRNAs used; >98% per four sgRNAs; numbers chosen to demonstrate equivalence to yeast, flies, and mosquitoes wherein >90% cutting has been reported.

3.1.3.2.3: Test multiplexing via Cpf1 crRNA-processing.

Using the same methodology as 3.1.3.2.2, test loss of fluorescence or generation of visible phenotype upon delivering an array with Cpf1 crRNAs between SpCas9 guide RNAs along with a plasmid expressing SpCas9-2A-Cpf1. Similarly, test arrays consisting of solely Cpf1 crRNAs. Goal: >90% cutting of target per sgRNA/crRNA pair; >98% per two pairs.

3.1.3.2.4: Make strains expressing SpCas9, Cpf1, or SpCas9-2A-Cpf1.

Use the marker & docking strain from 3.1.3.2.1 to insert a cassette with a constitutive actin promoter driving expression of each nuclease.

3.1.3.2.5: Create transgenic strains to test Pol III promoters.

Use the marker & docking strain from 3.1.3.2.1 to integrate U6/7SK promoters from *C. brenneri and close relatives such as C. elegans, C. briggsae *that drive an array of guide RNAs vs fluorescent genes in order to identify the best Pol III promoters to use in daisy drive systems.

3.1.3.2.6: Quantify Pol III promoter efficacy.

Cross the guide array strains from 3.1.3.2.5 with the nuclease strains from 3.1.3.2.4. Quantify cutting efficacy and thus promoter performance by measuring the incidence of fluorescent or visible marker phenotypes in the offspring. Identify at least three promoters sufficient to cut at

90% efficiency per two guide RNAs.

3.1.3.2.7: Quantify fitness costs from nuclease/guide RNA expression and off-target cuts.

Measure the cost of constitutive expression of CRISPR nucleases combined with guide RNA arrays relative to those imposed by the nucleases or guide RNAs alone. Engineer the docking strain from 3.1.3.2.1 to encode a fluorescent marker, constitutively expressed nucleases, and highly active Pol III-driven arrays of guide RNAs from 3.1.3.2.6. A second set shall encode only the marker and the nucleases, a third set shall encode the marker and guide RNA arrays, and a control strain will only express the fluorescent marker with a mutation that abolishes fluorescence.

Quantify fitness by competing the fluorescent strains with nuclease+guide RNAs or nuclease alone against the non-fluorescent control strain. Competitions shall involve populations of >10,000 worms over 8 generations, with strain frequencies determined by measuring the fraction of the population that is fluorescent. If fitness differences are apparent early and continuing the assay for the full 8 generations provides no additional information, future experiments shall use only as many generations as are informative.

Because modelling suggests that high fitness costs dramatically impair drive performance, cargo costs are likely additive with nuclease costs, and many cargos are likely to be costly, the target cost for nuclease expression (without guide RNAs) must be s < 0.1. If this target cannot be met with constitutive promoters, the Recipient shall subsequently perform equivalent experiments for germline-specific nuclease expression. Cost measurements combined with multiplexing efficacy (3.1.3.2.1/2) will inform the decision to use SpCas9, Cpf1, or both in subsequent experiments.

3.1.3.2.8 Quantify off-target cutting.

If the results of sub-task 3.1.3.2.7 indicate that the addition of guide RNA arrays incurs a substantial fitness cost over and above that of nuclease expression alone (additional s>0.05), determine whether off-target cuts are responsible. Specifically, computationally identify the 10 most likely off-target cuts from each guide RNA, PCR-amplify those sequences, and measure off-target cutting at those loci through SURVEYOR assay or another standard method to pinpoint the likely cause. Should the modelling results from 3.1.1.2 and the results of this study both indicate the off-targets are not a problem unless costly, this will suggest that future gene drive development need only measure the fitness effects of combined nuclease and guide RNA expression relative to either alone, save for the highly unlikely yet theoretically possible in-trans migration of the drive system into an off-target site, which could push the frequency considerably higher and move into other species. This possibility shall be investigated on a case-by-case basis using large-population nematode experiments.

3.1.3.2.9 Create mutants with varying non-homologous end joining (NHEJ) rates by modulating cku-80 expression.

Create a series of mutant worms with differing HDR rates by changing the strength of the Kozak sequence in the native gene and/or employing RNAi to degrade the mRNA. Quantify the effects on HDR.* *

3.1.3.2.10 Measure brood sizes and identify genes to edit to match mosquito reproduction.

Measure brood size of nematode strains intended to match mosquitoes and compare to number of eggs laid by female mosquitoes of the species to be mimicked. Identify genes to be disrupted or edited, starting with those known to affect fertility in C. elegans, in order to adjust brood sizes to match mosquito reproduction.

Deliverables: Methods of expressing at least four guide RNAs from a single Pol III promoter, identifying at least four identified Pol III promoters that can result in >98% cutting, and a ranking of different nucleases by fitness cost with at least one costing s<0.1. At least 3 strains with gene drive inheritance frequencies that differ from wild-type nematodes due to altered expression of cku-80, to be used to match the relevant DNA-repair characteristics of mosquito species for the purpose of studying drive systems intended for other organisms. A comparison of nematode strain brood sizes with those of mosquitoes, and a list of candidate genes to be edited so that brood sizes will roughly match the typical number of eggs laid by mosquitoes.

3.1.3.3 Germline nuclease expression and ribosomal recoding for stability and precision.

3.1.3.3.1 Create 10 strains with recoded ribosomal genes.

Recode ten candidate haplo-insufficient ribosomal genes, including rpS14, rpS16, rpS18, rpS21, rpS22, rpS26. Recodings must remove at least 4 target sites for SpCas9 and 4 target sites of Cpf1. Fluorescent marker genes shall be inserted downstream along with guide RNAs targeting the wild-type gene.

3.1.3.3.2 Evaluate fitness of recoded ribosomal genes.

Compete control strains bearing similar guide RNA cassettes and the same fluorescent marker gene as the recoded ribosomal genes but with a mutation that inactivates fluorescence versus recoded to quantify the fitness cost of recoding. Quantification shall be accomplished by measuring the relative population increase of control nematodes relative to recoded ones over generations by tracking fluorescence.

3.1.3.3.3 Create 5 transgenic strains expressing SpCas9 in germline.

Express the CRISPR nuclease just downstream of target recoded ribosomal genes using promoter/3'UTRs predicted to result in varied developmental expression timing. Focus on 3'UTRs with conserved orthologs to mosquitoes and other species of interest. Top candidates are nos-3, fbf-1, *daz-1, gld-1 *for early germline (useful for maximizing drive system fitness) and him-3 for putative exclusive meiotic expression.

3.1.3.3.4 Create 5 strains expressing Cpf1 in the germline.

Use expression signals for nos-3, fbf-1daz-1, gld-1, him-3 unless given reason to choose others.

3.1.3.3.5 Create 5 strains expressing SpCas9-2A-Cpf1 in the germline.

Use expression signals for nos-3, fbf-1daz-1, gld-1, him-3 unless given reason to choose others.

3.1.3.3.6 Test nuclease activity and ribosomal haplo-insufficiency during gametogenesis.

Cross recoded ribosomal gene strains containing downstream guide RNAs and nuclease-expressing strains. Quantify incidence of marker gene in offspring as well as the total number of offspring produced for crosses relative to each strain crossed to itself to determine fertility-based fitness cost of the drive, if any. Perform tests for both combinations of males+females to determine whether nuclease expression and drive efficiency differs by the sex of the nuclease-encoding parent.

3.1.3.3.7 Test underdominance.

Swap the positions of the two least costly extensively recoded ribosomal genes by simultaneous reciprocal exchange via CRISPR or Cre recombinase. Each gene shall be inserted (including promoter and 3'UTR) in place of the middle of the other's coding sequence, disrupting it. Cross heterozygotes with wild-type worms and quantify offspring fertility and genotypes. Expect ~50% normal fertility with surviving progeny consisting of half heterozygote and half wild-type worms.

Deliverables:** **Strains with germline-expressed nuclease that cut target genes at >98%. A determination of whether targeting haploinsuffiscient genes in the early proliferative germline can increase inheritance to 100%. Identify recoded ribosomal genes with costs <0.05 according to brood size measurements. Demonstrate swapped recoded ribosomal genes that exhibit underdominance when crossed to wild-type worms (50% +/- 5% offspring survival) and might be spread by daisy drive in a daisy quorum system.

3.1.3.4 Build daisy drive and daisyfield systems

3.1.3.4.1 Generate nuclease cargos for daisy drive and daisy field.

Insert the optimal germline-expressing nuclease cassette 3' of the least costly recoded ribosomal gene.

3.1.3.4.2 Generate nuclease cargos for daisy quorum/underdominance.

Insert the optimal germline-expressing nuclease cassette 3' of the swapped recoded ribosomal genes.

3.1.3.4.3 Generate neutral site daisy elements.

Insert Pol III-driven guide RNA cassettes that target 1) the next neutral site daisy element and 2) the cargo ribosomal target gene. Neutral daisy elements are the default because building recoded ribosomal daisy elements may not be possible in every organism and may not be advantageous depending on the parameters (see below).

3.1.3.4.4: Generate recoded ribosomal daisy elements.

Modeling studies in 3.1.1.2 will determine whether drive-resistance alleles that block daisy element copying are a serious problem. Incorporating experimentally determined fitness costs from 3.1.3.3.2 and the efficacy of ribosomal recoding for germline haplo-insufficiency in increasing the effective homing rate from 3.1.3.3.6 into the models will determine whether building daisy elements with recoded ribosomal genes should be more effective than neutral site daisy elements for anticipated release numbers and cargo costs. If the models indicate that the answer is yes, daisy elements shall be created by inserting guide RNA cassettes downstream of recoded ribosomal genes, with the guide RNAs targeting the next daisy element's wild-type ribosomal gene as well as the nuclease cargo's ribosomal gene(s).

3.1.3.4.5: Generate daisyfield elements.

Introduce a guide RNA array targeting the nuclease cargo(s) into repeated regions. Targets will be the repeated sequences to be determined based on genome sequencing/scaffolding.

3.1.3.4.6: Cross daisy elements and cargos to generate intact daisy drive systems.

Build and test inheritance of daisy-chain and daisy-field drive systems by quantifying the frequency of inheritance of different fluorescently labeled elements in the progeny of matings with wild-type worms.

Deliverables:** **At least one daisy-chain drive system with 3+ elements for which the cargo element (encoding the nuclease) is inherited at >90% frequency, the daisy elements are each inherited at >80% efficiency when a prior element is present, and elements are inherited at normal Mendelian frequency when there is no prior element. The fitness cost of the complete daisy-chain drive system as measured by brood size measurements must be <0.2. Each daisy element must encode at least four guide RNAs targeting the next element in the chain. The nuclease must be labeled with a fluorescent marker to readily quantify inheritance.

At least one daisyfield system with 8+ elements for which the cargo element (encoding the nuclease) is inherited at >90% frequency and the elements are inherited at 50% frequency on average (e.g. reduced by approximately half each generation). The fitness cost of the complete daisyfield drive system as measured by brood size measurements must be <0.2. Each daisy element must encode at least four guide RNAs targeting the cargo element. The nuclease must be labeled with a fluorescent marker to readily quantify inheritance.

3.1.3.5 Build massively linked populations using liquid-handling robot.

3.1.3.5.1 Stably propagate labeled nematodes on the robot.

Propagate 600 populations, 200 for each of 3 nematode strains bearing different fluorescent markers, in adjacent wells (60 total populations per plate) to quantify the rate of cross-contamination over 10 generations.

3.1.3.5.2 Quantify nematode populations on the robot.

Develop a method of using the plate reader to quantify the total worm population per well and the fraction bearing each fluorescent marker in mixed populations. Check the results by comparison to automated software-based identification of worms on plates via microscopy.

3.1.3.5.3 Repeat an earlier competitive fitness experiment on the robot.

Use the robot to duplicate an earlier competition experiment and compare the results to those obtained manually in larger cultures.

3.1.3.5.4 Simulate a metapopulation with predetermined gene flow rates on the robot.

Simulate 8 subpopulations of differing sizes with known gene flow rates between them. Each subpopulation shall consist of 1 or more wells of nematodes maintained on the robot that are thoroughly mixed each generation such that they are effectively panmictic. The robot shall additionally be programmed to transfer nematodes bidirectionally or unidirectionally between simulated populations at fixed rates. Seed initial populations with nematodes carrying different fluorescent markers, and monitor the frequency of each marker in each population using the plate reader. Continue for 10 generations.

Deliverables:** **System to test spatial models and daisy quorum in linked metapopulations over many generations. Populations shall exhibit <10% loss due to contamination with bacteria or other organisms over 8 generations, or demonstrate a means of resolving contamination. Populations shall not cross-contaminate more than 0.1% per generation. The system shall accurately simulate a predefined series of subpopulations of different sizes linked by arbitrary gene flow rates as determined by fluorescent monitoring of allelic exchange and abundance using the plate reader.

3.1.3.6 Test daisy drive stability and dynamics in large populations and metapopulations.

3.1.3.6.1 Evaluate daisy drive system stability in large populations.

Employ ~1L liquid cultures with gene flow between them for >20 generations to determine whether recombination of the best daisy-chain drive system and the best daisy-field drive system from 3.1.3.4.6 can create a global drive system within a certain minimum detectable frequency. Global drive creation will be evidenced by the spread of a fluorescent marker gene to dominate the last culture in the series of linked populations, well beyond what is predicted by the available daisy elements and the initial introduction. If this does occur, sequence the cargo to identify the nature of the change, then repeat the experiment at least twice at different scales to estimate frequency of global drive creation through recombination.

3.1.3.6.2 Build daisy-chain drive systems with 5+ elements.

Add additional daisy-chain elements to daisy-chain drive systems generated in 3.1.3.4.6.

3.1.2.6.3 Build a daisy quorum drive system.

Cross worms with appropriate daisy-chain and daisy-field drive systems to worms carrying swapped recoded ribosomal genes to generate daisy quorum drive systems.

3.1.3.6.4: Early tests of drive system spread in metapopulations.

Begin quantifying the spread of daisy-chain, daisy-field, and daisy quorum drive systems through small subpopulations of worms that are linked with pre-programmed gene flow rates of 1-10%. Seed one starting subpopulation with daisy drive worms at an initial frequency of 1-20% (exact parameters to bechosen based on modeling results). Observe dynamics of spread for at least 8 generations.

Deliverables: Daisy-chain and daisy-field drive systems with minimum quantified stability (<1 event causing formation of a daisy necklace that could exhibit self-sustaining drive per 10^9 ^organism-generations).

At least one daisy-chain drive system with 5+ elements for which the cargo element (encoding the nuclease) is inherited at >90% frequency, the daisy elements are each inherited at >80% efficiency when a prior element is present, and elements are inherited at normal Mendelian frequency when there is no prior element. The fitness cost of the complete daisy-chain drive system as measured by brood size measurements shall be <0.2. Each daisy element shall encode at least four guide RNAs targeting the next element in the chain. The nuclease shall be labeled with a fluorescent marker to readily quantify inheritance.

At least one functional daisy quorum drive system that alters 100% of a target subpopulation.

Technical Area 2 – Population suppression and genetic remediation.

**3.1.4 Nematodes (TA2 Task 2) **

The Recipient shall:

3.1.4.1 Build and overwrite “rogue drives” with corresponding immunizing reversal drives.

3.1.4.1.1 Create functional molecularly confined “rogue drive” systems.

Make three drive systems targeting a synthetic BFP marker gene in an engineered strain using SpCas9, Cpf1, and SaCas9 (or a new nuclease as applicable). Add a GFP marker as the “rogue cargocargo”, which is equivalent to a slightly costly gene that would be responsible for the undesired effects without being linked to the function of the rogue drive. Nucleases can be constitutively expressed per the task described in 3.1.3.2.4. Note that synthetic site targeting is the safest form of molecular confinement and these nematodes are not found in the New England area.

3.1.4.1.2: Create standard reversal and immunizing reversal drives.

Build reversal drive systems that target sequences within GFP and/or the guide RNA cassette of the rogue drive system, causing them to be replaced with mCherry and the guide RNA cassette of the reversal drive system in heterozygotes. Build immunizing reversal drive systems that are identical to the “rogue” drive systems save that they carry mCherry instead of GFP and additionally target sequences within GFP and/or the guide RNA cassette of the rogue drive system, causing them to be replaced with mCherry and the guide RNA cassette of the immunizing reversal drive system in heterozygotes.

3.1.4.1.3: Measure rogue, reversal, and immunizing reversal drive inheritance.

Cross each drive system with wild-type and measure inheritance frequencies, defined as the frequency at which BFP is lost and GFP is inherited in offspring (for rogue + recipient pairs), BFP is lost and mCherry is inherited (for IR + recipient), and GFP is lost and mCherry is inherited (for rogue + R or IR pairs). Inheritance shall be >90% for at least one drive system in each case.

3.1.4.1.4: Build and test a daisy immunizing reversal element.

Construct a daisy immunizing reversal element corresponding to one of the rogue drive systems. It shall encode guide RNAs that use the rogue drive nuclease to drive all of the elements of either a daisy-chain or a daisy-field drive system as well as itself, defined as cutting and replacing at least one critical genetic component of the rogue drive system with the immunizing reversal element. In addition, it shall express guide RNAs that use an orthogonal daisy drive nuclease to copy itself onto the 'wild-type' (in this case BFP-encoding) chromosome, thereby removing the target site(s) used by the rogue drive system. The immunizing reversal element shall additionally encode mCherry as a marker. Crossing a daisy immunizing reversal element with an appropriate rogue drive system should demonstrate inheritance of the former at

90%. Similarly, crossing a daisy immunizing reversal element with a strain carrying the cargo element from an appropriate daisy drive system as well as the targeted BFP marker gene should demonstrate inheritance of the former at >90%.

*Deliverables: *A 'rogue' global drive system inherited at >90% in strains with a targeted synthetic BFP gene and 0% in strains lacking BFP, with fitness costs<0.2 as measured by brood size. A global immunizing reversal drive system that overwrites the rogue drive system, to be inherited at >90%. A daisy immunizing reversal element that overwrites the rogue drive system using its own nuclease, to be inherited at >90%. A daisy immunizing reversal element that is inherited at >90% when crossed with a strain carrying an orthogonal nuclease previously employed in a functional daisy drive system as well as the targeted synthetic BFP gene.

3.1.4.2 Build and test components for a daisy suppression drive system.

3.1.4.2.1 Build drive elements that disrupt recessive sex-specific fertility genes.

Insert a fluorescent marker and guide RNA cassettes targeting wild-type sequences into candidate genes, to be identified primarily by seeking analogues of genes identified in An. gambiae. Quantify fertility of heterozygous and homozygous females by measuring brood sizes.

3.1.4.2.2 Demonstrate biased inheritance in the presence of a nuclease cargo element.

Cross carriers of demonstrated fertility genes to worms bearing a corresponding cargo nuclease-encoding element. Cross the resulting worms to one another or back to the nuclease-only parent. Quantify the fertility of offspring.

Deliverables:** **At least one identified sex-specific fertility gene, and a self-targeting guide RNA element disrupting that gene which is inherited at >90% by the progeny of worms encoding the corresponding nuclease gene.

3.1.4.3 Build functional daisy suppression drive system.

3.1.4.3.1 Construct a complete daisy suppression drive.

Cross carriers of proven infertility genes to worms carrying daisy-chain and daisyfield drive systems to generate daisy suppression drive systems.

3.1.4.3.2 Test direct daisy suppression on wild-type nematode populations.

Test population suppression via loss of fertility by releasing worms carrying daisy suppression drives into existing wild-type populations as fixed percentages of the existing population. Exact numbers to be determined by modeling of the drive system.

3.1.4.3.3 Test suppression of populations altered by daisy quorum systems.

Introduce worms carrying a daisy suppression element into a population of worms exhibiting daisy quorum as a step towards genetic remediation to wild-type. Evaluate the population decline of the daisy quorum alleles over at least 8 generations.

Deliverables: At least one daisy suppression drive system capable of reducing a population by 99% upon being introduced at a frequency <20%. A demonstration that populations altered by daisy quorum drive systems can be suppressed by introducing a daisy suppression allele.

Technical Area 1 - Control of genome editing activity.

3.1.5 High-Throughput Transgenesis in Nematodes (TA1 Task 3)

The Recipient shall:

3.1.5.1 Develop next generation microinjection platform using nematodes.

3.1.5.1.1 Design & fabricate anti-vibration hardware mounting platform.

3.1.5.1.2 Computer control of ultrasonic robotic manipulator motion.

Build software development kit enabling rapid robotic positioning with sub-micron accuracy and communication delays less than 5 milliseconds.

3.1.5.1.3 Dynamic control system for fast-switching of air pressure levels.

Computer control must rapidly switch from low back pressure to high injection pressure to deliver genetic fluid to organism and clear any needle clogs.

3.1.5.1.4 Software interface to control microinjection platform.

Integration of all hardware components under computer control cameras, robotic manipulators, dynamic air pressure regulation

3.1.5.1.5 Safety enclosure with temperature control.

3.1.5.1.6 Safety mechanism that disengages air pressure/robotics.

3.1.5.1.7 Refine user software interface for ease of use and efficiency.

3.1.5.1.8 Create training materials for operating the machine and provide on-site training and assistance to the Esvelt Lab with operation.

Deliverables: A computer-assisted system capable of injecting 2 worms/minute. This platform will be delivered to the Esvelt Lab to begin producing transgenic nematodes operated by Esvelt Lab technicians with assistance from Hive Biosystems. The system will use pre-pulled microneedles purchased from Eppendorf (femtotips) at this stage. The microneedle will be manually aligned to the high magnification objective and the user will click on the software interface to visit gonad targets. The microneedle pressure will be manually adjusted for optimal gonad filling. Once calibrated the system will be capable of microinjecting 2 worms per minute by a highly skilled operator.

3.1.5.2 Enhanced automation using needle alignment to camera, piezo-based penetration.

3.1.5.2.1 Continue providing on-site assistance to project team members.

3.1.5.2.2 Replicate first system to begin next-gen platform development.

3.1.5.2.3 Develop on-axis needle penetration by piezo-based vibration.

Use fast piezoelectric oscillations with sub-micron steps similar to a jackhammer motion with minimal off-axis vibrations to prevent unwanted damage.

3.1.5.2.4 Algorithm: automated microneedle alignment to camera

3.1.5.2.5 Optimize quartz microneedle shape & settings for delicacy during microinjections.

Deliverables: A microinjection system with piezo-based jackhammer penetration with automated algorithm for quartz needle alignment to camera. The user will click on the screen and use the mouse scroll wheel to focus on the target then engage the needle for piezo-penetration and dispense of fluid. The entire process is controlled by the software interface.

3.1.5.2.6: Algorithm: detect organism position in low magnification.

3.1.5.2.7: Algorithm: high magnification 3D image stack of target.

3.1.5.2.8: Algorithm: detect/respond to clogged needles with pressure.

3.1.5.2.9: Algorithm: insert microneedle deliver fluid w/ video feedback.

3.1.5.2.10: Develop rotation stage to orient target for microinjection.

Develop a high-precision and light-weight stage using ultrasonic motors. The custom fabrication shall orient the gonad target to the needle for optimal angle of penetration. The stage shall rapidly and precisely move in XYZ and rotation. This technology is also required for the transgenesis in mosquitoes option.

Deliverables: A high-throughput user-operated platform that can microinject 3 worms/minute*. *Capable of microinjecting 3 worms per minute after automated needle alignment and nematode gel immobilization. The precision and delicacy required for *C. brenneri *necessitates a computer-assisted microinjection platform.

Requirements:

  1. An algorithm will automatically calibrate the microneedle to the high magnification objective; 2) An algorithm will adjust the microneedle back pressure and pulse pressure for optimal fluid delivery; 3) An algorithm will automatically select coarse nematode gonad targets in low magnification; 4) An algorithm will automatically visit each coarse nematode location in high magnification; 5) The user will focus and select the gonad target in high magnification using the software interface; 6) An algorithm will select the optimal angle of injection and perform the rotation and XY translation with feedback in low magnification to present the gonad target to the microneedle in high magnification; 7) An algorithm will automatically engage the microneedle to penetrate the cuticle and gonad sheath with on-axis piezoelectric jackhammer motion; 8) An algorithm with video feedback will dispense the optimal amount of fluid into the gonad then retract; 9) An algorithm will detect microneedle clogs and increase to maximum pressure to clear the clog and return to operation. If high pressure does not clear the clog then the microneedle will be replaced and automatically calibrated to return to service; 10) This process will be repeated for the desired number of injections while recording temperature, duration, and humidity; 11) After setup calibrations the system should be able to microinject 3 nematodes per minute.

3.1.5.2.11 System to safely degrade DNA by heat inside the microneedle.

3.1.5.2.12 Piezo-based syringe to front-load plasmid into microneedle.

3.1.5.2.13 Post-injection recovery technique for optimal survival.

3.1.5.2.14 Refine techniques for transformation efficiency & speed.

3.1.5.2.15 Optimize user software interface for ease of use & efficiency.

3.1.5.2.16 Create training materials and on-site support to project team members.

Deliverables: High-throughput platform capable of microinjecting 4 worms/minute. The system shall automatically align the needle to the camera and then select the nematode targets in low magnification. The system shall visit each gonad target site where the user will focus and select the gonad target in high magnification. The injection needle shall automatically penetrate and dispense fluid into the gonad then retract. This process shall repeat to inject the desired number of worms for the given plasmid.

3.1.6 Aedes aegypti (TA1 Task 4)

The Recipient shall:

3.1.6.1 Optimize CRISPR guide RNA expression and multiplexing in vitro

3.1.6.1.1 Characterize pol III promoters.

The activity of at least three putative Aedes polIII promoters (e.g. U6, 7SK) shall be quantified in vitro relative to other Aedes, Anopheles and/or Culex equivalents by expressing guide RNAs used to target a reporter gene. Obtain rank order of promoter strength of at least three putative *Aedes *polIII promoter fragments.

3.1.6.1.2 Characterize tRNA- and crRNA-based multiplexing.

Different tRNA and crRNAs architectures shall be tested in Aedes cells* *by expressing guide RNAs used to target a reporter gene. At least three different mosquito tRNAs shall be tested and structure of processed products analyzed. If other RNA-processing methods appear advantageous, e.g. Cpf1-based processing, such structures may be substituted.

3.1.6.1.3 Characterize guide RNAs targeting sites of interest.

Guide RNAs selected via BreakingCas, or other means, that target ribosomal genes or downstream neutral sites shall be evaluated via cutting/insertion assay (at least 2 guide RNAs per target ribosomal gene). At least two candidate ribosomal genes shall be investigated with at least 2 guide RNAs per target ribosomal gene. Candidate ribosomal genes shall be prioritized based on specific criteria, as known, including: gene size, proximal upstream gene, chromosome location, single identifiable homolog with other insects.

Deliverables: Data on sgRNA sufficient to predict via modelling (Task 1) performance of construct daisy drive systems constructed using these components. Target: at least one polIII promoter active in Aedes aegypti cells; ability to multiplex at least two guide RNAs while retaining activity (endonuclease activity at target site in combination with appropriate protein component, e.g. Cas9); at least one guide RNA shown to be capable of directing cutting of a specified site in Aedes aegypti genome.

3.1.6.2 Optimize germline CRISPR nuclease expression conditions.

3.1.6.2.1: Generate transgenic strains expressing Cas9 or Cpf1.

Generate at least two transgenic lines carrying nuclease expression constructs. The nuclease shall be Cas9 or other suitable alternative such as Cpf1; based on data from this project, other Safe Genes projects and the broader community. Use promoters anticipated to be germline-specific – germline expression is essential for gene drive systems. Based on the literature and current data (principally RNAseq data and analyses), initial candidates are AAEL010097 and AAEL007097 for male-and-female germline expression; for expression at the M locus the lead candidate is AAEL010268 since only expression in the male germline is relevant for that application. Somatic activity may lead to fitness costs, depending on configuration details; such fitness cost may be undesirable depending on specific design.

3.1.6.2.2 Characterize expression of strains expressing Cas9 or Cpf1 in the germline.

Evaluate expression timing and the frequency of homologous recombination induced by nuclease-expressing strains and/or use lines from other Safe Genes projects, e.g. by crossing candidates to a line expressing guide RNAs and a reporter gene. Initially, germline expression will be assessed by analysing nuclease mRNA levels in testes and ovaries relative to gonadectomised carcass; for lines showing evidence of germline more complex and biologically relevant bio assay will be conducted measuring nuclease activity in male germline, female germline and somatic cells via rate of cutting/homing of target sequence. Combine at least one nuclease-expressing insertion with a model sgRNA construct or exogenous sgRNAs and use to measure cutting and/or homing rates in male and female germline and in somatic cells. As time and resources permit, assess cutting rates by measuring mutation rate by high-throughput sequencing or, if available, based on sgRNA insertion site, phenotypic effect.

Deliverables: CRISPR nuclease activity in the germline demonstrated by detecting products of cutting and/or homing arising in offspring of such individuals.

3.1.6.3 Identify ribosomal gene recodings & express nucleases.

3.1.6.3.1: Identify viable recodings.

Strains with recoded ribosomal genes shall be generated for at least two different ribosomal genes. Viability of strains heterozygous and homozygous for recoded ribosomal genes shall be assessed.

3.1.6.3.2: Encode germline-expressing nucleases in recoded strains.

Encode CRISPR nuclease gene(s) downstream of a recoded ribosomal gene so that the nuclease is expressed in the germline.

Deliverables: At least two strains with recoded ribosomal genes, and at least one strain that expresses a nuclease in the germline with the nuclease gene near a recoded ribosomal gene.

3.1.6.4 Develop a model daisy drive cargo element.

3.1.6.4.1 Generate transgenic line incorporating sgRNA construct into model target gene.

A model cargo element to test other daisy drive elements and overall system performance. Develop a transgenic line incorporating an sgRNA construct inserted into the kmo gene.

3.1.6.4.2 Combine strains from above to provide daisy drive strain(s).

Based on data from construct and strain performance from earlier sub-tasks, and design of cage trials below, combine strains to provide daisy drive components. One daisy drive configuration of particular interest involves conversion of wild type population to Cas9+, allowing later spread of one or more cargo elements into that population specifically; strains at this step may therefore comprise two of three daisy drive elements

3.1.6.4.3 Small cage proof of concept demonstration of daisy drive.

The key proof of principle for Phase I is to show basal-element dependent drive of an intermediate daisy drive element following introduction into a wild-type population. Optionally, associated spread of a cargo element in the presence of intermediate elements may also be investigated.

Deliverables: Daisy drive components shown in combination to be capable of base-element-dependent drive of subsequent elements, demonstrated by increase in allele frequency of subsequent elements relative to initial introduction frequency.

3.1.7 Culex quinquefasciatus (TA1 Task 5)

The Recipient shall:

3.1.7.1 Genome sequence and assembly of *Culex quinquefasciatus *lab strain.

3.1.7.1.1: Sequence genome to enable correct targeting.

Deliverable: Sequenced and assembled genome at >30x coverage.** **

3.1.7.2 Optimize CRISPR guide RNA expression and multiplexing in vitro.

3.1.7.2.1 Characterize pol III promoters.

The activity of at least two putative *Culex *polIII promoters (e.g. from U6 or 7SK genes) shall be quantified *in vitro *relative to *Aedes *equivalents by expressing guide RNAs used to target a reporter gene.

3.1.7.2.2 Characterize tRNA- and crRNA-based multiplexing.

Different tRNA and crRNAs architectures shall be tested in mosquito cells* *by expressing guide RNAs used to target a reporter gene. At least three different mosquito tRNAs shall be tested and the structure of processed RNA products analyzed. If other RNA-processing methods appear advantageous, e.g. Cpf1-based processing, such structures may be substituted. At time of writing, Cas9 is the preferred option; the possibility of changing nuclease is included to recognise the rapid activity in this field. Nuclease choice shall be reassessed for each round of nuclease construct development.

3.1.7.2.3 Characterize guide RNAs targeting sites of interest.

Guide RNAs selected via BreakingCas or other means that target ribosomal genes or downstream neutral sites shall be evaluated via cutting/insertion assay. At least three candidate ribosomal genes shall be investigated; candidates prioritized based on specific criteria, as known, including: gene size, proximal upstream gene, chromosome location, single identifiable homolog with other insects.

Deliverables: Data on sgRNA sufficient to predict via modelling performance (Task 1) of constructed daisy drive systems using these components. Target: at least one polIII promoter active in Culex cells; ability to multiplex at least two guide RNAs while retaining activity (endonuclease activity at target site in combination with appropriate protein component, e.g. Cas9); at least one guide RNA shown to be capable of directing cutting of a specified site in Aedes aegypti genome.

3.1.7.3 Express CRISPR nuclease in Culex germline.

3.1.7.3.1 Identify candidate germline promoters.

Identify candidate germline promoters from genome sequence based on similarity to genes/promoters known to have this property in other insects, e.g. nanos, zpg.

3.1.7.3.2: Generate strains expressing Cas9 or Cpf1.

Recipient will generate about three (at least one) transgenic lines carrying nuclease expression constructs. The nuclease will be Cas9, or other suitable alternative such as Cpf1; selection based on data from this project, other Safe Genes projects and the broader community. Recipient will use promoters anticipated to be germline-specific – germline expression is essential for gene drive systems. Somatic activity may lead to fitness costs, depending on configuration details; such fitness cost may be undesirable, again depending on specific design.

Deliverables: At least one transgenic line expressing either Cas9 or Cpf1 in the germline.

3.1.8 Anopheles gambiae (TA1 Task 6) (OPTIONAL REQUIREMENTS)

**SOW Section 3.1.8 is optional tasking to be exercised at the discretion of the DARPA Program Manager. **

The Recipient shall:

3.1.8.1 Optimize CRISPR guide RNA expression and multiplexing in vitro.

3.1.8.1.1 Characterize pol III promoters.

The activity of at least one putative *Anopheles *polIII promoters (e.g. from U6 or 7SK genes) shall be quantified *in vitro *by expressing guide RNAs used to target a reporter gene.

3.1.8.1.2: Characterize tRNA- and crRNA-based multiplexing.

Test different tRNA and crRNAs architectures in mosquito cells* *by expressing guide RNAs used to target a reporter gene. At least three different mosquito tRNAs shall be tested and the structure of processed RNA products analyzed. If other RNA-processing methods appear advantageous, e.g. Cpf1-based processing, such structures may be substituted. At time of writing, Cas9 is the preferred option; the possibility of changing nuclease is included to recognise the rapid activity in this field. Nuclease choice shall be reassessed for each round of nuclease construct development.

3.1.8.1.3 Characterize guide RNAs targeting sites of interest.

Evaluate guide RNAs selected via BreakingCas or other means that target ribosomal genes or downstream neutral sites via cutting/insertion assay. At least three candidate ribosomal genes shall be investigated; candidates prioritized based on specific criteria, as known, including: gene size, proximal upstream gene, chromosome location, single identifiable homolog with other insects.

Deliverables: Data on sgRNA sufficient to predict via modelling (Task 1) performance of constructed daisy drive systems constructed using these components. Target: at least one polIII promoter active in Anopheles cells; ability to multiplex at least two guide RNAs while retaining activity (endonuclease activity at target site in combination with appropriate protein component, e.g. Cas9); at least one guide RNA shown to be capable of directing cutting of a specified site in Anopheles genome.

3.1.8.2 Express CRISPR nuclease in Anopheles germline

3.1.8.2.1 Identify candidate germline promoters.

Identify candidate germline promoters from genome sequence based on the literature and/or similarity to genes/promoters known to have this property in other insects, e.g. nanos, zpg.

3.1.8.2.2 Generate strains expressing Cas9 or Cpf1.

Generate about three (at least one) transgenic lines carrying nuclease expression constructs. The nuclease shall be Cas9, or other suitable alternative such as Cpf1; selection based on data from this project, other Safe Genes projects and the broader community. Use promoters anticipated to be germline-specific – germline expression is essential for gene drive systems. Somatic activity may lead to fitness costs, depending on configuration details; such fitness cost may be undesirable, again depending on specific design.

Deliverables: At least one transgenic line expressing either Cas9 or Cpf1 in the germline.

3.1.9 Data Sharing (TA1 Task 7)

3.1.9.1 The Recipient shall share summaries of experimental plans, data, and analyses.

3.1.9.1.1 Share all summaries, experimental plans, data and analyses with DARPA every 3 months, and subsequently on other open publication platforms, as appropriate. The team will additionally participate in data sharing engagement among Safe Genes Teams as organized by DARPA. All findings relevant to gene drive are anticipated to be publicly shared following safety review.

Deliverable: Regular updates of experimental plans, data, analyses, and lay summaries.

**3.1.10 Ethical, Legal, Social and Risk Analysis (TA1 Task 8) **

3.1.10.1 Develop a framework for a technical and regulatory workshop in Washington, DC.

3.1.10.1.1 Define regulatory precedence related to technical safeguards for genome editing to inform and justify a technical and regulatory workshop. Engage with regulators as in Task 3.1.10.8.1 to inform framework. Define how an additional workshop will provide novel guidance for this Safe Genes project and/or the program overall. Obtain DARPA PM approval of framework and LEEDR Panel guidance prior to workshop planning.

Deliverable: Framework for technical and regulatory workshop in DC.

(OPTION) 3.1.10.2 Technical and regulatory workshop in Washington, DC.

Secure stakeholder feedback on planned primary technical applications and technical safeguards such as daisy drives in order to improve the quality of technical work, the quality of risk assessments and the prospects for satisfying regulatory standards for acceptance of said applications and safeguards.

3.1.10.2.1 Organize and hold technical/regulatory workshop.

Conduct a meeting of scientists, regulators, and a subset of local leaders from target communities to be held in Washington, DC, with the goals of identifying critical technical concerns environmental concerns, and knowledge/training needs, and building a roster of local civil society members, regulators, and scientists in preparation for local community meetings that will discuss proposed technical applications, risks, and safeguards. Include preparation of briefing materials, selection of venue, identification and invitation of participants, arrangement of travel, and direction of the workshop.* *

Deliverable: Reporting on technical, regulatory, and social issues, with emphasis on recommendations from meeting(s) on how to improve the development of safeguard technologies, to enhance the rigor of tests of safeguards and to improve the quality of assessments of how deployment of safeguards would affect risks. Recommendations from these meetings are expected to be of particular value in guiding the design and evaluation of safety/stability/spread tests in nematodes and mosquito cage trials.

3.1.10.3 Coordinate with DARPA PM, LEEDR Panel, and existing local outreach effort organizers in an appropriate location to be agreed upon with the DARPA PM to create an engagement framework.

3.1.10.3.1 Coordination to include discussion of Safe Genes gene drive technologies to inform broad-based deliberation on potential future practical applications in the context of a selected local community. It is noted at the outset that the Safe Genes project will not conduct any external field release.

3.1.10.3.2 Hold monthly phone calls with DARPA LEEDR Panel to create a plan for community engagement in coordination with existing gene drive outreach efforts in local communities or geographic locations. Incorporate discussion of daisy drive technology.

*Deliverables: *A detailed framework on proposed local engagement, including plans for integration with existing engagement efforts, as appropriate.

(OPTION) 3.1.10.4 Upon DARPA PM approval and with LEEDR Panel guidance, implement engagement strategy determined in 3.1.10.3.

3.1.10.4.1 Engagement strategy may include meeting with existing outreach organizers in a specific location to discuss their current engagement efforts and information regarding local and external civil society, regulators, and scientists; eliciting feedback on technical development and testing plans for proposed daisy drive applications and safeguards; and preparing briefing materials, selecting venue, identifying and inviting participants, arranging travel, and directing the meeting. External funds may be used to support engagement and community meetings if appropriate. The team will reach out to related projects working with communities to gather ecological data to determine whether collaboration is appropriate.

Deliverables: A post-engagement meeting report that details local concerns and offers recommendations on how to integrate these concerns into technical development, testing, and risk assessment.

3.1.10.5 Ethics of ecological engineering technologies and gene drives.

3.1.10.5.1 Conduct literature review on relevant technologies and ethics.

3.1.10.5.2 Outline the key questions in ethics and gene drive technologies.

3.1.10.5.3 Identify specifics of daisy drive systems and ethical relevance.

Deliverable: An ethical framework specific to the development and deployment of daisy drive systems. The ethics of daisy drive and its derivatives (daisy quorum, daisy field) are distinct from other CRISPR drive systems, because they designed to be self-exhausting and only act at a local scale. However remote, the possibility of genetic rearrangement into a drive system with altered functionality (including enhancement of drive) also merits ethical analysis, to include the development of risk mitigation strategies to address a diversity of possible outcomes.

3.1.10.6 Ethics consultation

3.1.10.6.1 Provide ethics consultation to MIT’s Safe Genes members.

An ethicist embedded as a full-time team member will identify and address issues and dilemmas by assisting the team in deciding ethically preferred courses of action. Advise on the ethics of community engagement and involvement efforts, including choice of which communities to involve in the development processdaisy drive technology will be applied at least biannually, or more frequently as appropriate. The meetings shall focus on sharing programmatic and technical progress of the daisy project under the Safe Genes program and the construction of a framework for evaluation of future applications in mosquitoes, but should also apply to other organisms as appropriate. The goal shall be to allow for integrated planning of experiments and risk assessments that will be required to be in compliance with laws relevant to the future application of daisy drives. The regulators will be updated on progress and facilitate changes to technology development and risk assessment as necessary to increase the likelihood that products arising from Safe Genes will obtain regulatory approval. The investigators will also meet with regulators in other nations to discuss daisy drive technologies, tests, and risk assessments to improve the quality of technical work and increase the likelihood that this project will generate data and analysis on localization of gene drives that are relevant to future discussions of international acceptance of U.S. regulatory approvals.

3.2 OPTION PERIOD (PHASE 2)

Technical Area 2 – Population suppression and genetic remediation.

3.2.1 Mathematical Modelling. (TA2 Task1 Phase2)

The Recipient shall:

3.2.1.2 Assess complete genetic remediation strategies in well-mixed populations

3.2.1.2.1 Assess daisy immunizing reversal with quorum (DW+IRQ).

Extend the DW+IR model to include a quorum underdominance effect added to the daisy immunizing reversal drive system.

3.2.1.2.2 Assess genetic remediation of a rogue drive system (DW+IRQ+S).

To the DW+IRQ model of a rogue drive countered by daisy immunizing reversal with quorum, add the subsequent release of a daisy suppression element specific to the quorum-altered population, and track predicted remediation to wild-type.

3.2.1.2.3 Iteratively refine models based on experimental results.

Guide the design of nematode experiments and iteratively refine the model with experimental data.

Performance Metrics and Assessment: In sub-sub tasks 3.2.1.2.1 and 3.2.1.2.2, performance of remediation strategies will be quantified by absolute frequency of rogue drive 20 generations after release of the remediation drive. Also assess robustness of the absolute frequency measure with respect to design and release parameters (drive efficiency, fitness costs, initial release frequency of remediation drive and frequency attained by rogue drive before release of remediation drive).

*Deliverables: *Model extensions (DW+IRQ, DW+IRQ+S) which assess genetic remediation strategies in well-mixed populations, including internally panmictic linked subpopulations, and focusing on complete genetic remediation to wild-type in the latter case. Spatial dynamics shall serve as important baselines for the subsequent spatially explicit models in sub task 3.2.1.3.

3.2.1.3 Refine well-mixed metapopulation models to study a theoretical island release.

3.2.1.3.1 Compile mosquito abundance/migration data for a given island locale(s) from literature.

Collect data on mosquito drive parameters (cutting efficiency, HR and NHEJ rates) as well as population-level parameters specific to ecology of the chosen site, as available (population sizes and migration rates).

3.2.1.3.2 Fine-tune predictive models with empirical mosquito/nematode parameters.

Employ drive- and population parameters such as homing and fitness cost, guiding and refined by empirical nematode experiments mimicking those drive systems, to fine-tune the linked panmictic subpopulation models to produce informative models of candidate island releases.

Deliverables: Linked panmictic metapopulation models of daisy drive systems to predict outcomes of various potential release scenarios, to be updated at 48mo with additional nematode and mosquito data.

3.2.1.4 Construct spatially explicit daisy drive modeling frameworks.

3.2.1.4.1 Deterministic reaction-diffusion models.

Include diffusion components to model dynamics in two spatial dimensions. These models shall directly extend the previous models (DW, DW+S, DW+IR, DW+IRQ, DW+IRQ+S) and serve as a baseline for stochastic models developed in 3.2.1.4.2.

3.2.1.4.2 Stochastic spatial models.

Develop stochastic spatial variants of the DW+ models from sub-sub task 3.2.1.2.1. Stochasticity will be particularly important to consider in cases of very small populations or very low migration rates.

Deliverables: Extensions of the DW+ models to include explicit spatial dynamics. These models will be iteratively extended through 3.2.1.5 and 3.2.1.6 to eventually result in spatial models to study theoretical island releases.

3.2.1.5 Assess genetic remediation strategies in spatially explicit models.

3.2.1.5.1 Assess targeted daisy suppression.

Model targeted suppression of a population harboring a rogue gene or drive using a daisy suppression drive alone. Identify the release size and frequency required to achieve remediation, informed by nematode data from 3.2.2.3.

3.2.1.5.2 Assess daisy immunizing reversal with quorum in spatially explicit systems.

Evaluate the daisy immunizing reversal with quorum strategy for genetic remediation. Compare to the well-mixed metapopulation models of 3.2.1.4, and inform modeling by nematode data from 3.2.2.3.

Deliverables: Evaluation of the capacity of daisy suppression alone to slow or completely eliminate and remediate engineered genes or rogue drives in spatial systems. Evaluation of the capacity of daisy immunizing reversal with quorum followed by suppression to eliminate an engineered gene or a rogue drive and restore wild-type genetics.

3.2.1.6 Refine spatially explicit models to study theoretical island release.

3.2.1.6.1 Compile selected island(s) mosquito abundance/migration data from literature.

Collect data on mosquito drive parameters (cutting efficiency, HR and NHEJ rates) as well as population-level parameters specific to local island ecology (population sizes and migration rates). Also consult with local ecologists for any data that is not publicly available.

3.2.1.6.2 Fine-tune predictive models with empirical mosquito parameters.

Employ drive- and population parameters obtained in sub-sub task 3.2.1.6.1 to fine-tune the spatially explicit models from sub task 3.2.1.4 to produce informative models of local releases. If mosquito population data is not available, use parameters developed in conjunction with local ecologists and informed by community-gathered data.

Deliverables: Spatially explicit models of island-based daisy drive systems. A comparison of linked panmictic subpopulation models, spatially explicit models, and nematode metapopulation experiments.

**Technical Area 1 – Control of genome editing activity. **

**3.2.2 Nematodes (TA1 Task 2) **

The Recipient shall:

3.2.2.1 Empirical optimization of daisy-chain and daisy-drive systems.

3.2.2.1.1: Build daisy-chain drive systems with 7+ elements.

3.2.2.1.2: Build parallel daisy-chain systems to evade resistance.

Construct at least two different versions of the same daisy-chain gene drive system with guide RNAs targeting different sites within the neutral sequence replaced by the next guide RNA. Resistance alleles to one daisy-chain drive system shall not be capable of blocking another version of that system.

3.2.2.1.3 Build daisy-field drive systems with 32+ elements.

3.2.2.1.4 Build strains to maintain and propagate daisy-field.

Build 3+ strains expressing different guide RNAs targeting the repeat sequence replaced by daisy-field elements. Iteratively mate these with daisy-field strains to convert repeat sequences lacking a daisy-field element into daisy-field elements. Quantify the number of repeats by qPCR.

3.2.2.1.5 Create extended daisy-chain and high-copy daisy-field quorum strains. Cross the quorum/underdominance strains with daisy-chain and daisy-field systems.

Deliverables: A daisy-chain drive system with 7+ elements, with and without quorum. A daisy-field drive system with 32+ elements, with and without quorum. At least 3 strains expressing different guide RNAs targeting the repeat sequence replaced by daisy-field elements, whose use in concert can maintain (when outcrossing) and increase the number of daisy-field elements in a strain.

3.2.2.2 Long-term population genetics comparisons of daisy drive systems.

3.2.2.2.1 Rigorously test stability of optimized strains.

Repeat the daisy stability tests for the high-powered daisy-chain strains.

3.2.2.2.2 Long-term metapopulation studies of daisy-chain systems.

Quantify the eventual fraction of metapopulation altered per organism released as a function of number of daisy elements by propagating for 30 generations in a metapopulation consisting of at least twenty linked subpopulations. Monitor the rise of drive-resistant alleles over time and their effects on spread via sequencing. Design to be informed by 3.2.1.2 and 3.2.1.4, and results to be used to refine those models in turn.

3.2.2.2.3 Long-term metapopulation studies of daisy-field systems.

Similar requirements as 3.2.2.2.2 for daisy-chain drive systems above.

3.2.2.2.4 Long-term metapopulation studies of daisy systems with quorum.

Similar requirements as 3.2.2.2.2 for daisy-chain and daisy-field above, but with quorum.

Deliverables: Minimum stability metrics of daisy-chain drive systems with respect to potential recombination into a global drive daisy necklace. Empirical studies of spread through linked metapopulation to inform models, ultimately granting the ability to reliably predict the spread of a given class of daisy drive system through different population structures as a frequency of the number of daisy elements, cost, HDR rate, and release number, frequency, and spatial distribution (with 3.2.1.2 and 3.2.1.4).

3.2.2.3 Empirical models of mosquito drive systems in nematodes.

3.2.2.3.1 Construct nematode equivalents of mosquito species.

Choose nematode strains with HDR rates approximating those of target mosquito species. Make worm strains carrying daisy-relevant DNA sequences from those species. Specifically, recode nematode ribosomal genes to carry target sequences from mosquito equivalents. If necessary, encode these sequences in the nematode ribosomal gene intron and 3'UTR such that cutting both deletes the last exon and disrupts the gene. Insert sequences corresponding to those targeted by mosquito daisy drive elements into neutral sites. Also replace a nematode fertility gene whose disruption causes effects similar to a mosquito fertility gene with a fusion of the two genes, constituted as mosquito5'-2A-nematode-2A-mosquito3', such that targeting the mosquito gene also disrupts the nematode gene.

3.2.2.3.2 Construct nematode daisy drive equivalent of mosquito drive systems.

Construct an analogue of each mosquito daisy drive system in the nematode strain designed to mimic that species of mosquito. While the promoters used will necessarily differ, both guide RNAs and targeted sequences shall be identical.

3.2.2.3.3 Quantify minimal stability of mosquito daisy drive systems.

Determine whether mosquito daisy drive systems are vulnerable to recombination into global daisy necklaces. Note that this measurement cannot be performed in mosquitoes due to limited population size.

3.2.2.3.4: Test mosquito daisy drive dynamics in nematodes mimicking one from potential island site(s).

Construct a metapopulation of nematodes with mosquito DNA on the robot that mimics the best-understood mosquito population structure on chosen islands with respect to the size of each subpopulation and rates of gene flow between them. Introduce a nematode strain with a daisy drive system analogous to that constructed in the target mosquito species into the subpopulations at a feasible release rate and pattern. Observe and record the rate of spread and population levels of mosquitoes over time. In particular, quantify any evolution of drive-resistant alleles or lingering spread of daisy components to populations mimicking those of other islands. Compare results to mathematical models and use to refine subsequent mathematical models and repeated experiments.

Deliverables: Nematode strains engineered to correspond to each mosquito species by HDR rate, brood size, and sequences targeted by daisy drive systems. Nematode strains carrying daisy drive systems analogous to those constructed in each mosquito species where analogous drive systems have different expression signals but identical guide RNAs and target sequences. Empirical data measuring the safety and stability of daisy drive systems up to a confidence level limited by the maximum number of mosquito-mimicking nematodes that can be grown.

**Technical Area 2 – Population suppression and genetic remediation **

3.2.3 Genetic remediation and restoration. (TA2 Task2)

The Recipient shall:

3.2.3.1 Genetic remediation and restoration.

3.2.3.1.1 Test genetic remediation of daisy quorum to wild-type via daisy suppression.

Starting from linked subpopulations of worm, wherein that one subpopulation has become nearly fixed for a daisy quorum drive system, and that population is linked through a series of others to one that remains wholly wild-type, introduce worms carrying a daisy suppression element into the altered population exhibiting underdominance. Track the population decline of the daisy quorum alleles and their hypothesized eventual replacement by wild-type worms migrating from linked populations by fluorescence.

3.2.3.1.2 Test genetic remediation of an engineered gene to wild-type.

Test elimination of an engineered gene in a metapopulation, specifically, a BFP allele introduced into a linked series of otherwise wild-type subpopulations, where the BFP element frequency is very high in some subpopulations and much lower in others. Build a daisy quorum reversal system consisting of the swapped haploinsufficient ribosomal genes with nucleases attached, daisy elements that drive the quorum elements, and a guide RNA element + mCherry element that directs cutting and replacement of BFP with itself using the daisy nuclease and also drives the quorum elements. Introduce the construct into the metapopulation at a frequency deemed reasonable for wild-type releases (e.g. 0.5-5%). Measure the incidence of the original BFP allele and its replacement by mCherry over 15 generations, then release a daisy suppression element to remove the quorum-affected population and track elimination of engineered alleles from the population by monitoring fluorescence.

3.2.3.1.3 Test daisy suppression as a method of eliminating a rogue drive system.

Using the rogue drive system of 3.2.2.1.1 and daisy suppression systems with few or many daisy elements (3 elements from 3.2.2.2 or 7+ from crossing the daisy drive system of 3.2.1.7 with a suppression element) and guided by the modeling efforts of 3.1.1.5 and 3.2.1.2, experimentally determine the feasibility of eliminating a rogue drive system using daisy suppression alone.

3.2.3.1.4 Test daisy immunizing reversal as a method of eliminating a rogue drive system.

Guided by the models of 3.2.1.2, test the feasibility of completely eliminating a rogue drive system released into a metapopulation consisting of at least 10 linked subpopulations with a daisy immunizing reversal drive system. Use enhanced daisy drives from 3.1.2.7 as well as weaker ones from 3.2.2.2. Track drive system incidence by fluorescence.

3.2.3.1.5 Test daisy immunizing reversal with quorum vs rogue drive, then suppression.

As 3.2.3.1.4 but use a daisy quorum drive system.

Deliverables: Experimental quantification of the number of organisms that must be released in a particular spatial distribution to restore populations altered with introduced engineered genes or with rogue drive systems to wild-type genetics using daisy suppression alone or daisy immunizing reversal with quorum followed by suppression. Studies to be undertaken in metapopulations of nematodes maintained on a liquid-handling robot that consist of at least 10 linked subpopulations (that is, a minimum of 9 transitions via programmed gene flow to move between the two most distantly connected subpopulations, with additional branching populations connected to one or more of that primary chain). Each subpopulation must comprise at least 100,000 organisms. A daisy drive system capable of eliminating every copy of the rogue drive system when released in numbers comprising <5% of the population.

Technical Area 1 – Control of genome editing activity.

**3.2.4. Aedes Aegypti. (TA1 Task 3 Phase 2) **

The Recipient shall:

3.2.4.1 Create daisy drive elements and strains.

3.2.4.1.1 Assess fitness of recoded strains expressing nucleases (life history).

Assess fitness of recoded strains expressing nucleases by measuring relevant life history parameters (e.g. egg-to-adult survival, developmental period, fecundity). Screen at least two strains, if available.

3.2.4.1.2 Assess fitness of recoded strains expressing nucleases (allele frequency evolution).

Assess fitness of up to three best-performing strain(s) by evaluating multi-generation allele frequency evolution in mixed-genotype populations. This more complex, lengthy and labor-intensive assay is sensitive to fitness costs at any/all life cycle stage.

3.2.4.1.3 Insert guide RNA gene(s) into neutral or recoded loci.

Create 1+ neutral or recoded daisy guide RNA elements based on data from modeling, recoding, and Ae. aegypti,.

3.2.4.1.4 Generate daisy drive platform strains.

Internally combine daisy element strains and cross with the nuclease expressing strain to make daisy drive platform strains.

Deliverables: 1+ strain with recoded homozygous viable ribosomal gene(s), 1+ strains that express CRISPR nucleases in the germline adjacent to a recoded ribosomal gene, 1+ daisy drive strains with 2+ elements. For nuclease gene insertions: homozygous viable (homozygous survival to adult >0.75 relative to wild type); fitness of each element >0.75 (per allele, relative to wild type). These fitness parameters shall be refined in the light of model outputs from Task 1

3.2.4.2 Assess impact of off-target endonuclease activity

3.2.4.2.1 Assess rate of mutagenesis at target sites.

Sequence a sample of wild-type alleles by deep sequencing to detect the nature and frequency of induced mutation at the target site. The small cage experiment provides an opportunity to investigate the rate of non-canonical events at the target locus, e.g. mutation via non-homologous end joining (NHEJ) rather than homology-dependent repair and homing.

3.2.4.2.2: Assess rate of cutting of off-target sites.

Using an sgRNA and nuclease-expressing insertion lines from above, identify the most closely related sites in the genome based on sequence similarity (up to three). Measure rate of mutagenesis at these sites relative to cutting (or homing) rate at nominal target site.

Deliverables: Quantitative data on rates of induced sequence change at target and non-target sites.

Technical Area 2 – Population suppression and genetic remediation.

3.2.5 Genetic load daisy elements for suppression and remediation. (TA2 Task 3 Phase 2)

The Recipient shall:

3.2.5.1 Create genetic load daisy elements for suppression and remediation.

3.2.5.1.1 Identify candidate female fertility genes.

Based on sequence similarity with other insects (e.g. Aedes aegyptiAnopheles gambiaeand Drosophila melanogaster), identify at least three putative female fertility genes in Culex quinquefasciatus.

3.2.5.1.2 Validate putative female fertility genes.

Disrupt genes by gene editing with target-specific guide RNA cassettes; assess fertility of heterozygous and homozygous mutants. Also, validate guide RNAs targeting these genes.

3.2.5.1.3 Insert guide RNA expression constructs in female fertility genes.

Insert guide RNA expression construct into at least one female fertility gene

3.2.5.1.4 Assess fertility/fitness effects of mutations/insertions into female fertility genes.

Test fertility of males and females, especially homozygous females, by measuring per-female larval productivity, i.e. egg production and hatch rate of such eggs.

Deliverables: 1+ strains with a guide RNA cassette disrupting a female fertility gene and measurements of fertility and fitness for hetero/homozygotes of both sexes. Fertility (per-female larval productivity) <10% of wild-type as homozygotes and >70% of wild type as heterozygotes. These fertility parameters will be refined in the light of model outputs from Task 1

3.2.5.2 Genetic load daisy drive strains for suppression and remediation

3.2.5.2.1 Insert guide RNAs at Male-determining (M) locus

Guide RNAs shall ideally target the basal element from a previously generated daisy drive strain. Test activity by crossing with the daisy drive strain and observing progeny inheritance. Expression from this locus may not succeed, but will be attempted.

3.2.5.2.2 Build a daisy suppression strain.

Cross the genetic load daisy elements to the daisy drive strain, ideally with the M-locus basal element.

3.2.5.2.3 Test population suppression.

Run multiple parallel cage trials of the strain(s) to quantify the rate of spread of the female-fertility allele(s) and extent of population suppression.

Deliverables: 1+ daisy drive strains with 3+ elements, assessment of spread of daisy drive elements within isolated populations and suppression via genetic load.

Technical Area 1 – Control of genome editing activity.

**3.2.6 Culex Quinquetasciatus. (TA1 Task 4) **

The Recipient shall:

3.2.6.1 Culex quinquefasciatus.

3.2.6.1.1 Characterize expression of strains designed to express Cas9 or Cpf1 in the germline.

Evaluate expression timing and the frequency of homologous recombination by crossing candidate lines to a line expressing guide RNAs targeting a reporter gene, e.g. kmo. Activity in male germline, female germline and somatic cells shall be assessed.

Deliverables: CRISPR nuclease activity in the germline demonstrated by detecting products of cutting and/or homing arising in offspring of such individuals.

3.2.6.2 Identify ribosomal gene recodings & express nucleases.

3.2.6.2.1 Identify homozygous viable recodings.

Generate strains with recoded ribosomal genes for 3 (minimum 1) different ribosomal genes. Asses viability of strains homozygous for recoded ribosomal genes.

3.2.6.2.2 Encode germline-expressing nucleases in recoded strains.

Encode CRISPR nuclease gene(s) downstream of a recoded ribosomal gene so that they are expressed in the germline.

Deliverables: At least one strain with recoded ribosomal genes, and at least one strain that expresses a nuclease in the germline with the nuclease gene near a recoded ribosomal gene.

3.2.6.3 Create daisy drive elements and strains.

3.2.6.3.1 Assess fitness of recoded strains expressing nucleases (life history parameters).

Assess fitness of recoded strains expressing nucleases by multi-generation allele frequency evolution in mixed-genotype populations. Up to three strains shall be assessed, if available.

3.2.6.3.2 Assess fitness of recoded strains expressing nucleases (allele frequency evolution).

Assess fitness of up to three best-performing strain(s) by multi-generation allele frequency evolution in mixed-genotype populations. This more complex, lengthy and labor-intensive assay is sensitive to fitness costs at any/all life cycle stage.

3.2.6.3.3 Insert guide RNAs into neutral or recoded loci.

Create 1+ neutral or recoded daisy guide RNA elements based on data from modeling, recoding, and Ae. aegypti experiments.

3.2.6.3.4 Generate daisy drive platform strains.

Internally combine daisy element strains and cross with a nuclease expressing strain to make daisy drive platform strains.

Deliverables: 1+ strain with recoded homozygous viable ribosomal gene(s), 1+ strains that express CRISPR nucleases in the germline adjacent to a recoded ribosomal gene, 1+ daisy drive strains with 2+ elements.

Technical Area 2 – Population suppression and genetic remediation.

3.2.7 Genetic load daisy elements for suppression and remediation. (TA2 Task 4)

The Recipient shall:

3.2.7.1 Create genetic load daisy elements for suppression and remediation.

3.2.7.1.1 Identify candidate female fertility genes.

Based on sequence similarity with other insects, identify at least three putative female fertility genes in Culex quinquefasciatus.

3.2.7.1.2 Validate putative female fertility genes.

Disrupt genes by gene editing with target-specific guide RNA cassettes; assess fertility of heterozygous and homozygous mutants. Also validate guide RNAs targeting these genes.

3.2.7.1.3 Insert guide RNA expression constructs into female fertility genes.

Insert guide RNA expression construct into at least one female fertility gene

3.2.7.1.4 Verify female infertility.

Homozygose the above strain(s) and test fertility.

Deliverables: 1+ strains with a guide RNA cassette disrupting a female fertility gene and measurements of fertility and fitness for hetero/homozygotes of both sexes. Fertility (per-female larval productivity) <10% of wild-type as homozygotes and >70% of wild type as heterozygotes. These fertility parameters shall be refined in the light of model outputs from Task 1

3.2.7.2 Create genetic load daisy drive strains for suppression and remediation.

3.2.7.2.1 Build a daisy suppression strain.

Cross the genetic load daisy elements to the daisy drive strain.

3.2.7.2.2 Test population suppression.

Run 4 parallel cage trials of the strain(s) to quantify the extent of suppression.

Deliverables: 1+ daisy drive strains with 3+ elements, optionally assessment of suppression via genetic load.

Technical Area 1 – Control of genome editing activity.

3.2.8 – Anopheles Gambiae (TA1 Task 5)

The Recipient shall:

3.2.8.1 Identify germline nuclease expression signals.

3.2.8.1.1 Characterize expression of strains designed to express Cas9 or Cpf1 in the germline 30mo

Evaluate expression timing and the frequency of homologous recombination by crossing candidate lines to a line expressing guide RNAs targeting a reporter gene, e.g. kmo. Activity in male germline, female germline and somatic cells shall be assessed.

Deliverables: CRISPR nuclease activity in the germline demonstrated by detecting products of cutting and/or homing arising in offspring of such individuals** 30mo**

3.2.8.2 Identify ribosomal gene recodings & express nucleases.

3.2.8.2.1 Identify homozygous viable recodings.

Generate strains with recoded ribosomal genes for 3 (minimum 1) different ribosomal genes. Viability of strains homozygous for recoded ribosomal genes shall be assessed.

3.2.8.2.2: Encode germline-expressing nucleases in recoded strains.

Encode CRISPR nuclease gene(s) downstream of a recoded ribosomal gene so that they are expressed in the germline.

Deliverables: At least one strain with recoded ribosomal genes, and at least one strain that expresses a nuclease in the germline with the nuclease gene near a recoded ribosomal gene.

3.2.8.3 Create daisy drive elements and strains.

3.2.8.3.1 Assess fitness of recoded strains expressing nucleases (life history parameters).

Assess fitness of recoded strains expressing nucleases by measuring relevant life history parameters (e.g. egg-to-adult survival, developmental period, fecundity) and/or multi-generation allele frequency evolution in mixed-genotype populations. At least two strains shall be assessed, if available.

3.2.8.3.2 Assess fitness of recoded strains expressing nucleases (allele frequency evolution).

Assess fitness of up to three best-performing strain(s) by multi-generation allele frequency evolution in mixed-genotype populations. This more complex, lengthy and labor-intensive assay is sensitive to fitness costs at any/all life cycle stage.

3.2.8.3.3 Insert guide RNAs into neutral or recoded loci.

Based on data from modeling, recoding, and Ae. aegypti, create 1+ neutral or recoded daisy guide RNA elements.

3.2.8.3.4 Generate daisy drive platform strains.

Internally combine daisy element strains and cross with the nuclease expressing strain from 4.1.6.4.2 to make daisy drive platform strains.

Deliverables: 1+ strain with recoded homozygous viable ribosomal gene(s), 1+ strains that express CRISPR nucleases in the germline adjacent to a recoded ribosomal gene, 1+ daisy drive strains with 2+ elements.

Technical Area 2 – Population suppression and genetic remediation.

3.2.9 Genetic load daisy elements for suppression and remediation. (TA2 Task 5 Phase2)

3.2.9.1 Create genetic load daisy elements for suppression and remediation.

3.2.9.1.1 Identify candidate female fertility genes.

Based on sequence similarity with other insects (e.g. Aedes aegypti and Drosophila melanogaster), identify at least three putative female fertility genes in Anopheles gambiae.

3.2.9.1.2 Validate putative female fertility genes.

Disrupt genes by gene editing with target-specific guide RNA cassettes; assess fertility of heterozygous and homozygous mutants. Validate guide RNAs targeting these genes.

3.2.9.1.3 Insert guide RNA expression constructs into female fertility genes.

Insert guide RNA expression construct into at least one female fertility gene

3.2.9.1.4 Verify female infertility.

Homozygose the above strain(s) and test fertility.

Deliverables: 1+ strains with a guide RNA cassette disrupting a female fertility gene and measurements of fertility and fitness for hetero/homozygotes of both sexes. Fertility (per-female larval productivity) <10% of wild-type as homozygotes and >70% of wild type as heterozygotes. These fertility parameters will be refined in the light of model outputs from Task 1.

3.2.9.2 Create genetic load daisy drive strains for suppression and remediation.

3.2.9.2.1: Insert guide RNAs at Male-determining (M) locus (optional).

Guide RNAs should target the basal element from a daisy drive strain. Test activity by crossing with the daisy drive strain and observing progeny inheritance.

3.2.9.2.2: Build a daisy suppression strain.

Cross the genetic load daisy elements to the daisy drive strain, ideally with the M-locus basal element.

3.2.9.2.3: Test population suppression.

Run 4 parallel cage trials of the strain(s) to quantify the extent of suppression.

Deliverables: 1+ daisy drive strains with 3+ elements, optionally assessment of suppression via genetic load.

Technical Area 1 – Control of genome editing activity

3.2.10 High-Throughput Transgenesis in Mosquitoes (TA1 Task 6 Phase 2) (OPTIONAL REQUIREMENT)

**SOW Section 3.2.10 is optional tasking to be exercised at the discretion of DARPA BTO. **

The Recipient shall:

3.2.10.1 Develop high-throughput transgenesis of Aedes aegypti.

3.2.10.1.1 Replicate second system, begin mosquito system development.

3.2.10.1.2 Egg immobilization/mounting technique for Aedes aegypti.

3.2.10.1.3 Optimize microneedle shape & parameters for delicacy.

3.2.10.1.4 Algorithm to detect organism position in low magnification.

3.2.10.1.5 Algorithm to acquire high mag 3D image stack.

3.2.10.1.6 Algorithm: insert microneedle, deliver fluid, feedback/control.

3.2.10.1.7 Develop post injection recovery technique: optimal survival.

3.2.10.1.8 Refine techniques for transformation efficiency and speed.

Deliverables: A high-throughput platform that can microinject 6 *Aedes aegypti *eggs/min with survivability equal to or greater than manual rates (5%).

3.2.10.2 Develop high-throughput transgenesis of Anopheles gambiae mosquitoes.

3.2.10.2.1 Egg immobilization/mounting technique for Anopheles gambiae.

3.2.10.2.2 Optimize microneedle shape & parameters for delicacy.

3.2.10.2.3 Algorithm to detect organism position in low magnification.

3.2.10.2.4 Algorithm to acquire high mag 3D image stack.

3.2.10.2.5 Algorithm: insert microneedle, deliver fluid, feedback/control.

3.2.10.2.6 Develop post injection recovery technique: optimal survival.

3.2.10.2.7 Refine techniques for transformation efficiency and speed.

Deliverables: A high-throughput platform that can microinject 6 *Anopheles gambiae *eggs/min with survivability equal to or greater than manual rates (5%).

3.2.10.3 High-throughput transgenesis of Culex quinquefasciatus mosquitoes

3.2.10.3.1 Egg immobilization/mounting technique for Culex quinquefasciatus.

Culex mosquitoes lay rafts of dozens of eggs attached together that must be separated for injection targeting. Their chorion develops causing the outer shell to harden and break needles. A bleaching procedure can reduce the chorion to allow additional microinjections but this process reduces survivability. Culex is highly sensitive to desiccation, so gel conditions must be optimized for buoyancy and hydration.

3.2.10.3.2 Optimize microneedle shape & parameters for delicacy.

3.2.10.3.3 Algorithm to detect organism position in low magnification.

3.2.10.3.4 Algorithm to acquire high mag 3D image stack.

3.2.10.3.5 Algorithm: insert microneedle, deliver fluid, feedback/control.

3.2.10.3.6 Develop post injection recovery technique: optimal survival.

3.2.10.3.7 Refine techniques for transformation efficiency and speed.

Deliverables: A high-throughput platform that can microinject 6 Culex quinquefasciatuseggs/min with survivability equal to or great than manual rates (5%).

3.2.11 Data Sharing (Task 7)

3.2.11.1 The Recipient shall share relevant experimental plans, data, and analyses on Responsive Science, and/or as preprints on bioRxiv, save for those involving microinjection system development.

Deliverable: Regular updates of experimental plans, data, analyses, and lay summaries.

3.2.12 Ethical, Legal, Social and Risk Analysis (Task 8 Phase 2)

3.2.12.1 Develop a framework for the second technical and regulatory workshop in DC.

3.2.12.1.1 Define outcomes from first regulatory workshop and justification for second workshop on technical safeguards for genome editing. Continue to engage with regulators as in Task 3.2.12.8 to inform framework. Define how the second workshop will provide additional guidance for this Safe Genes project. Obtain DARPA PM approval of framework and LEEDR Panel guidance prior to workshop planning.

Deliverable: Framework for second technical and regulatory workshop in DC.

OPTION 3.2.12.2 Organize and run second technical and regulatory workshop.

To secure broad feedback on technical work-in-progress on daisy drive systems in order to further improve the quality of technical work, the quality of risk assessments and especially to improve the likelihood of regulatory acceptance of applications.

OPTION 3.2.12.2.1 Organize and hold technical workshop.

Conduct a meeting of scientists, regulators, and a subset of local leaders from target communities to be held in Washington, DC, with the goals of identifying critical technical concerns environmental concerns, and knowledge/training needs, and building a roster of local civil society, regulators, and scientists in preparation for local community meetings that will discuss proposed technical applications and safeguards. Include preparation of briefing materials, selection of venue, identification and invitation of participants, arrangement of travel, and direction of the workshop.* *

OPTION 3.2.12.2.2 Report on technical workshop*.*

Deliverable: Report on technical, regulatory, and social challenges and recommendations on testing, integration of key recommendations into technical development and testing programs and risk assessment. In particular, recommendations to guide safety/stability/spread tests in nematodes and mosquito cage trials.

3.2.12.3 Convene and run local daisy drive meeting in a DARPA PM and LEEDR Panel approved location determined in Phase I.

3.2.12.3.1 Upon DARPA PM approval, with LEEDR Panel guidance, and in coordination with existing local engagement efforts, introduce in an even-handed, non-advocacy manner the idea of daisy suppression and quorum drive systems, including the best information about potential benefits and risks understood to date, to local communities, actively invite concerns and criticism from local community members, use the expressed concerns to direct safety research and risk assessment, and ensure that future meetings and discussions of gene drive technology use include discussions of daisy drive. Ideally, at least one team member will participate in future meetings.

3.2.12.3.2 Participate in or hold local community engagement/involvement meeting.

Upon DARPA PM approval, with LEEDR Panel guidance and in coordination with existing local engagement efforts, attend or convene meeting with local and external civil society, regulators and scientists at a location previously determined in Phase I. Obtain feedback on technical developments to date, on the risk analysis for proposed daisy drive applications, on the ethical framework for community guidance of the research, and on prospective field trials. Sub-tasks include preparation of briefing materials, selecting venue, identifying and inviting participants, arranging travel, and directing the meeting. To save funds, this may be a collaboration with other meetings on Safe Genes technology topics.

3.2.12.3.3 Report on local meeting.

Deliverables: Conduct meeting and deliver a meeting report detailing recommendations guiding technical development and field trials beyond Safe Genes; responses to address the technical and community concerns and opportunities that arose at the meeting.

3.2.12.4 Conceptual Models.

3.2.12.4.1 Outline case study: Selected Island Location

Following Kapucsinki and Dana on Problem Formulation and Options Assessment Methodology (PFOA) on improving the quality of technical risk assessments through representative stakeholder engagement, the risk assessment team shall work with local ecologists and interested community members to parameterize ecological settings put forth through group consensus as field trial candidates that will allow optimal testing of technologies developed in this proposal, identify species that interact with the target populations, develop potentially feasible conditions of release, and list potential interactions for each. Ideally, the risk assessment team will attend at least one local meeting on mosquito removal sponsored by an outside group early in Phase 2 to further this task.

3.2.12.4.2 Literature reviews.

Gather available data on population dynamics, habitat requirements, characteristics, trophic interactions for target species and off-target species leading to potential unintended consequences. Consult with local ecologists and regulators frequently, as some data may not be publicly available.

3.2.12.4.3 Conceptual models.

Develop conceptual models and identify endpoints/data for quantitative risk estimation. Conceptual models are graphical depictions of relationships among model components; also referred to as influence diagrams. Models shall be initially qualitative, but shall ultimately form the basis of quantitative connections between network nodes in the risk model.

Deliverables: Database of relevant literature by category; graphical conceptual models (figures) detailing risk pathways to serve as the basis for future quantitative assessments.

3.2.12.5 Risk analysis.

3.2.12.5.1 Parameterize risk model for selected island location.

Using the results from previous tasks, develop quantitative fault trees and Bayesian networks based on data from the literature reviews and informed by local meetings. Build influence diagrams that quantify relationships across nodes in the risk model. Incorporate available data from mathematical modelling, safety testing such as stability data, and from the ethical analyses as appropriate.* *The model shall inform tests of mosquito drive system spread in the nematode experiments.

3.2.12.5.2 Develop probabilistic estimates of potential impact.

Final model development and quantitative results. Results presented in terms of probabilities of unintended (adverse) consequences (i.e., risk).

3.2.12.5.3 Present the risk assessment at a meeting in in coordination with DARPA PM, LEEDR Panel, regulators, and local organizers

Deliverables: Draft and final reports outlining risk methodology, data sources, and results; one or more peer-reviewed manuscripts. Deliverables shall be capable of serving as a framework for future risk assessments of daisy drive applications in other contexts. Model shall be delivered in software form (likely Netica but may be MatLab or Analytica).

3.2.12.6 Ethics consultation.

3.2.12.6.1 Provide ethics consultation to MIT’s Safe Genes members.

Continuous ethics consultation. Technologies involving CRISPR and gene drive are changing so rapidly that new developments posing ethical challenges are certain to arise over the course of the Safe Genes daisy drive effort. This continuous activity shall aim to identify and address issues and dilemmas by assisting the team in deciding ethically preferred courses of action. An embedded person with expertise in social, legal and ethical implications of technology (“ethicist”) who has already earned the trust of the researchers will have many advantages in this role relative to an external panel.

3.2.12.6.2 Engage with DARPA LEEDR Panel.

The ethicist shall be the primary point of contact with the external LEEDR Panel in addressing issues identified by team members or the outside panel and assist in deciding ethically preferred courses of action in case of dilemmas, which cannot be predicted at the start of the project. As an embedded team member, the ethicist shall help mediate in the event of disagreements between the rest of the team and the external panel. These could readily have implications for the conduct and eventual success or failure of the research and likelihood of community acceptance, but necessarily cannot be quantified.

Deliverables: Continuous engagement and mediation between the team and the external LEEDR Panel. The ethicist shall engage the external LEEDR Panel at least twice yearly, plus whenever specific unanticipated issues arise. Key deliverables shall be records of these discussions and summaries of key issues including changes to the technical work plan. Additionally, a white paper shall report on the successes and challenges of embedding ethicists within research teams, which may advise DARPA's choice of ELSI involvement in future.

3.2.12.7 Regulatory engagement regarding daisy drive mosquitoes.

3.2.12.7.1 Periodic meetings with regulators (e.g., FDA, USDA, EPA) and U.S. Fish and Wildlife.

Engage with federal regulators and community-level regulatory bodies relevant to organisms employing daisy drive technology at least biannually, or more frequently as appropriate. The meetings shall focus on sharing programmatic and technical progress of the daisy project under the Safe Genes program and the construction of a framework for evaluation of future applications in mosquitoes, but shall also apply to other organisms as appropriate. The goal of meetings with regulators is to allow for integrated planning of experiments and risk assessments that shall be required to be in compliance with laws relevant to the application. The regulators shall be updated on progress and facilitate changes to technology development and risk assessment as necessary to increase the likelihood that products arising from Safe Genes will obtain regulatory approval.

3.2.12.7.2 Refine a framework for regulatory engagement and potential approval for the use of daisy drive systems. Work with regulators to refine the plan for using model systems to answer regulatory questions concerning the evolutionary stability and likely outcomes of deployed daisy drive systems in light of technical results. Ensure technological innovations (daisy drive systems) comply with regulatory rules and are suitable for use in specific locations in the event that there is sufficiently broad community support for such trials.

Deliverables: Records of meetings with regulators. An updated framework, now informed by data, outlining how model systems and laboratory experimentation may be used to answer critical regulatory questions concerning the safety and stability of gene drive systems. An updated concrete plan to streamline technological development of daisy drive systems in mosquitoes and other organisms to enable regulatory approval.

**4.0 Milestones **

4.1 BASE PERIOD (PHASE 1)

4.1.1 Mathematical Modeling (TA1 Task1 Phase1)

4.1.1.1 Thoroughly analyze baseline, deterministic well-mixed (DW) model of daisy drive.

4.1.1.1.1 Month 6: Understand sensitivities of CRISPR/repair parameters.

4.1.1.1.2 Month 6: Understand sensitivities of fitness parameters.

4.1.1.1.3 Month 6: Understand sensitivities of population-level parameters.

4.1.1.2 Analyze the importance of off-target cutting.

4.1.1.2.1 Month 9: Understand sensitivities of drives to off-target cutting.

4.1.1.2.2 Month 9: Understand the effects of driven off-target changes.

4.1.1.2.3 Month 9: Research outcomes: Determination of off-target cutting relevance.

4.1.1.3 Evaluate the importance of existing or created drive-resistant alleles.

4.1.1.3.1 Month 12: Extended basic daisy drive model (DW) to include resistant alleles (DW+R).

4.1.1.4 Study daisy-field and underdominance architectures.

4.1.1.4.1 Month 15: Extended DW model to include an arbitrary number of parallel daisy elements.

4.1.1.4.2 Month 18: Extended DW model to include underdominance at the cargo element.

4.1.1.4.3 Month 12: Iterative refinement of the DW model based on experimental results.

4.1.1.4.3 Month 18: Deliverables: Model extensions including daisyfield (DW+F) and quorum/underdominance (DW+Q) architectures.

4.1.1.5 Construct models for multiple-population nematode experiments.

4.1.1.5.1 Month 24: Metapopulation extensions of previous models.

4.1.1.5.2 Month 24: Determination of spatial parameter regimes for experimentation.

4.1.1.5.3 Month 12: Iterative refinement of metapopulation models based on experimental results.

4.1.1.5.4 Month 24: Deliverables: Models that directly connect earlier insights to nematode work, thereby informing experiments and refining models.

4.1.2 Mathematical Modelling (TA2 Task 1)

4.1.2.1 Construct genetic remediation models in well-mixed populations.

4.1.2.1.1 Month 20: Extend the DW model to encompass population suppression (DW+S).

4.1.2.1.2 Month 22: Extend the DW model to include immunizing reversal (DW+IR).

4.1.2.1.3 Month 22: *Deliverables: *Model extensions (DW+S, DW+IR) which assess genetic remediation strategies in well-mixed populations.

4.1.2.2 Assess genetic remediation strategies in linked populations with gene flow.

4.1.2.2.1 Month 21: Assess targeted daisy suppression in linked populations.

4.1.2.2.2 Month 24: Assess daisy immunizing reversal in linked populations.

4.1.2.2.3 Month 24: Research Outcomes:** **Assessment of genetic remediation strategies in linked, well-mixed populations.

4.1.3 Nematodes (TA1 Task 2)

4.1.3.1 Whole-genome sequence and assembly of C. brenneri lab strain.

4.1.3.1.1 Month 3: Deliverable:** *Assembled genome of C. brenneri with >30X coverage. *

4.1.3.2 Characterize the necessary components for daisy drive.

4.1.3.2.1 Month 2: Marker & docking strain for testing germline transgene expression.

4.1.3.2.2 Month 4: Optimize multiplexing via tRNA-processing for SpCas9 sgRNAs.

4.1.3.2.3 Month 4: Test multiplexing via Cpf1 crRNA-processing.

4.1.3.2.4 Month 5: Make strains expressing SpCas9, Cpf1, or SpCas9-2A-Cpf1.

4.1.3.2.5 Month 6: Create transgenic strains to test Pol III promoters.

4.1.3.2.6 Month 8: Quantify Pol III promoter efficacy.

4.1.3.2.7 Month 9: Quantify fitness costs from nuclease/guide RNA expression and off-target cuts.

4.1.3.2.8 Month 11: Quantify off-target cutting.

4.1.3.2.9 Month 15: Create mutants with varying non-homologous end joining (NHEJ) rates by modulating cku-80 expression.

4.1.3.2.10 Month 15: Measure brood sizes and identify genes to edit to match mosquito reproduction.

4.1.3.2.11 Month 15: Deliverable: Methods of expressing at least four guide RNAs from a single Pol III promoter.

4.1.3.3 Germline nuclease expression and ribosomal recoding for stability and precision.

4.1.3.3.1 Month 8: Create 10 strains with recoded ribosomal genes.

4.1.3.3.2 Month 10: Evaluate fitness of recoded ribosomal.

4.1.3.3.3 Month 11: Create 5 transgenic strains expressing SpCas9 in germline.

4.1.3.3.4 Month 11: Create 5 strains expressing Cpf1 in the germline.

4.1.3.3.5 Month 11: Create 5 strains expressing SpCas9-2A-Cpf1 in the germline.

4.1.3.3.6 Month 12: Test nuclease activity and ribosomal haplo-insufficiency during gametogenesis.

4.1.3.3.7 Month 15: Test underdominance.

4.1.3.3.8 Month 15: Deliverables:** **Strains with germline-expressed nuclease that cut target genes at >98%.

4.1.3.4 Build daisy drive and daisyfield systems

4.1.3.4.1 Month 13: Generate nuclease cargos for daisy drive and daisy field.

4.1.3.4.2 Month 15: Generate nuclease cargos for daisy quorum/underdominance.

4.1.3.4.3 Month 16: Generate neutral site daisy elements.

4.1.3.4.4 Month 16: Generate recoded ribosomal daisy elements.

4.1.3.4.5 Month 16: Generate daisyfield elements.

4.1.3.4.6 Month 17: Cross daisy elements and cargos to generate intact daisy drive systems.

4.1.3.4.7 Month 18: Deliverables:** **At least one daisy-chain drive system with 3+ elements.

4.1.3.5 Build massively linked populations using liquid-handling robot.

4.1.3.5.1 Month 12: Stably propagate labeled nematodes on the robot.

4.1.3.5.2 Month 13: Quantify nematode populations on the robot.

4.1.3.5.3 Month 14: Repeat an earlier competitive fitness experiment on the robot.

4.1.3.5.4 Month 16: Simulate a metapopulation with predetermined gene flow rates on the robot.

4.1.3.5.5 Month 16: Deliverables:** **System to test spatial models and daisy quorum in linked metapopulations over many generations.

4.1.3.6 Test daisy drive stability and dynamics in large populations and metapopulations.

4.1.3.6.1 Month 24: Evaluate daisy drive system stability in large populations.

4.1.3.6.2 Month 21: Build daisy-chain drive systems with 5+ elements.

4.1.3.6.3 Month 22: Build a daisy quorum drive system.

4.1.3.6.4 Month 24: Early tests of drive system spread in metapopulations.

4.1.3.6.5 Month 24: Deliverables: Daisy-chain and daisy-field drive systems with minimum quantified stability.

**4.1.4 Nematodes (TA2 Task 2) **

4.1.4.1 Build and overwrite “rogue drives” with immunizing reversal drives.

4.1.4.1.1 Month 7: Create functional molecularly confined “rogue drive” systems.

4.1.4.1.2 Month 9: Create reversal and immunizing reversal drives.

4.1.4.1.3 Month 11: Measure rogue, reversal, and immunizing reversal drive inheritance.

4.1.4.1.4 Month 18: Build and test a daisy immunizing reversal element.

4.1.4.1.4 Month 18: *Deliverables: *A 'rogue' global drive system inherited at >90% in strains with a targeted synthetic BFP gene and 0% otherwise, with fitness cost s<0.2 as measured by brood size.

4.1.4.2 Build and test components for a daisy suppression drive system.

4.1.4.2.1 Month 12: Build drive elements that disrupt recessive sex-specific infertility genes.

4.1.4.2.2 Month 18: Demonstrate biased inheritance in the presence of a nuclease element.

4.1.4.2.2 Month 18: Deliverables:** **At least one identified sex-specific fertility gene, and a self-targeting guide RNA element disrupting that gene which is inherited at >90% by the progeny of worms encoding the corresponding nuclease gene.

4.1.4.3 Build functional daisy suppression drive system.

4.1.4.3.1 Month 19: Construct a complete daisy suppression drive.

4.1.4.3.2 Month 21: Test direct daisy suppression on wild-type nematode populations.

4.1.4.3.3 Month 24: Test suppression of populations altered by daisy quorum systems.

4.1.4.3.4 Month 24: Deliverables: At least one daisy suppression drive system capable of reducing a population by 99% upon being introduced at a frequency <20%.

4.1.5 High-Throughput Transgenesis in Nematodes (TA1 Task 3)

4.1.5.1 Develop next generation microinjection platform using nematodes.

4.1.5.1.1 Month 1: Design & fabricate anti-vibration hardware mounting platform.

4.1.5.1.2 Month 2: Computer control of ultrasonic robotic manipulator motion.

4.1.5.1.3 Month 2: Dynamic control system for fast-switching of air pressure levels.

4.1.5.1.4 Month 3: Software interface to control microinjection platform.

4.1.5.1.5 Month 4: Safety enclosure with temperature control.

4.1.5.1.6 Month 5: Safety mechanism that disengages air pressure/robotics.

4.1.5.1.7 Month 6: Refine user software interface for ease of use and efficiency.

4.1.5.1.8 Month 6: Create training materials for operating the machine and provide on-site training and assistance to the Esvelt Lab with operation.

4.1.5.1.9 Month 6: Deliverables: A computer-assisted system capable of injecting 2 worms/minute.

4.1.5.2 Enhanced automation using needle alignment to camera, piezo-based penetration.

4.1.5.2.1 Month 8: Continue providing on-site assistance to project team members.

4.1.5.2.2 Month 8: Replicate first system to begin next-gen platform development.

4.1.5.2.3 Month 9: Develop on-axis needle penetration by piezo-based vibration.

4.1.5.2.4 Month 10: Algorithm: automated microneedle alignment to camera.

4.1.5.2.5 Month 12: Optimize quartz microneedle shape & settings for delicacy during microinjections.

4.1.5.2.6 Month 12: Deliverables: A microinjection system with piezo-based jackhammer penetration with automated algorithm for quartz needle alignment to camera.

4.1.5.2.7 Month 14: Algorithm: detect organism position in low magnification.

4.1.5.2.8 Month 16: Algorithm: high magnification 3D image stack of target.

4.1.5.2.9 Month 17: Algorithm: detect/respond to clogged needles with pressure.

4.1.5.2.10 Month 18: Algorithm: insert microneedle deliver fluid w/ video feedback.

4.1.5.2.11 Month 18: Develop rotation stage to orient target for microinjection.

4.1.5.2.12 Month 18: Deliverables: A high-throughput user-operated platform that can microinject 3 worms/minute*. *

4.1.5.2.13 Month 20: System to safely degrade DNA by heat inside the microneedle.

4.1.5.2.14 Month 22: Piezo-based syringe to front-load plasmid into microneedle.

4.1.5.2.15 Month 23: Post-injection recovery technique for optimal survival.

4.1.5.2.16 Month 24: Refine techniques for transformation efficiency & speed.

4.1.5.2.17 Month 24: Optimize user software interface for ease of use & efficiency.

4.1.5.2.18 Month 24: Create training materials and on-site support to project team members.

4.1.5.2.19 Month 24: Deliverables: High-throughput platform capable of microinjecting 4 worms/minute.

4.1.6 Aedes aegypti (TA1 Task 4)

4.1.6.1 Optimize CRISPR guide RNA expression and multiplexing in vitro.

4.1.6.1.1 Month 6: Characterize pol III promoters.

4.1.6.1.2 Month 9: Characterize tRNA- and crRNA-based multiplexing.

4.1.6.1.3 Month 9: Characterize guide RNAs targeting sites of interest.

4.1.6.1.4 Month 9: Research Outcomes: Data on sgRNA sufficient to predict via modelling (Task 1) performance of construct daisy drive systems constructed using these components.

4.1.6.2 Optimize germline CRISPR nuclease expression conditions.

4.1.6.2.1 Month 15: Generate transgenic strains expressing Cas9 or Cpf1.

4.1.6.2.2 Month 18: Characterize expression of strains expressing Cas9 or Cpf1 in the germline.

4.1.6.2.3 Month 21: Research Outcome: CRISPR nuclease activity in the germline demonstrated by detecting products of cutting and/or homing arising in offspring of such individuals.

4.1.6.3 Identify ribosomal gene recodings & express nucleases.

4.1.6.3.1 Month 21: Identify viable recodings.

4.1.6.3.2 Month 24: Encode germline-expressing nucleases in recoded strains.

4.1.6.3.3 Month 24: Deliverables: At least two strains with recoded ribosomal genes, and at least one strain that expresses a nuclease in the germline with the nuclease gene near a recoded ribosomal gene.

4.1.6.4 Develop a model daisy drive cargo element.

4.1.6.4.1 Month 15: Generate transgenic line incorporating sgRNA construct into model target gene.

4.1.6.4.2 Month 18: Combine strains from above to provide daisy drive strain(s).

4.1.6.4.3 Month 24: Small cage proof of concept demonstration of daisy drive.

4.1.6.4.4 Month 24: Deliverables: Daisy drive components shown in combination to be capable of base-element-dependent drive of subsequent elements.

3.1.7 Culex quinquefasciatus (TA1 Task 5)

4.1.7.1 Genome sequence and assembly of *Culex quinquefasciatus *lab strain

4.1.7.1.1 Month 3: Sequence genome to enable correct targeting.

4.1.7.1.2 Month 3: Deliverable: Sequenced and assembled genome at >30x coverage.** **

4.1.7.2 Optimize CRISPR guide RNA expression and multiplexing in vitro.

4.1.7.2.1 Month 9: Characterize pol III promoters.

4.1.7.2.2 Month 12: Characterize tRNA- and crRNA-based multiplexing.

4.1.7.2.3 Month 18: Characterize guide RNAs targeting sites of interest.

4.1.7.2.4 Month 18: Research Outcomes: Data on sgRNA sufficient to predict via modelling performance (Task 1) of constructed daisy drive systems using these components.

4.1.7.3 Express CRISPR nuclease in Culex germline.

4.1.7.3.1 Month 20: Identify candidate germline promoters.

4.1.7.3.2 Month 24: Generate strains expressing Cas9 or Cpf1.

4.1.7.3.3 Month 24: Deliverables: At least one transgenic line expressing either Cas9 or Cpf1 in the germline.

4.1.8 Anopheles gambiae (TA1 Task 6) – (OPTIONAL TASKING)

4.1.8.1 Optimize CRISPR guide RNA expression and multiplexing in vitro.

4.1.8.1.1 Month 9: Characterize pol III promoters.

4.1.8.1.2: Month 12: Characterize tRNA- and crRNA-based multiplexing.

4.1.8.1.3 Month 18: Characterize guide RNAs targeting sites of interest.

4.1.8.1.4 Month 18: Research Outcomes: Data on sgRNA sufficient to predict via modelling (Task 1) performance of constructed daisy drive systems constructed using these components.

4.1.8.2 Express CRISPR nuclease in Anopheles germline

4.1.8.2.1 Month 20: Identify candidate germline promoters.

4.1.8.2.2 Month 24: Generate strains expressing Cas9 or Cpf1.

4.1.8.2.3 Month 24: Deliverables: At least one transgenic line expressing either Cas9 or Cpf1 in the germline.

4.1.9 Data Sharing (TA1 Task 7)

4.1.9.1 Deliver experimental data collected in this effort within four months of completion of data collection.

4.1.9.2 Month 12: Post all data collected up until month 9 into a data sharing platform to be identified by the PM.

4.1.9.3 Month 18: Post all data collected up until month 15 into a data sharing platform to be identified by the PM.

4.1.9.4 Month 24: Post all data collected up until month 21 into a data sharing platform to be identified by the PM.

Deliverable: Regular updates of experimental plans, data, analyses, and lay summaries.

4.1.10 Ethical, Legal, Social and Risk Analysis (TA1 Task 8)

4.1.10.1 Develop a framework for a technical and regulatory workshop in Washington, DC.

4.1.10.1.1 Month 1: Framework for technical and regulatory workshop in DC.

4.1.10.2 Technical and regulatory workshop in Washington, DC. (OPTIONAL TASKING)

4.1.10.2.1 Month 3: Organize and hold technical/regulatory workshop.

4.1.10.2.2 Month 6: Report on technical/regulatory workshop.

4.1.10.2.3 Month 6: Deliverable: Report on technical, regulatory, and social challenges and recommendations on testing, integration of key recommendations into technical development and testing programs and risk assessment.

4.1.10.3 Coordinate with DARPA LEEDR Panel on local outreach efforts in a predetermined location from Phase I.

4.1.10.3.1 Month 6: Engage with DARPA Safe Genes PM, LEEDR Panel, and existing local outreach effort organizers to create an engagement framework.

4.1.10.3.2 Month 18: Deliverables: A detailed framework on proposed local engagement, including plans for integration with existing engagement efforts.

4.1.10.3.3 Month 22: Upon DARPA PM approval and with LEEDR Panel guidance, hold meeting with existing outreach organizers in the chosen location to discuss their current engagement efforts and information regarding local and external civil society, regulators and scientists. Upon DARPA PM approval and with LEEDR Panel guidance, implement engagement strategy determined in 4.1.10.3.2.

4.1.10.3.4* Month 24: Deliverables:* A meeting report detailing local outreach organizer recommendations on testing and on local concerns; integration of key recommendations into technical development and risk assessment.

4.1.10.4 Ethics of ecological engineering technologies and gene drives.

4.1.10.4.1 Month 6: Literature review on ecological engineering technologies and ethics.

4.1.10.4.2 Month 6: Outline the key questions in ethics and gene drive technologies.

4.1.10.4.3 Month 7: Identify specifics of daisy drive systems and ethical relevance.

4.1.10.4.4 Month 7: Deliverable: Ethical framework specific to the development and deployment of daisy drive systems.

4.1.10.5 Ethics consultation

4.1.10.5.1 Month 24: Provide ethics consultation to Recipient’s Safe Genes members.

4.1.10.5.2 Month 24: Research Outcomes: Continuous consultation service and advice during all project meetings. Reports on ethical issues as they arise during meetings.

4.1.10.5.3 Month 24: Engage with DARPA LEEDR Panel.

4.1.10.5.4 Month 24: Research Outcomes: Continuous engagement and mediation between the team and the external LEEDR Panel.

4.1.10.6 Regulatory engagement.

4.1.10.6.1 Every 4 to 6 months: Periodic meetings with regulators from FDA, U.S. Department of Fish and Wildlife, and community regulatory bodies.

4.1.10.6.2 Month 18: Deliverables: Records of meetings with regulators.

4.2 OPTION PERIOD (PHASE 2)

4.2.1 Mathematical Modelling. (TA2 Task 1 Phase 2)

4.2.1.2 Assess complete genetic remediation strategies in well-mixed populations.

4.2.1.2.1 Month 25: Assess daisy immunizing reversal with quorum (DW+IRQ).

4.2.1.2.2 Month 26: Assess genetic remediation of a rogue drive system (DW+IRQ+S).

4.2.1.2.3 Month 26: Iteratively refine models based on experimental results.

4.2.1.2.4 Month 30: *Deliverables: *Model extensions (DW+IRQ, DW+IRQ+S) which assess genetic remediation strategies in well-mixed populations.

4.2.1.3 Refine well-mixed metapopulation models to study candidate island release.

4.2.1.3.1 Month 32: Compile mosquito abundance/migration data from literature.

4.2.1.3.2 Month 32: Fine-tune predictive models with empirical mosquito/nematode parameters.

4.2.1.3.3 Month 36: Deliverables: Linked panmictic metapopulation models of daisy drive systems to predict outcomes of various potential release scenarios, to be updated at 48mo with additional nematode and mosquito data.

4.2.1.4 Construct spatially explicit daisy drive modeling frameworks.

4.2.1.4.1 Month 34: Deterministic reaction-diffusion models.

4.2.1.4.2 Month 38: Stochastic spatial models.

4.2.1.4.3 Month 38: Deliverables: Extensions of the DW+ models to include explicit spatial dynamics.

4.2.1.5 Assess genetic remediation strategies in spatially explicit models.

4.2.1.5.1 Month 40: Assess targeted daisy suppression.

4.2.1.5.2 Month 42: Assess daisy immunizing reversal with quorum in spatially explicit systems.

4.2.1.5.3 Month 42: Deliverables: Evaluation of the capacity of daisy suppression alone to slow or completely eliminate and remediate engineered genes or rogue drives in spatial systems.

4.2.1.6 Refine spatially explicit models to study local island release.

4.2.1.6.1 Month 44: Compile mosquito abundance/migration data from literature.

4.2.1.6.2 Month 48: Fine-tune predictive models with empirical mosquito parameters.

4.2.1.6.2 Month 48: Deliverables: Spatially explicit models of local island daisy drive systems.

**4.2.2 Nematodes (TA1 Task 2 Phase 2) **

4.2.2.1 Empirical optimization of daisy-chain and daisy-drive systems.

4.2.2.1.1 Month 27: Build daisy-chain drive systems with 7+ elements.

4.2.2.1.2 Month 30: Build parallel daisy-chain systems to evade resistance.

4.2.2.1.3 Month 30: Build daisy-field drive systems with 32+ elements.

4.2.2.1.4 Month 31: Build strains to maintain and propagate daisy-field.

4.2.2.1.5 Month 32: Create extended daisy-chain and high-copy daisy-field quorum strains. Cross the quorum/underdominance strains with daisy-chain and daisy-field systems.

4.2.2.1.6 Month 32: Deliverables: A daisy-chain drive system with 7+ elements, with and without quorum. A daisy-field drive system with 32+ elements, with and without quorum. At least 3 strains expressing different guide RNAs targeting the repeat sequence replaced by daisy-field elements.

4.2.2.2 Long-term population genetics comparisons of daisy drive systems.

4.2.2.2.1 Month 34: Rigorously test stability of optimized strains.

4.2.2.2.2 Month 36: Long-term metapopulation studies of daisy-chain systems.

4.2.2.2.3 Month 38: Long-term metapopulation studies of daisy-field systems.

4.2.2.2.4 Month 42: Long-term metapopulation studies of daisy systems with quorum.

4.2.2.2.4 Month 42: Deliverables: Minimum stability metrics of daisy-chain drive systems with respect to potential recombination into a global drive daisy necklace.

4.2.2.3 Empirical models of mosquito drive systems in nematodes.

4.2.2.3.1 Month 33: Construct nematode equivalents of mosquito species.

4.2.2.3.2 Month 38: Construct nematode daisy drive equivalent of mosquito drive systems.

4.2.2.3.3 Month 40: Quantify minimal stability of mosquito daisy drive systems.

4.2.2.3.4 Month 48: Test mosquito daisy drive dynamics in nematodes.

4.2.2.3.4 Month 48: Deliverables: Nematode strains engineered to correspond to each mosquito species by HDR rate, brood size, and sequences targeted by daisy drive systems.

4.2.3 Genetic remediation and restoration. (TA2 Task 2 Phase 2)

4.2.3.1 Genetic remediation and restoration.

4.2.3.1.1 Month 27: Test genetic remediation of daisy quorum to wild-type via daisy suppression.

4.2.3.1.2 Month 30: Test genetic remediation of an engineered gene to wild-type.

4.2.3.1.3 Month 32: Test daisy suppression as a method of eliminating a rogue drive system.

4.2.3.1.4 Month 36: Test daisy immunizing reversal as a method of eliminating a rogue drive system.

4.2.3.1.5 Month 40: Test daisy immunizing reversal with quorum vs rogue drive, then suppression.

4.2.3.1.6 Month 40: Deliverables: Experimental quantification of the number of organisms that must be released in a particular spatial distribution to restore populations altered with introduced engineered genes or with rogue drive systems to wild-type genetics.

**4.2.4 Aedes aegypti. (TA1 Task 3 Phase 2) **

4.2.4.1 Create daisy drive elements and strains.

4.2.4.1.1 Month 30: Assess fitness of recoded strains expressing nucleases (life history).

4.2.4.1.2 Month 36: Assess fitness of recoded strains expressing nucleases (allele frequency evolution).

4.2.4.1.3 Month 36: Insert guide RNA gene(s) into neutral or recoded loci.

4.2.4.1.4 Month 39: Generate daisy drive platform strains.

4.2.4.1.5 Month 39: Deliverables: 1+ strain with recoded homozygous viable ribosomal gene(s), 1+ strains that express CRISPR nucleases in the germline adjacent to a recoded ribosomal gene, 1+ daisy drive strains with 2+ elements.

4.2.4.2 Assess impact of off-target endonuclease activity

4.2.4.2.1 Month 30: Assess rate of mutagenesis at target sites.

4.2.4.2.2 Month 42: Assess rate of cutting of off-target sites.

4.2.4.2.3 Month 42: Deliverables: Quantitative data on rates of induced sequence change at target and non-target sites.

4.2.5 Genetic load daisy elements for suppression and remediation. (TA2 Task 3 Phase 2)

4.2.5.1 Create genetic load daisy elements for suppression and remediation.

4.2.5.1.1 Month 30: Identify candidate female fertility genes.

4.2.5.1.2 Month 36: Validate putative female fertility genes.

4.2.5.1.3 Month 39: Insert guide RNA expression constructs in female fertility genes.

4.2.5.1.4 Month 42: Assess fertility/fitness effects of mutations/insertions into female fertility genes.

4.2.5.1.5 Month 42: Deliverables: 1+ strains with a guide RNA cassette disrupting a female fertility gene and measurements of fertility and fitness for hetero/homozygotes of both sexes.

4.2.5.2 Genetic load daisy drive strains for suppression and remediation

4.2.5.2.1 Month 39: Insert guide RNAs at Male-determining (M) locus.

4.2.5.2.2 Month 42: Build a daisy suppression strain.

4.2.5.2.3 Month 48: Test population suppression.

4.2.5.2.3 Month 48: Deliverables: 1+ daisy drive strains with 3+ elements, assessment of spread of daisy drive elements within isolated populations and suppression via genetic load.

**4.2.6 Culex quinquefasciatus. (TA1 Task 4 Phase 2) **

4.2.6.1 Culex quinquefasciatus.

4.2.6.1.1 Month 30: Characterize expression of strains designed to express Cas9 or Cpf1 in the germline.

4.2.6.1.2 Month 30: Research Outcomes: CRISPR nuclease activity in the germline demonstrated by detecting products of cutting and/or homing arising in offspring of such individuals.

4.2.6.2 Identify ribosomal gene recodings & express nucleases.

4.2.6.2.1 Month 30: Identify homozygous viable recodings.

4.2.6.2.2 Month 36: Encode germline-expressing nucleases in recoded strains.

4.2.6.2.3 Month 36: Deliverables: At least one strain with recoded ribosomal genes, and at least one strain that expresses a nuclease in the germline with the nuclease gene near a recoded ribosomal gene.

4.2.6.3 Create daisy drive elements and strains.

4.2.6.3.1 Month 42: Assess fitness of recoded strains expressing nucleases (life history parameters).

4.2.6.3.2 Month 42: Assess fitness of recoded strains expressing nucleases (allele frequency evolution).

4.2.6.3.3 Month 42: Insert guide RNAs into neutral or recoded loci.

4.2.6.3.4 Month 45: Generate daisy drive platform strains.

4.2.6.3.5 Month 45: Deliverables: 1+ strain with recoded homozygous viable ribosomal gene(s), 1+ strains that express CRISPR nucleases in the germline adjacent to a recoded ribosomal gene, 1+ daisy drive strains with 2+ elements.

4.2.7 Genetic load daisy elements for suppression and remediation. (TA2 Task 4 Phase 2)

4.2.7.1 Create genetic load daisy elements for suppression and remediation.

4.2.7.1.1 Month 27: Identify candidate female fertility genes.

4.2.7.1.2 Month 33: Validate putative female fertility genes.

4.2.7.1.3 Month 39: Insert guide RNA expression constructs into female fertility genes.

4.2.7.1.4 Month 42: Verify female infertility.

4.2.7.1.5 Month 42: Deliverables: 1+ strains with a guide RNA cassette disrupting a female fertility gene and measurements of fertility and fitness for hetero/homozygotes of both sexes.

4.2.7.2 Create genetic load daisy drive strains for suppression and remediation.

4.2.7.2.1 Month 48: Build a daisy suppression strain.

4.2.7.2.2 Month 48: Test population suppression.

4.2.7.2.3 Month 48: Deliverables: 1+ daisy drive strains with 3+ elements, optionally assessment of suppression via genetic load.

4.2.8 – Anopheles gambiae (TA1 Task 5 Phase 2) (OPTIONAL REQUIREMENT)

4.2.8.1 Identify germline nuclease expression signals.

4.2.8.1.1 Month 30: Characterize expression of strains designed to express Cas9 or Cpf1 in the germline.

4.2.8.1.2 Month 42: Deliverables: CRISPR nuclease activity in the germline demonstrated by detecting products of cutting and/or homing arising in offspring of such individuals** **

4.2.8.2 Identify ribosomal gene recodings & express nucleases.

4.2.8.2.1 Month 30: Identify homozygous viable recodings.

4.2.8.2.2 Month 36: Encode germline-expressing nucleases in recoded strains.

4.2.8.2.3 Month 36: Deliverables: At least one strain with recoded ribosomal genes, and at least one strain that expresses a nuclease in the germline with the nuclease gene near a recoded ribosomal gene.

4.2.8.3 Create daisy drive elements and strains.

4.2.8.3.1 Month 42: Assess fitness of recoded strains expressing nucleases (life history parameters).

4.2.8.3.2 Month 42: Assess fitness of recoded strains expressing nucleases (allele frequency evolution).

4.2.8.3.3 Month 42: Insert guide RNAs into neutral or recoded loci.

4.2.8.3.4 Month 45: Generate daisy drive platform strains.

4.2.8.3.5 Month 45: Deliverables: 1+ strain with recoded homozygous viable ribosomal gene(s), 1+ strains that express CRISPR nucleases in the germline adjacent to a recoded ribosomal gene, 1+ daisy drive strains with 2+ elements.

4.2.9 Genetic load daisy elements for suppression and remediation (TA2 Task 5 Phase2)

4.2.9.1 Create genetic load daisy elements for suppression and remediation.

4.2.9.1.1 Month 27: Identify candidate female fertility genes.

4.2.9.1.2 Month 33: Validate putative female fertility genes.

4.2.9.1.3 Month 39: Insert guide RNA expression constructs into female fertility genes.

4.2.9.1.4 Month 42: Verify female infertility.

4.2.9.1.5 Month 42: Deliverables: 1+ strains with a guide RNA cassette disrupting a female fertility gene and measurements of fertility and fitness for hetero/homozygotes of both sexes.

4.2.9.2 Create genetic load daisy drive strains for suppression and remediation.

4.2.9.2.1 Month 39: Insert guide RNAs at Male-determining (M) locus (optional).

4.2.9.2.2 Month 48: Build a daisy suppression strain.

4.2.9.2.3 Month 48: Test population suppression.

4.2.9.2.3 Month 48: Deliverables: 1+ daisy drive strains with 3+ elements, optionally assessment of suppression via genetic load.

4.2.10 High-Throughput transgenesis in mosquitoes (TA1 Task 6 Phase 2) (OPTIONAL REQUIREMENT)

4.2.10.1 Develop high-throughput transgenesis of Aedes aegypti.

4.2.10.1.1 Month 25: Replicate second system, begin mosquito system development.

4.2.10.1.2 Month 25: Egg immobilization/mounting technique for Aedes aegypti.

4.2.10.1.3 Month 27: Optimized microneedle shape & parameters for delicacy.

4.2.10.1.4 Month 30: Algorithm to detect organism position in low magnification.

4.2.10.1.5 Month 30: Algorithm to acquire high mag 3D image stack.

4.2.10.1.6 Month 31: Algorithm: insert microneedle, deliver fluid, feedback/control.

4.2.10.1.7 Month 32: Develop post injection recovery technique: optimal survival.

4.2.10.1.8 Month 33: Refine techniques for transformation efficiency and speed.

4.2.10.1.9 Month 33: Deliverables: A high-throughput platform that can microinject 6 *Aedes aegypti *eggs/min.

4.2.10.2 Develop high-throughput transgenesis of Anopheles gambiae mosquitoes.

4.2.10.2.1 Month 28: Egg immobilization/mounting technique for Anopheles gambiae.

4.2.10.2.2 Month 30: Optimize microneedle shape & parameters for delicacy.

4.2.10.2.3 Month 33: Algorithm to detect organism position in low magnification.

4.2.10.2.4 Month 33: Algorithm to acquire high mag 3D image stack.

4.2.10.2.5 Month 34: Algorithm: insert microneedle, deliver fluid, feedback/control.

4.2.10.2.6 Month 35: Develop post injection recovery technique: optimal survival.

4.2.10.2.7 Month 36: Refine techniques for transformation efficiency and speed.

4.2.10.2.8 Month 37: Deliverables: A high-throughput platform that can microinject 6 *Anopheles gambiae *eggs/min.

4.2.10.3 High-throughput transgenesis of Culex quinquefasciatus mosquitoes

4.2.10.3.1 Month 31: Egg immobilization/mounting technique for Culex quinquefasciatus.

4.2.10.3.2 Month 33: Optimize microneedle shape & parameters for delicacy.

4.2.10.3.3 Month 36: Algorithm to detect organism position in low magnification.

4.2.10.3.4 Month 36: Algorithm to acquire high mag 3D image stack.

4.2.10.3.5 Month 37: Algorithm: insert microneedle, deliver fluid, feedback/control.

4.2.10.3.6 Month 38: Develop post injection recovery technique: optimal survival.

4.2.10.3.7 Month 39: Refine all techniques for transformation efficiency and speed.

4.2.10.3.8 Month 39: Deliverables: A high-throughput platform that can microinject 6 Culex quinquefasciatus eggs/min.

4.2.11 Data Sharing (Task 7 Phase 2)

4.2.11.1 Deliver experimental data collected in this effort within four months of completion of data collection

4.2.11.2 Month 30: Post all data collected up until month 27 into a data sharing platform to be identified by the PM.

4.2.11.3 Month 36: Post all data collected up until month 33 into a data sharing platform to be identified by the PM.

4.2.11.4 Month 42: Post all data collected up until month 39 into a data sharing platform to be identified by the PM.

4.2.11.5 Month 48: Post all data collected up until month 45 into a data sharing platform to be identified by the PM.

4.2.12 Ethical, Legal, Social and Risk Analysis (Task 8 Phase 2)

4.2.12.1 Develop a framework for the second technical and regulatory workshop in DC.

4.2.12.1.1 Month 27 Deliverables: Define outcomes from first regulatory workshop and justification for second workshop on technical safeguards for genome editing. Continue to engage with regulators to inform framework. Define how the second workshop will provide additional guidance for this Safe Genes project.

4.2.12.1.2 Obtain DARPA PM approval of framework and LEEDR Panel guidance prior to workshop planning.

4.2.12.1.3 Month 30* Deliverable*: Framework for second technical and regulatory workshop in DC.

4.2.12.2 Organize and run second technical and regulatory workshop. (OPTIONAL REQUIREMENT)

4.2.12.2.1 Month 36 Deliverable: Report on technical workshop, regulatory, and social challenges and recommendations on testing, integration of key recommendations into technical development and testing programs and risk assessment.

4.2.12.3 Conduct or participate in daisy drive meeting in coordination with DARPA PM, LEEDR Panel, and local outreach organizers.

4.2.12.3.1 Participate in or organize local community engagement/involvement meeting.

4.2.12.3.2 Month 43 Deliverable: Report on local meeting detailing recommendations guiding technical development and field trials beyond Safe Genes; responses to address the technical and community concerns and opportunities that arose at the meeting.

4.2.12.4 Conceptual Models.

4.2.12.4.1 Month 28: Outline case study for chosen island(s)

4.2.12.4.2 Month 32: Literature reviews.

4.2.12.4.3 Month 34: Conceptual models.

4.2.12.4.4 Month 36: Deliverables: Database of relevant literature by category; graphical conceptual models (figures) detailing risk pathways to serve as the basis for future quantitative assessments

4.2.12.5 Risk analysis.

4.2.12.5.1 Month 40: Parameterize risk model.

4.2.12.5.2 Month 42: Develop probabilistic estimates of potential impact.

4.2.12.5.3 Month 42: Present the risk assessment at the second local meeting. (OPTION)

4.2.12.5.4 Month 48: Deliverables: Draft and final reports outlining risk methodology, data sources, and results; one or more peer-reviewed manuscripts.

4.2.12.6 Ethics consultation.

4.2.12.6.1 Month 48: Provide ethics consultation to MIT’s Safe Genes members.

4.2.12.6.2 Month 48: Engage with DARPA LEEDR Panel.

4.2.12.6.2 Month 48: Deliverables: Continuous engagement and mediation between the team and the external LEEDR Panel.

4.2.12.7 Regulatory engagement and planning potential future field trials of daisy drive mosquitoes.

4.2.12.7.1 Month 48: Periodic meetings with regulators from local community regulatory bodies, FDA and U.S. Fish and Wildlife.

4.2.12.7.5 Month 48: Deliverables: Records of meetings with regulators. An updated framework, newly informed by data, outlining how nematodes may be used to answer critical regulatory questions concerning the safety and stability of gene drive systems.

**5.0 PROGRAM MANAGEMENT AND REVIEW **

The Government will actively monitor, review and approve the Recipient's performance to ensure all the performers are in sync and matched with the Government's requirements. The Government will ensure that each of the performers share experimental data across the program. The Government will further ensure that the performers develop techniques and capabilities that are compatible and integrate with each other. The Recipient shall collaborate and cooperate with other performers in the program under the direction of the Government. At the Government PI meeting, all performers shall demonstrate their technical capabilities and engage each other in a cooperative yet challenging environment. Along these lines, the Government will ensure that the performers each share technical information with each other to enable the testing/challenging of each other's capabilities. The Government will further oversee the program and will review, approve, and participate in the demonstrations, as well as determine which performers will advance to the Option Period (Phase 2). The Government shall review all releases to the public and/or press that reference DARPA or the Defense Department or this Program (excluding peer-reviewed publications of scientific findings). No such material shall be publically released without Government approval.

5.1 KICK-OFF MEETING

The Recipient shall participate in a kick-off meeting within 60 days of award of this cooperative agreement. In this meeting, the Recipient shall present a program management plan and financial tracking plan. The Recipient shall provide to the DARPA PM, ahead of time, presentation materials for the meeting.

5.2 MONTHLY FINANCIAL REPORTS

The Recipient shall provide monthly financial progress reports to SPAWARSYSCEN-PACIFIC (SSC-PAC) and the DARPA Biological Technologies Office (BTO). The purpose of these reports is to provide a brief on project progress and inform the DARPA BTO of any potential issues.

5.3 QUARTERLY FINANCIAL REPORTS.

The Recipient shall provide quarterly financial progress reports to the SSC-PAC and DARPA BTO. The purpose of these reports is to provide a brief project progress and inform the SSC-PAC and DARPA BTO of any potential issues.

5.4 MONTHLY TECHNICAL UPDATES

The Recipient shall provide monthly technical updates by teleconference to SSC-PAC and the DARPA BTO. The purpose of these updates is to present a summary of work completed and milestones met; discuss any problems encountered; update the program schedule; present the program financial status; and discuss remaining work regarding technical tasks, data sharing, and ELSI activities. Monthly updates shall also include all technical data items generated under this agreement, including but not limited to experimental data; processed data, methods and processed used; research reports and publications; and software (source code and executables).

5.5 QUARTERLY TECHNICAL REPORTING

The Recipient shall provide detailed technical quarterly progress reports to SSC-PAC and the DARPA BTO. The purpose of these reports is to present a summary of work completed and milestones met; discuss any problems encountered; update the program schedule; present the program financial status; and discuss remaining work regarding technical tasks, data sharing, and ELSI activities. Quarterly reports shall also include all technical data items generated under this agreement but not limited to experimental data; processed data, along with methods of processing used; research reports and publications; and software (source code and executables).

5.6 ANNUAL PI MEETING

The Recipient shall participate in annual meetings with all Principal Investigators of the Safe Genes Program. In this meeting, the Recipient shall present a summary report of all work completed and milestones accomplished and present program management and financial tracking plans. The Recipient shall provide, ahead of time, presentation materials for the meeting.

5.7 ANNUAL SITE VISITS

The Recipient shall host annual site visits of SSC-PAC and the DARPA BTO. The purpose of this visit is to present a summary report of all work completed and milestones accomplished and to discuss ongoing or future technical tasks, data sharing, and ELSI activities

**5.8 FINAL COOPERATIVE AGREEMENT REVIEW **

The Recipient shall prepare and deliver a final cooperative agreement report that completely and clearly describes all work completed, the results obtained, and the implications of those results. The Recipient shall discuss milestones accomplished and relevant and/or related future efforts that may be pursued. The Recipient shall provide a copy of the final report to SSC-PAC and the DARPA BTO.

6.0 GOVERNMENT FURNISHED EQUIPMENT (GFE) / RECIPIENT ACQUIRED PROPERTY (CAP)

7.0 TRAVEL

Travel is anticipated for meetings between collaborators, meetings with DARPA, and to professional conferences to include annual PI meetings, ELSI activities, and annual site visits.

8.0 PLACE OF PERFORMANCE

Massachusetts Institute of Technology
Office of Sponsored Programs
77 Massachusetts Ave
Building NE18-901
Cambridge, MA 02139

9.0 SECURITY

This work is unclassified.