Building Resilience Against Split-Orders

Building resilience in nucleic acid screening systems against split orders by integrating assembly pipelines into screening systems and validating resilience through test sets

Endorsements made here support IBBIS (International Biosecurity and Biosafety Initiative for Science).

Building resilience in nucleic acid screening systems against split orders by integrating assembly pipelines into screening systems and validating resilience through test sets

Endorsements made here support IBBIS (International Biosecurity and Biosafety Initiative for Science).

Project Details

Background

There is increasing concern and risk that threat actors will attempt to evade nucleic acid synthesis screening tools by splitting orders for sequences of concern (SOCs) into short oligos and assembling those fragments in vitro. It is thus a priority to test for resilience from split-order attempts and subsequently build resilience within screening systems against such threats by being able to detect SOCs from oligo pools.

Project

The first part of the project will involve collectively developing high-quality assembly test sets, utilizing the Sequence Biosecurity Risk Consortium (SBRC) test sets of full fragments to select high-consequence SOCs while taking into account various assembly methods. The second part of the project will involve building pipelines within IBBIS’ common mechanism (commec) screening tool to detect split orders and pass the test sets.

Consultants with relevant experience will be working on this project, overseen by me and another lead, within the umbrella of the SBRC, a consortium moderated by me and two other leads. The deliverables developed are listed below, and there will be regular SBRC working group meetings also overseen by senior members of the SBRC as well as a senior bioinformatician working on commec.

Deliverables

Deliverable 1: Working group meetings will be setup and consultants will attend weekly split-order working group meetings with other SBRC members who are working on developing test sets to devise methodologies and plans of action to develop assembly test sets.

Deliverable 2: Collectively develop and iterate to create high-quality split-order test sets from SOCs within the SBRC test set and benign sequences, taking into account various assembly methods (at least 3 among a prioritised set of assembly methods, including some Gibson, Golden Gate, Sidewinder, polymerase cycling assembly, Darwin) and high-consequence SOCs.

Deliverable 3: Draft a short report on the outputs and methods used by the working group.

Deliverable 4: Build an efficient pipeline (<1 min / pool) to assemble oligo pools. Share code within the SBRC and integrate within commec, passing the constructed test sets (>95% recall), while avoiding increasing false positives, to build resilience against assembly threats.

Theory of Impact

AI-enabled biodesign tools and frontier models are now increasingly capable of designing short oligos from a whole gene that users can order from a synthesis provider and subsequently assemble in vitro. Such short oligos are currently difficult to detect by current screening tools, making them prone to evasion if someone intends to assemble a sequence of concern.

This exacerbates biosecurity risk from a malicious actor being more easily able to order short oligos to assemble sequences of concern from pathogens of pandemic potential from such fragments.

This project will ensure that screening tools have robust test sets that they can evaluate against to ensure they can detect potential assembly attempts from short oligos and provide a proof of concept for resilience by building a pipeline in our own screening tool that can detect assembly attempts through split orders.

People

Rassin Lababidi

Team Member

Tessa Alexanian

Technical Lead

Grants Received– no grants recorded

Add grant

Funding Asks

grantmaking.ai Launch Round

Applied

Building resilience in nucleic acid screening systems against split orders by integrating assembly pipelines into screen systems and validating resilience through test sets

Minimum

$20,000

Ideal

$36,000

What this ask is

Background

Project

Consultants with relevant experience will be working on this project, overseen by myself and another technical lead, within the umbrella of the SBRC, a consortium moderated by me and two other leads. The deliverables developed are listed below, and there will be regular SBRC working group meetings also overseen by senior members of the SBRC as well as a senior bioinformatician working on commec.

Deliverables

Deliverable 3: Draft a short report on the outputs and methods used by the working group.

How the money will be spent

The majority of the grant will go into hiring relevant consultants (3-4) who have worked on the oligo assembly problem to deliver high-quality test sets and pipelines to build resilience against oligo assembly. ($18,000)
Budget for compute and access AI models ($2,000)
If a larger budget is given, more consultants can be hired, or we would be able to hire more senior consultants as well as dedicate time to test against other screening tools ($9,000)
- The SBRC moderators would be able to spend more dedicated hours to develop the deliverables as well if the stretch amount is given ($7,000)

Discussion

No comments yet. Be the first to share your thoughts.