Bypassing Current Biosecurity Screens with AI Designed Proteins and Closing the Gap with Edge-AI Functional Screening
- 1Yale University
Abstract
DNA synthesis machines are vital tools in medicine and research, but they can be misused to produce sequences which are transcribed and translated to produce toxic proteins. Synthesis providers may screen orders for dangerous sequences, but most screening software performs sequence-based alignment, which is vulnerable to AI redesign attacks. Additionally, software-only synthesis screening requires the operator of the machine to be trusted. We present Capiti, an edge-deployed screening system that runs with no internet connection, listens directly to a synthesizer's valve control signals, and aborts a run when it detects incipient pathogenic function. Rather than performing sequence alignment, Capiti is specifically trained to recognize fold. ProteinMPNN was used to generate putatively functional variants of known pathogenic proteins, and a CNN was trained to separate them from active-site knockouts, scrambles, and decoys. We release two models, Capiti-E (known pathogenic proteins) and Capiti-C (resistance genes for antibiotics of last resort), alongside emusynth, a DNA synthesizer emulator. On our test sets, IBBIS commec and SeqScreen incorrectly passed 18.5% and 44.7% of putatively functional pathogenic variants respectively; Capiti passed 0.24%.
Related
- A Cell-based Screening Method Using an Intracellular Antibody for Discovering Small Molecules Targeting Hard-to-drug Proteins
- A Rationally Designed Transgene Drives CAR T Functional Persistence and Durable Regression of Solid Tumors
- Identification of Genes Encoding Seminal Fluid Proteins in Coccinella septempunctata and Functional Analysis of the Gene Encoding Trypsin
- MERS-CoV accessory ORF proteins in innate immune evasion: an evidence-graded functional framework
- Functional diversity of phage sponge proteins that sequester host immune signals
BibTeX
@misc{oualline2026bypassing,
title = {{Bypassing Current Biosecurity Screens with AI Designed Proteins and Closing the Gap with Edge-AI Functional Screening}},
author = {G. Oualline and M. Noon},
howpublished = {bioRxiv},
year = {2026},
month = oct,
doi = {10.64898/2026.10.14.759233},
url = {https://www.biorxiv.org/content/10.64898/2026.10.14.759233v1},
note = {CC BY-NC},
}
RIS
TY - UNPB
TI - Bypassing Current Biosecurity Screens with AI Designed Proteins and Closing the Gap with Edge-AI Functional Screening
AU - G. Oualline
AU - M. Noon
PY - 2026
DA - 2026/10/14
PB - bioRxiv
DO - 10.64898/2026.10.14.759233
UR - https://www.biorxiv.org/content/10.64898/2026.10.14.759233v1
AB - DNA synthesis machines are vital tools in medicine and research, but they can be misused to produce sequences which are transcribed and translated to produce toxic proteins. Synthesis providers may screen orders for dangerous sequences, but most screening software performs sequence-based alignment, which is vulnerable to AI redesign attacks. Additionally, software-only synthesis screening requires the operator of the machine to be trusted. We present Capiti, an edge-deployed screening system that runs with no internet connection, listens directly to a synthesizer's valve control signals, and aborts a run when it detects incipient pathogenic function. Rather than performing sequence alignment, Capiti is specifically trained to recognize fold. ProteinMPNN was used to generate putatively functional variants of known pathogenic proteins, and a CNN was trained to separate them from active-site knockouts, scrambles, and decoys. We release two models, Capiti-E (known pathogenic proteins) and Capiti-C (resistance genes for antibiotics of last resort), alongside emusynth, a DNA synthesizer emulator. On our test sets, IBBIS commec and SeqScreen incorrectly passed 18.5% and 44.7% of putatively functional pathogenic variants respectively; Capiti passed 0.24%.
ER -