Skip to content
DraconiticResearch
Search Sources API
194,067 studies31 sourcesUpdated 12 min ago
PreprintCC BYChemRxiv

AI Agents for CRISPR Diagnostics: From Automation to Bounded Autonomy

  • Sina Jamalzadegan1,2,3,4,5,6,7
  • Mahsa Bagi6,7
  • Sepehr Soltani8,9
  • Zahra Joudaki6,7
  • Akhil Penumudy6,7
  • David Haussler1,2,3,4
  • Mohammed A. Mostajo-Radji1,2,3
  • Mircea Teodorescu1,2,3,4,5
  • Qingshan Wei6,7
9 affiliations
  1. 1Genomics Institute
  2. 2University of California Santa Cruz
  3. 3Live Cell Biotechnology Discovery Lab
  4. 4Department of Biomolecular Engineering
  5. 5Department of Electrical and Computer Engineering
  6. 6Department of Chemical and Biomolecular Engineering
  7. 7North Carolina State University
  8. 8Department of Industrial Engineering
  9. 9Cullen College of Engineering, University of Houston
Open at sourceDOICite

Abstract

Recent years have witnessed rapid advances in clustered regularly interspaced short palindromic repeats (CRISPR) diagnostics, artificial intelligence (AI), and self-driving laboratories for automation. CRISPR diagnostics have emerged as a promising strategy for earlier disease detection, with Cas12a and Cas13a targeting DNA and RNA, respectively, and converting that recognition into collateral reporter cleavage. In parallel, different subsets of AI have advanced rapidly, with machine learning (ML) interpreting biosensor readouts, and large language models (LLMs) and AI agents applied to automation. Laboratory automation and self-driving platforms reduce the workload of human experimentalists and enable scale-up, having run long unattended campaigns in materials discovery. However, how these advances in AI, CRISPR diagnostics and automation could be integrated to enable a future of autonomous biosensing remains unresolved. In this perspective, we evaluate CRISPR diagnostics as autonomous systems, where analytical metrics alone cannot ensure robustness, failure-mode predictability, recoverability and auditability. We then propose a bounded-autonomy layer in which multimodal evidence, scored by ML, gates a Sense, quality control (QC), Decide, Act and Report loop, with guardrailed LLM agents invoked only when QC flags a run, selecting pre-validated actions under deterministic validation. By tying detectable failure modes to validated recovery actions, we outline a path toward bounded autonomy in CRISPR diagnostics, where a device unable to act safely abstains and escalates.

Related

  1. Engineering gRNA for CRISPR biosensing: chemical, structural, and programmable designs for ultrasensitive diagnostics
    Wenyu Fu, Rui Sang, Ruier Xue +4
    TrAC Trends in Analytical ChemistryPeer-reviewedCC BY6 citedMolecular biology & gene regulationOpenAlex
  2. Automation of the Azapeptide Submonomer Synthesis Method
    Tyler G. Hobart, Molly E. Helton, Maxwell O. Bowles +1
    Journal of peptide science : an official publication of the European Peptide SocietyPeer-reviewedCC BY-NC-NDFull textMolecular biology & gene regulationEurope PMC
  3. A rapid, automation-friendly, cell-free and purification-free pipeline for binding and kinetic characterisation of de novo designed binders with high quality standardised per-design data
    D. Ruiz Espi, G. Avelar Molina, O. Sonne Buch +9
    bioRxivPreprintCC BYFull textMolecular biology & gene regulationbioRxiv
  4. Ion Channel Nano-Diagnostics for ER+ Breast Cancer
    Manos Gkikas, Evangelos Dadiotis, Mehreen Zaka +3
    Journal of Biological ChemistryPeer-reviewedCC BY 4.0Molecular biology & gene regulationCrossref
  5. Development of CRISPR/Cas13-based analytical tools to study RNA–protein interactions
    Nucleic acids researchPeer-reviewedCC BYFull textMolecular biology & gene regulationEurope PMC

Cite

BibTeXRIS
BibTeX
@misc{jamalzadegan2026agents,
  title = {{AI Agents for CRISPR Diagnostics: From Automation to Bounded Autonomy}},
  author = {Sina Jamalzadegan and Mahsa Bagi and Sepehr Soltani and Zahra Joudaki and Akhil Penumudy and David Haussler and Mohammed A. Mostajo-Radji and Mircea Teodorescu and Qingshan Wei},
  howpublished = {ChemRxiv},
  year = {2026},
  month = oct,
  doi = {10.26434/chemrxiv.15010042/v1},
  url = {https://chemrxiv.org/doi/full/10.26434/chemrxiv.15010042/v1},
  note = {CC BY},
}
RIS
TY  - UNPB
TI  - AI Agents for CRISPR Diagnostics: From Automation to Bounded Autonomy
AU  - Sina Jamalzadegan
AU  - Mahsa Bagi
AU  - Sepehr Soltani
AU  - Zahra Joudaki
AU  - Akhil Penumudy
AU  - David Haussler
AU  - Mohammed A. Mostajo-Radji
AU  - Mircea Teodorescu
AU  - Qingshan Wei
PY  - 2026
DA  - 2026/10/07
PB  - ChemRxiv
DO  - 10.26434/chemrxiv.15010042/v1
UR  - https://chemrxiv.org/doi/full/10.26434/chemrxiv.15010042/v1
AB  - Recent years have witnessed rapid advances in clustered regularly interspaced short palindromic repeats (CRISPR) diagnostics, artificial intelligence (AI), and self-driving laboratories for automation. CRISPR diagnostics have emerged as a promising strategy for earlier disease detection, with Cas12a and Cas13a targeting DNA and RNA, respectively, and converting that recognition into collateral reporter cleavage. In parallel, different subsets of AI have advanced rapidly, with machine learning (ML) interpreting biosensor readouts, and large language models (LLMs) and AI agents applied to automation. Laboratory automation and self-driving platforms reduce the workload of human experimentalists and enable scale-up, having run long unattended campaigns in materials discovery. However, how these advances in AI, CRISPR diagnostics and automation could be integrated to enable a future of autonomous biosensing remains unresolved. In this perspective, we evaluate CRISPR diagnostics as autonomous systems, where analytical metrics alone cannot ensure robustness, failure-mode predictability, recoverability and auditability. We then propose a bounded-autonomy layer in which multimodal evidence, scored by ML, gates a Sense, quality control (QC), Decide, Act and Report loop, with guardrailed LLM agents invoked only when QC flags a run, selecting pre-validated actions under deterministic validation. By tying detectable failure modes to validated recovery actions, we outline a path toward bounded autonomy in CRISPR diagnostics, where a device unable to act safely abstains and escalates.
ER  -