Machine Learning for Genomics Explorations (MLGenX)

Call for Papers

MLGenX 2026 invites submissions from researchers at the intersection of machine learning, genomics, and biological discovery. This year, the workshop will feature four tracks designed to welcome a diverse range of contributions: the Main Track, the Special Track on Lab-in-the-Loop and Self-Evolving Systems, the Tiny Papers Track, and a new AI-Generated Track. By providing dedicated spaces for applied ML in biology, methodological innovation, early-stage ideas, and AI serving as a primary author, MLGenX 2026 aims to highlight the next frontier of target discovery—closing the loop between reasoning, experimentation, and adaptation.

Main Track

The Main Track welcomes contributions spanning both machine learning methodology and biologically grounded applications at the intersection of ML, biology, and genomics. Representative topics include, but are not limited to:

• Foundation models and agentic AI: scalable training and fine-tuning, in-context learning for biological and multi-omics data, and agentic systems for reasoning and tool use.
• Causality and mechanistic interpretability: causal discovery, counterfactual modeling, and biologically grounded representations to reveal mechanism rather than correlation.
• Generalizability and uncertainty quantification: robust transfer across biological domains, modalities, and experimental settings.
• Design of regulatory sequence elements: ML for DNA, RNA, and gene or cell therapeutics, including sequence-to-function modeling and AI-assisted CRISPR and RNA design.
• Perturbative biology and cellular organization: modeling cellular responses to perturbations and inferring cell states, cell–cell interactions, and tissue organization from multi-omics data.

Special Track on Lab-in-the-Loop and Self-Evolving Systems

This track focuses on systems that learn through interaction with experimental feedback. Representative topics include, but are not limited to:

• Active learning with experimental feedback: models that adapt based on wet-lab or simulation outcomes, optimizing experimental design and discovery cycles.
• Lab-in-the-loop architectures: integration of AI reasoning systems with robotic labs, automated pipelines, and real-time data streams.
• Self-evolving agentic frameworks: autonomous agents capable of iterative hypothesis generation, testing, and refinement using foundation models and multi-agent coordination.

Tiny Papers Track

The Tiny Papers Track provides a venue for short (up to 4-page) submissions presenting early-stage ideas, negative results, position papers, or preliminary findings. This track aims to lower the barrier for participation and highlight creative or exploratory directions. AI-generated papers are not permitted in this track.

AI-Generated Track

MLGenX 2026 introduces a new AI-Generated Track for submissions where AI systems serve as a substantial contributor or primary author, under clear human oversight. Papers in this track should align with the Main Track or Special Track themes and follow standard ICLR formatting guidelines. Submissions must clearly disclose the role of AI systems and adhere to ICLR policies on large language model usage and research ethics.

Journal Partnership

Partnership with Nature Biotechnology. Top workshop contributions will be eligible for a fast-track review process at Nature Biotechnology. Participation is optional and offered after acceptance notifications. Selected authors will be invited to expand evaluations and address reviewer comments prior to transfer to the journal.

Submission Instructions

Similar to the main ICLR conference, submissions will be double blind. We use OpenReview to host papers. There is no strict page limit for submissions to the Main Track or the Special Track. Submissions to the AI-Generated Track must not exceed 8 pages for the main text, and submissions to the Tiny Papers Track must not exceed 4 pages for the main text. References and appendices may be of unlimited length. To prepare your submission, please use the official ICLR template.

Submissions that are identical to versions that have been previously published, or accepted to the main ICLR conference are not allowed. However, papers that cite previous related work by the authors and papers that have appeared on non-peer reviewed websites (like arXiv) do not violate the policy. Submission of the paper to archival repositories such as arXiv is allowed during the review period.

Note: Authors are permitted to submit works that are currently under review by other venues. Additionally, accepted workshop papers are non-archival and may be subsequently published in conferences or journals.

We plan to offer Best Paper Award(s), and exceptional submissions may be selected for oral presentations. Accepted papers will be featured on the workshop website.

Additional Track-Specific Notes: Submissions to the AI-Generated Track must clearly disclose the role of AI systems and human contributors and must comply with ICLR policies on large language model usage and research ethics. AI-generated submissions are not permitted in the Tiny Papers Track.

Note: Official reviews are anonymous. Unlike the main conference, submissions and reviews will not be made public until acceptance.

Important Dates

All deadlines are 11:59 pm UTC -12h ("Anywhere on Earth"). All authors must have an OpenReview profile when submitting.

• Main & Special Track Submission Deadline: February 2, 2026
• Tiny Papers Track Submission Deadline: February 8, 2026
• Acceptance Notification (all tracks): March 1, 2026

Tentative Speakers & Panelists

Charlotte Bunne
EPFL

Bo Wang
University of Toronto

Smita Krishnaswamy
Yale University

Kathy Wei
310.ai

Hani Goodarzi
UCSF

James Zou
Stanford University

Kyunghyun Cho
New York University, Genentech

Catherine Tong
Isomorphic Labs

Organizers

Ehsan Hajiramezanali

Aviv Regev

Fabian Theis

Mihaela van der Schaar

Arman Hasanzadeh

Wei Qiu

Tommaso Biancalani

Acknowledgement

We would also like to appreciate Gabriele Scalia and Aicha BenTaieb for their insightful input, ongoing collaboration, and continued engagement. We also extend our sincere thanks to Barbara Cheifet, Chief Editor of Nature Biotechnology, for her generous help and guidance in establishing the partnership with the journal.