Logo image
Back
Large Language Models for Molecular Biology (LLMBio): Editorial
Conference proceeding   Open access

Large Language Models for Molecular Biology (LLMBio): Editorial

Gavin Hearne, Glen Rogers, Chaz Allegra, Robi Polikar and Gail Rosen
Companion Proceedings of the 16th ACM International Conference on Bioinformatics, Computational Biology and Health Informatics, pp 1-1
11 Oct 2025
DOI: https://doi.org/10.1145/3768322.3774785
url
https://doi.org/10.1145/3768322.3774785View
Published, Version of Record (VoR) Open

Abstract

Applied computing -- Life and medical sciences -- Computational biology Computing methodologies -- Artificial intelligence -- Natural language processing
The rapid success of large language models (LLMs) in natural language processing has inspired their extension to molecular biology, where biological sequences can similarly be treated as languages. Advances in sequencing and multiomic technologies are producing increasingly vast and complex datasets that strain traditional methods, creating an urgent need for new approaches. LLMs provide a powerful framework for modeling nonlinear dependencies, integrating across modalities, and learning biologically meaningful representations. These LLM capabilities improve applications ranging from protein structure prediction and variant effect prioritization to gene expression modeling and drug discovery. Early successes with such models, capable of understanding and interpreting biological data, underscore the promise of transformer architectures in uncovering biological semantics directly from sequence data. This workshop brings together researchers in computational biology, machine learning, and related disciplines to discuss opportunities and challenges in applying LLMs to biological tasks, such as in genome-scale and multiomic data. Core themes include pretraining strategies, foundation models, benchmarking, interpretability, uncertainty quantification, and ethical considerations in working with sensitive biomedical data. The program features lightning talks, a poster session, and a panel discussion with experts across genomics, bioinformatics, and AI. By fostering cross-disciplinary dialogue and supporting early-career researchers, the workshop seeks to build a community around LLM-driven molecular biology and accelerate the development of interpretable, trustworthy, and impactful models for both basic science and translational applications.

Metrics

2 Record Views

Details

Logo image