eSIG-Net: Accurate prediction of single-mutation induced perturbations on protein interactions using a language model
(edgetic mutation Sequence-based Interaction Grammar Network), a novel sequence-based "Interaction Language Model" for predicting protein interaction alterations caused by single mutations. eSIG-Net combines various protein sequence embeddings, introduces a mutation-encoding module with syntax and evolutionary insights, and employs contrastive learning to evaluate mutation-induced interaction changes. eSIG-Net significantly outperforms current state-of-the-art sequence-based and structure-based prediction methods at predicting mutational impact on protein interactions. We highlight examples where eSIG-Net nominates causal variants with high confidence and elucidates their functional role under relevant biological contexts. Together, eSIG-Net is a first-in-kind "interaction language model" that can accurately predict interaction-specific rewiring by single mutations with only sequence information, and exhibits generalizability across biological contexts.
Authors
- Yuntao Yang (ORCID: https://orcid.org/0000-0003-4829-6077)
- Aditya Shrawat (ORCID: https://orcid.org/0000-0002-6115-552X)
- Daniel J. McGrail (ORCID: https://orcid.org/0000-0002-6669-6069)
- Nidhi Sahni (ORCID: https://orcid.org/0000-0002-9155-5882)
- Daniel F. Jarosz (ORCID: https://orcid.org/0000-0003-3497-5888)
- Jianzhi Huang (ORCID: https://orcid.org/0009-0005-0640-2013)
- Xingxin Pan (ORCID: https://orcid.org/0000-0002-4429-4878)
- Erxi Wu
- Sidi Chen
- Sidharth Raghavan
- Chuanpeng Dong
- Gloria M Sheynkman
- S. Gail Eckhardt
- W. Jim Zheng
- Zhao Li
- Juan I Fuxman Bass
- S. Stephen Yi
Institutions
- Boston University (US)
- Livestrong Foundation (US)
- Cleveland Clinic (US)
- Baylor University (US)
- Baylor College of Medicine (US)
- Yale Cancer Center (US)
- Cleveland Clinic Lerner College of Medicine (US)
- Dana-Farber Cancer Institute (US)
- Systems Biology Institute (JP)
- Houston Institute for Clinical Research (US)
- University of Virginia (US)
- Case Western Reserve University (US)
- The University of Texas at Austin (US)
- Texas A&M University (US)
- The University of Texas Health Science Center at Houston (US)
- Stanford University (US)
Publication Details
- Journal
- bioRxiv (Cold Spring Harbor Laboratory)
- Published
- 2026-03-31
- DOI
- https://doi.org/10.64898/2026.03.27.714913
- Primary Topic
- Bioinformatics and Genomic Networks
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- U.S. Department of Defense
- Cancer Research Institute
- National Institutes of Health
- National Human Genome Research Institute
- National Cancer Institute