Prediction of Residue-Level Base Preference for DNA-Binding Proteins Using a Multimodal Learning Framework
4 affiliations
- 1Shanghai Jiao Tong University , , ,
- 2Xinhua Hospital Affiliated to Shanghai Jiao Tong University School of Medicine , , ,
- 3Songjiang Hospital Affiliated to Shanghai Jiao Tong University School of Medicine , , ,
- 4Key Laboratory of Gene Editing and Cell-Based Immunotherapy for Hematological Diseases , ,
Abstract
Specific recognition between proteins and DNA governs essential cellular functions, yet the binding specificity of most DNA-binding proteins (DBPs) remains experimentally uncharacterized and high-throughput experimental methods still lag far behind the ever-growing protein data set. Here, we present 3M-BSIP (3-modal base-specific interaction predictor), a multimodal deep learning framework capable of predicting the DNA-binding interface and residue-level base preference for DBPs. 3M-BSIP constructs a holistic protein representation by integrating three distinct modalities: a protein language model capturing evolutionary conservation; a graph neural network encoding 3D structural features; and a hierarchical encoder characterizing physicochemical properties of protein surface pointclouds. Comprehensive benchmarking demonstrates that 3M-BSIP achieves SOTA performance on both interface-residue identification and base-preference predictions. Crucially, the model maintains robust performance on computationally constructed protein structures and is also tolerant toward those undergoing profound structural rearrangements upon DNA binding. Two real-world applications of 3M-BSIP demonstrate that our model not only can accurately predict the specific binding motif of structurally unknown transcription factors but also, more importantly, can guide rapid reprogramming of the base preference of DBPs based on a purely computational pipeline. Therefore, 3M-BSIP holds great potential in annotating the binding motifs of all DBPs in a high-throughput manner and designing new DBPs with desired base specificities.
Related
- VAEBAC: a representation-learning framework for proteome-scale prediction of amyloidogenic proteins and functional stratification of nucleic-acid-binding proteins.
- GeoEPred: A multimodal structure-aware geometric deep learning framework for Gram-negative bacterial secreted effector prediction with sequence semantics
- A learning-to-rank framework for large-scale identification of unknown proteins from cryo-EM and cryo-ET maps
- When Clinical and Metabolomics Data Work Together: A Comparative Framework for Multimodal Disease Classification across Machine and Deep Learning.
- AbLWR: a context-aware listwise ranking framework for antibody-antigen binding affinity prediction via positive-unlabeled learning
BibTeX
@article{li2026prediction,
title = {{Prediction of Residue-Level Base Preference for DNA-Binding Proteins Using a Multimodal Learning Framework}},
author = {Siyuan Li and Wenyan He and Mengru Zhang and Xin Zhang and Wei-Qun Fang and Min-Juan Xu and Yi Shi and Yujia Cai and Yong Zhang and Lin-Tai Da},
journal = {JACS Au},
year = {2026},
month = oct,
doi = {10.1021/jacsau.6c00925},
url = {https://pubs.acs.org/jaaucr/article/6/10/5663/5443142/Prediction-of-Residue-Level-Base-Preference-for},
note = {CC BY-NC-ND 4.0},
}
RIS
TY - JOUR
TI - Prediction of Residue-Level Base Preference for DNA-Binding Proteins Using a Multimodal Learning Framework
AU - Siyuan Li
AU - Wenyan He
AU - Mengru Zhang
AU - Xin Zhang
AU - Wei-Qun Fang
AU - Min-Juan Xu
AU - Yi Shi
AU - Yujia Cai
AU - Yong Zhang
AU - Lin-Tai Da
PY - 2026
DA - 2026/10/01
JO - JACS Au
DO - 10.1021/jacsau.6c00925
UR - https://pubs.acs.org/jaaucr/article/6/10/5663/5443142/Prediction-of-Residue-Level-Base-Preference-for
AB - Specific recognition between proteins and DNA governs essential cellular functions, yet the binding specificity of most DNA-binding proteins (DBPs) remains experimentally uncharacterized and high-throughput experimental methods still lag far behind the ever-growing protein data set. Here, we present 3M-BSIP (3-modal base-specific interaction predictor), a multimodal deep learning framework capable of predicting the DNA-binding interface and residue-level base preference for DBPs. 3M-BSIP constructs a holistic protein representation by integrating three distinct modalities: a protein language model capturing evolutionary conservation; a graph neural network encoding 3D structural features; and a hierarchical encoder characterizing physicochemical properties of protein surface pointclouds. Comprehensive benchmarking demonstrates that 3M-BSIP achieves SOTA performance on both interface-residue identification and base-preference predictions. Crucially, the model maintains robust performance on computationally constructed protein structures and is also tolerant toward those undergoing profound structural rearrangements upon DNA binding. Two real-world applications of 3M-BSIP demonstrate that our model not only can accurately predict the specific binding motif of structurally unknown transcription factors but also, more importantly, can guide rapid reprogramming of the base preference of DBPs based on a purely computational pipeline. Therefore, 3M-BSIP holds great potential in annotating the binding motifs of all DBPs in a high-throughput manner and designing new DBPs with desired base specificities.
SN - 2691-3704
ER -