Sequence-based virtual screening using transformers

Abstract Protein-ligand interactions play central roles in myriad biological processes and are of key importance in drug design. Deep learning approaches are becoming cost-effective alternatives to high-throughput experimental methods for ligand identification. Here, to predict the binding affinity...

Full description

Saved in:
Bibliographic Details
Main Authors: Shengyu Zhang, Donghui Huo, Robert I. Horne, Yumeng Qi, Sebastian Pujalte Ojeda, Aixia Yan, Michele Vendruscolo
Format: Article
Language:English
Published: Nature Portfolio 2025-07-01
Series:Nature Communications
Online Access:https://doi.org/10.1038/s41467-025-61833-8
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849235186468782080
author Shengyu Zhang
Donghui Huo
Robert I. Horne
Yumeng Qi
Sebastian Pujalte Ojeda
Aixia Yan
Michele Vendruscolo
author_facet Shengyu Zhang
Donghui Huo
Robert I. Horne
Yumeng Qi
Sebastian Pujalte Ojeda
Aixia Yan
Michele Vendruscolo
author_sort Shengyu Zhang
collection DOAJ
description Abstract Protein-ligand interactions play central roles in myriad biological processes and are of key importance in drug design. Deep learning approaches are becoming cost-effective alternatives to high-throughput experimental methods for ligand identification. Here, to predict the binding affinity between proteins and small molecules, we introduce Ligand-Transformer, a deep learning method based on the transformer architecture. Ligand-Transformer implements a sequence-based approach, where the inputs are the amino acid sequence of the target protein and the topology of the small molecule to enable the prediction of the conformational space explored by the complex between the two. We apply Ligand-Transformer to screen and validate experimentally inhibitors targeting the mutant EGFRLTC kinase, identifying compounds with low nanomolar potency. We then use this approach to predict the conformational population shifts induced by known ABL kinase inhibitors, showing that sequence-based predictions enable the characterisation of the population shift upon binding. Overall, our results illustrate the potential of Ligand-Transformer to accurately predict the interactions of small molecules with proteins, including the binding affinity and the changes in the free energy landscapes upon binding, thus uncovering molecular mechanisms and facilitating the initial steps in drug design.
format Article
id doaj-art-eb02de2104b14254af2d596b5830c635
institution Kabale University
issn 2041-1723
language English
publishDate 2025-07-01
publisher Nature Portfolio
record_format Article
series Nature Communications
spelling doaj-art-eb02de2104b14254af2d596b5830c6352025-08-20T04:02:54ZengNature PortfolioNature Communications2041-17232025-07-0116111210.1038/s41467-025-61833-8Sequence-based virtual screening using transformersShengyu Zhang0Donghui Huo1Robert I. Horne2Yumeng Qi3Sebastian Pujalte Ojeda4Aixia Yan5Michele Vendruscolo6Centre for Misfolding Diseases, Yusuf Hamied Department of Chemistry, University of CambridgeCentre for Misfolding Diseases, Yusuf Hamied Department of Chemistry, University of CambridgeCentre for Misfolding Diseases, Yusuf Hamied Department of Chemistry, University of CambridgeCentre for Misfolding Diseases, Yusuf Hamied Department of Chemistry, University of CambridgeCentre for Misfolding Diseases, Yusuf Hamied Department of Chemistry, University of CambridgeCollege of Life Science and Technology, Beijing University of Chemical TechnologyCentre for Misfolding Diseases, Yusuf Hamied Department of Chemistry, University of CambridgeAbstract Protein-ligand interactions play central roles in myriad biological processes and are of key importance in drug design. Deep learning approaches are becoming cost-effective alternatives to high-throughput experimental methods for ligand identification. Here, to predict the binding affinity between proteins and small molecules, we introduce Ligand-Transformer, a deep learning method based on the transformer architecture. Ligand-Transformer implements a sequence-based approach, where the inputs are the amino acid sequence of the target protein and the topology of the small molecule to enable the prediction of the conformational space explored by the complex between the two. We apply Ligand-Transformer to screen and validate experimentally inhibitors targeting the mutant EGFRLTC kinase, identifying compounds with low nanomolar potency. We then use this approach to predict the conformational population shifts induced by known ABL kinase inhibitors, showing that sequence-based predictions enable the characterisation of the population shift upon binding. Overall, our results illustrate the potential of Ligand-Transformer to accurately predict the interactions of small molecules with proteins, including the binding affinity and the changes in the free energy landscapes upon binding, thus uncovering molecular mechanisms and facilitating the initial steps in drug design.https://doi.org/10.1038/s41467-025-61833-8
spellingShingle Shengyu Zhang
Donghui Huo
Robert I. Horne
Yumeng Qi
Sebastian Pujalte Ojeda
Aixia Yan
Michele Vendruscolo
Sequence-based virtual screening using transformers
Nature Communications
title Sequence-based virtual screening using transformers
title_full Sequence-based virtual screening using transformers
title_fullStr Sequence-based virtual screening using transformers
title_full_unstemmed Sequence-based virtual screening using transformers
title_short Sequence-based virtual screening using transformers
title_sort sequence based virtual screening using transformers
url https://doi.org/10.1038/s41467-025-61833-8
work_keys_str_mv AT shengyuzhang sequencebasedvirtualscreeningusingtransformers
AT donghuihuo sequencebasedvirtualscreeningusingtransformers
AT robertihorne sequencebasedvirtualscreeningusingtransformers
AT yumengqi sequencebasedvirtualscreeningusingtransformers
AT sebastianpujalteojeda sequencebasedvirtualscreeningusingtransformers
AT aixiayan sequencebasedvirtualscreeningusingtransformers
AT michelevendruscolo sequencebasedvirtualscreeningusingtransformers