Integrating QSAR modelling with reinforcement learning for Syk inhibitor discovery

Abstract Spleen tyrosine kinase (Syk) is a crucial mediator of inflammatory processes and a promising therapeutic target for the management of autoimmune disorders, such as immune thrombocytopenia. While several Syk inhibitors are known to date, their efficacy and safety profiles remain suboptimal,...

Full description

Saved in:
Bibliographic Details
Main Authors: Maria Zavadskaya, Anastasia Orlova, Andrei Dmitrenko, Vladimir Vinogradov
Format: Article
Language:English
Published: BMC 2025-04-01
Series:Journal of Cheminformatics
Subjects:
Online Access:https://doi.org/10.1186/s13321-025-00998-2
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850181791867994112
author Maria Zavadskaya
Anastasia Orlova
Andrei Dmitrenko
Vladimir Vinogradov
author_facet Maria Zavadskaya
Anastasia Orlova
Andrei Dmitrenko
Vladimir Vinogradov
author_sort Maria Zavadskaya
collection DOAJ
description Abstract Spleen tyrosine kinase (Syk) is a crucial mediator of inflammatory processes and a promising therapeutic target for the management of autoimmune disorders, such as immune thrombocytopenia. While several Syk inhibitors are known to date, their efficacy and safety profiles remain suboptimal, necessitating the exploration of novel compounds. The study introduces a novel deep reinforcement learning strategy for drug discovery, specifically designed to identify new Syk inhibitors. The approach integrates quantitative structure–activity relationship (QSAR) predictions with generative modelling, employing a stacking-ensemble model that achieves a correlation coefficient of 0.78. From over 78,000 molecules generated by this methodology, we identified 139 promising candidates with high predicted potency, binding affinity and optimal drug-likeness properties, demonstrating structural novelty while maintaining essential Syk inhibitor characteristics. Our approach establishes a versatile framework for accelerated drug discovery, which is particularly valuable for the development of rare disease therapeutics. Scientific contribution The study presents the first application of QSAR-guided reinforcement learning for Syk inhibitor discovery, yielding structurally novel candidates with predicted high potency. The presented methodology can be adapted for other therapeutic targets, potentially accelerating the drug development process.
format Article
id doaj-art-3f80a30a440b4fe2a9ec7c0049650a80
institution OA Journals
issn 1758-2946
language English
publishDate 2025-04-01
publisher BMC
record_format Article
series Journal of Cheminformatics
spelling doaj-art-3f80a30a440b4fe2a9ec7c0049650a802025-08-20T02:17:49ZengBMCJournal of Cheminformatics1758-29462025-04-0117111110.1186/s13321-025-00998-2Integrating QSAR modelling with reinforcement learning for Syk inhibitor discoveryMaria Zavadskaya0Anastasia Orlova1Andrei Dmitrenko2Vladimir Vinogradov3 Center for AI in Chemistry, ITMO University Center for AI in Chemistry, ITMO University Center for AI in Chemistry, ITMO University Center for AI in Chemistry, ITMO UniversityAbstract Spleen tyrosine kinase (Syk) is a crucial mediator of inflammatory processes and a promising therapeutic target for the management of autoimmune disorders, such as immune thrombocytopenia. While several Syk inhibitors are known to date, their efficacy and safety profiles remain suboptimal, necessitating the exploration of novel compounds. The study introduces a novel deep reinforcement learning strategy for drug discovery, specifically designed to identify new Syk inhibitors. The approach integrates quantitative structure–activity relationship (QSAR) predictions with generative modelling, employing a stacking-ensemble model that achieves a correlation coefficient of 0.78. From over 78,000 molecules generated by this methodology, we identified 139 promising candidates with high predicted potency, binding affinity and optimal drug-likeness properties, demonstrating structural novelty while maintaining essential Syk inhibitor characteristics. Our approach establishes a versatile framework for accelerated drug discovery, which is particularly valuable for the development of rare disease therapeutics. Scientific contribution The study presents the first application of QSAR-guided reinforcement learning for Syk inhibitor discovery, yielding structurally novel candidates with predicted high potency. The presented methodology can be adapted for other therapeutic targets, potentially accelerating the drug development process.https://doi.org/10.1186/s13321-025-00998-2Generative designDrug discoveryQSARMachine learningSyk inhibitorsReinforcement learning
spellingShingle Maria Zavadskaya
Anastasia Orlova
Andrei Dmitrenko
Vladimir Vinogradov
Integrating QSAR modelling with reinforcement learning for Syk inhibitor discovery
Journal of Cheminformatics
Generative design
Drug discovery
QSAR
Machine learning
Syk inhibitors
Reinforcement learning
title Integrating QSAR modelling with reinforcement learning for Syk inhibitor discovery
title_full Integrating QSAR modelling with reinforcement learning for Syk inhibitor discovery
title_fullStr Integrating QSAR modelling with reinforcement learning for Syk inhibitor discovery
title_full_unstemmed Integrating QSAR modelling with reinforcement learning for Syk inhibitor discovery
title_short Integrating QSAR modelling with reinforcement learning for Syk inhibitor discovery
title_sort integrating qsar modelling with reinforcement learning for syk inhibitor discovery
topic Generative design
Drug discovery
QSAR
Machine learning
Syk inhibitors
Reinforcement learning
url https://doi.org/10.1186/s13321-025-00998-2
work_keys_str_mv AT mariazavadskaya integratingqsarmodellingwithreinforcementlearningforsykinhibitordiscovery
AT anastasiaorlova integratingqsarmodellingwithreinforcementlearningforsykinhibitordiscovery
AT andreidmitrenko integratingqsarmodellingwithreinforcementlearningforsykinhibitordiscovery
AT vladimirvinogradov integratingqsarmodellingwithreinforcementlearningforsykinhibitordiscovery