This study introduces an innovative deep reinforcement learning framework for drug discovery, seamlessly integrating Quantitative Structure-Activity Relationship (QSAR) predictions with generative modeling. Focused on developing spleen tyrosine kinase inhibitors for immune thrombocytopenia, the approach leverages a stacking-ensemble QSAR model and deep learning to design promising therapeutic compounds.

Завадская М.С., Орлова А.А. (науч. рук. Виноградов В.В.) Qsar-enhanced reinforcement learning approach for novel syk inhibitor discovery // Сборник тезисов докладов конгресса молодых ученых. Электронное издание. – СПб: Университет ИТМО, [2025]. URL: https://kmu.itmo.ru/digests/article/14132