BSL: A Unified and Generalizable Multitask Learning Platform for Virtual Drug Discovery from Design to Synthesis

2025-08-02

AI Summary

Current virtual drug discovery platforms suffer from two major limitations: (1) incomplete task coverage and fragmented workflows, and (2) weak generalization—particularly for out-of-distribution (OOD) molecules. To address these challenges, we introduce BSL, an open-source deep learning platform that unifies seven core tasks—including molecular generation, property prediction, and activity assessment—within a single end-to-end multi-task learning framework. BSL innovatively integrates generative models with graph neural networks, adopts a modular and extensible architecture, and incorporates OOD-aware representation learning and evaluation mechanisms to significantly enhance cross-molecular-space generalization. Empirically, BSL achieves state-of-the-art performance across multiple benchmarks. Furthermore, it successfully identified three novel, experimentally validated active compounds targeting the GluN1/GluN3A subunits of the NMDA receptor—demonstrating both algorithmic advancement and practical utility in de novo drug discovery.

Technology Category

• Machine Learning: Deep Neural Architectures and Foundation Models
• Natural Language Processing: Generation
• Search and Optimization: Learning to Search

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Drug discovery is of great social significance in safeguarding human health, prolonging life, and addressing the challenges of major diseases. In recent years, artificial intelligence has demonstrated remarkable advantages in key tasks across bioinformatics and pharmacology, owing to its efficient data processing and data representation capabilities. However, most existing computational platforms cover only a subset of core tasks, leading to fragmented workflows and low efficiency. In addition, they often lack algorithmic innovation and show poor generalization to out-of-distribution (OOD) data, which greatly hinders the progress of drug discovery. To address these limitations, we propose Baishenglai (BSL), a deep learning-enhanced, open-access platform designed for virtual drug discovery. BSL integrates seven core tasks within a unified and modular framework, incorporating advanced technologies such as generative models and graph neural networks. In addition to achieving state-of-the-art (SOTA) performance on multiple benchmark datasets, the platform emphasizes evaluation mechanisms that focus on generalization to OOD molecular structures. Comparative experiments with existing platforms and baseline methods demonstrate that BSL provides a comprehensive, scalable, and effective solution for virtual drug discovery, offering both algorithmic innovation and high-precision prediction for real-world pharmaceutical research. In addition, BSL demonstrated its practical utility by discovering novel modulators of the GluN1/GluN3A NMDA receptor, successfully identifying three compounds with clear bioactivity in in-vitro electrophysiological assays. These results highlight BSL as a promising and comprehensive platform for accelerating biomedical research and drug discovery. The platform is accessible at https://www.baishenglai.net.

Problem

• Unified platform for multitask drug discovery workflows
• Improving generalization to out-of-distribution molecular data
• Integrating seven core tasks with advanced AI technologies

Innovation

• Unified multitask learning platform for drug discovery
• Generative models and graph neural networks integration
• Focus on generalization to out-of-distribution molecular structures