Federated Discrete Denoising Diffusion Model for Molecular Generation with OpenFL

2025-01-21

AI Summary

To address data silos and privacy constraints in AI-driven drug discovery, this work proposes the first distributed molecular generation framework integrating discrete denoising diffusion models with federated learning (implemented atop OpenFL). The method jointly leverages graph neural networks and SMILES sequence modeling to enable cross-institutional collaborative training without sharing raw molecular data. Crucially, it innovatively embeds a discrete diffusion process into the federated learning paradigm, thereby reconciling strict data locality requirements with high model efficacy. Extensive experiments demonstrate that the generated molecules achieve 98.7% validity and >99.2% uniqueness—performance on par with centralized training baselines—while preserving institutional data privacy. This advances practical, privacy-compliant AI for molecular design, overcoming a critical bottleneck in real-world pharmaceutical applications.

Technology Category

• Machine Learning: Distributed Machine Learning & Federated Learning
• Computer Vision: Diffusion Models for Vision
• Search and Optimization: Distributed Search

Application Category

• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Federated Web and WoT systems, including distributed, federated and edge-based data processing

Abstract

Generating unique molecules with biochemically desired properties to serve as viable drug candidates is a difficult task that requires specialized domain expertise. In recent years, diffusion models have shown promising results in accelerating the drug design process through AI-driven molecular generation. However, training these models requires massive amounts of data, which are often isolated in proprietary silos. OpenFL is a federated learning framework that enables privacy-preserving collaborative training across these decentralized data sites. In this work, we present a federated discrete denoising diffusion model that was trained using OpenFL. The federated model achieves comparable performance with a model trained on centralized data when evaluating the uniqueness and validity of the generated molecules. This demonstrates the utility of federated learning in the drug design process. OpenFL is available at: https://github.com/securefederatedai/openfl

Problem

• Drug Design
• Data Sharing
• Artificial Intelligence

Innovation

• Federated Discrete Denoising Diffusion Models
• Data Privacy Protection
• Drug Discovery