A hybrid deep learning framework for toxicity prediction

ToxD4C, a novel multi-modal deep learning framework, was developed for the simultaneous classification and regression of diverse molecular toxicity endpoints. ToxD4C uniquely integrates three-dimensional molecular geometries, graph attention networks, and SE(3)-equivariant Transformer architectures, effectively capturing complex stereochemical and electronic molecular features.

ToxD4C是一款新型多模态深度学习框架，专为多样化分子毒性终点的同步分类与回归任务开发。该框架创新性地融合了三维分子几何结构、图注意力网络（Graph Attention Networks）以及SE(3)等变Transformer（SE(3)-equivariant Transformer）架构，能够有效捕捉复杂的立体化学与电子分子特征。