发表机构
Eli Lilly and Company（礼来公司）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本研究提出WEECFP分子指纹与带SuRGE编码的Transformer架构，其融合模型在TDC ADMET等基准上表现优异，近乎无损的令牌化与高效图距离编码提升了分子属性预测性能。
AI 中文摘要
我们提出了WEECFP，一种无参数的1024维连续分子指纹，它将每个Morgan亚结构分散到单个向量的约32个带符号位置中；以及WEECFP-SuRGE，一种Transformer架构，其自注意力机制应用了SuRGE（亚结构旋转图距离编码）——一种由分子最短路径图距离参数化的类RoPE旋转——作用于WEECFP亚结构令牌。该架构的7模型融合版本（WEECFP-SuRGE融合版）在TDC ADMET排行榜上实现了最低的平均回归排名；在TDC ADMET排行榜上总体排名第2（仅落后于预训练的MapLight+GNN），且在不使用外部预训练的方法中排名第1；在整个22基准套件中，在Pgp、亲脂性、CYP2D6底物、微粒体清除率和LD50上获得了排行榜第1名（WEECFP-NoSuRGE融合版单独在HIA上达到第1名）——全程未使用任何外部预训练。在MoleculeNet上，WEECFP-SuRGE在4项回归任务中的3项（ESOL、亲脂性、QM9）上击败了所有经典指纹基线。我们进一步表明，WEECFP令牌化近乎无损：贪婪重叠重构在9个MoleculeNet数据集的分布内分子中恢复了99.9%的精确规范SMILES，在跨数据集保留集（HIV→亲脂性）中恢复了98.93%的分子；且图距离的三参考最远优先编码与真实成对距离的皮尔逊相关系数r=0.901，在匹配精度下实现了O(S)的位置内存。
英文摘要
Computational molecular property prediction requires representations that capture local chemistry, long-range interactions, and molecular topology. Conventional fingerprints provide efficient local substructure features, whereas learned graph and sequence models can represent broader context but often rely on pretraining or three-dimensional conformers. We introduce Wide Encoded Extended Connectivity Fingerprints (WEECFP) with Substructure Rotary Graph-distance Encoding (SuRGE), a tokenized hierarchical Morgan representation in which graph-distance-dependent rotations are applied at the input and within transformer self-attention. Across MoleculeNet and the Therapeutic Data Commons ADMET benchmarks, WEECFP-SuRGE is competitive with recent pretrained and geometry-aware methods without external pretraining or conformer generation. We also show that the rotated token representation remains structurally informative. A guided confirmed-handshake overlap procedure reconstructs the correct constitutional isomer for 92.6% of a 4,200-molecule self-library evaluation. Together, the predictive and reconstruction results indicate that WEECFP-SuRGE preserves local substructure identity while making relative topology available to the model.
