做分子性质预测的朋友看过来：新模型EnsembleEGNN把多个构象编码成一个嵌入，在环肽任务上比纯序列模型效果好，R²提升到0.538。
EnsembleEGNN 是一种分子集成基础模型，通过对每个构象使用共享的等变图神经网络 (EGNN) 编码，再用 Set Attention Block 池化得到整体表示。在 CREMP 环肽数据集上预训练，采用掩码 token 恢复、噪声坐标重建和成对距离重建的多任务自监督目标。在 CREMP-CycPeptMPDB 数据集上，从零训练的 EnsembleEGNN 性能极差 (R²=0.005)，而预训练模型达到 R²=0.477、Pearson r=0.699，超过仅序列的 BERT 基线 (R²=0.439, Pearson r=0.667)。与 BERT 序列编码器联合端到端训练后，混合模型进一步提升至 R²=0.538、Pearson r=0.737。
Graph Learning on Ensembles of Cyclic Peptides: An Investigation of Molecular Ensemble Modeling
Molecular property prediction from structure often uses a single representative conformation, even though many molecules exist as conformational ensembles in solution. We introduce EnsembleEGNN, a molecular ensemble foundation model that encodes an ensemble by first encoding each conformer with shared Equivariant Graph Neural Network (EGNN) layers, then pooling the resulting conformer representations with a Set Attention Block. We pretrain the model on CREMP, a cyclic peptide ensemble dataset, using a multi-task self-supervised objective combining masked token recovery, noisy-coordinate reconstruction, and pairwise distance reconstruction. On the CREMP-CycPeptMPDB dataset, training EnsembleEGNN from scratch fails entirely ($R^2=0.005$). However, the pretrained model reaches $R^2=0.477$ and Pearson $r=0.699$, outperforming the sequence-only BERT baseline ($R^2=0.439$, Pearson $r=0.667$). When EnsembleEGNN is co-trained end-to-end with the BERT sequence encoder, the hybrid model improves further to $R^2=0.538$ and Pearson $r=0.737$. These results demonstrate that encoding conformational ensembles into a single thermodynamically informed embedding improves cyclic-peptide property prediction.