Prediction of inter-chain distance maps of protein complexes with 2D attention-based deep neural networks

Residue-residue distance information is useful for predicting tertiary structures of protein monomers or quaternary structures of protein complexes. Many deep learning methods have been developed to predict intra-chain residue-residue distances of monomers accurately, but few methods can accurately predict inter-chain residue-residue distances of complexes. We develop a deep learning method CDPred (i.e., Complex Distance Prediction) based on the 2D attention-powered residual network to address the gap. Tested on two homodimer datasets, CDPred achieves the precision of 60.94% and 42.93% for top L/5 inter-chain contact predictions (L: length of the monomer in homodimer), respectively, substantially higher than DeepHomo’s 37.40% and 23.08% and GLINTER’s 48.09% and 36.74%. Tested on the two heterodimer datasets, the top Ls/5 inter-chain contact prediction precision (Ls: length of the shorter monomer in heterodimer) of CDPred is 47.59% and 22.87% respectively, surpassing GLINTER’s 23.24% and 13.49%. Moreover, the prediction of CDPred is complementary with that of AlphaFold2-multimer.

残基间距离（residue-residue distance）信息对于预测蛋白质单体的三级结构或蛋白质复合物的四级结构具有重要应用价值。目前已有诸多深度学习方法可精准预测单体的链内残基间距离，但鲜有方法能准确预测复合物的链间残基间距离。我们基于二维注意力增强残差网络开发了一款深度学习方法CDPred（Complex Distance Prediction，复合物距离预测），以填补这一研究空白。在两个同二聚体数据集上的测试结果显示，针对前L/5个链间接触预测（L为同二聚体中单体的长度），CDPred的精确率分别达到60.94%和42.93%，显著高于DeepHomo的37.40%、23.08%以及GLINTER的48.09%和36.74%。在两个异二聚体数据集上的测试中，针对前Ls/5个链间接触预测（Ls为异二聚体中较短单体的长度），CDPred的精确率分别为47.59%和22.87%，优于GLINTER的23.24%与13.49%。此外，CDPred的预测结果与AlphaFold2-multimer的预测结果具有互补性。