Peptide-binding specificity prediction using fine-tuned protein structure prediction networks.
Peptide-binding specificity prediction using fine-tuned protein structure prediction networks.
复制标题
利用微调蛋白结构预测网络进行肽结合特异性预测。
DOI：
10.1073/pnas.2216697120
复制
发表时间：
2023-02-28
影响因子：
11.1
通讯作者：
Bradley, Philip
中科院分区：
文献类型：
--
作者：
Motmaen, Amir;Dauparas, Justas;Baek, Minkyung;Abedi, Mohamad H.;Baker, David;Bradley, Philip
关键词：
肽结合蛋白在细胞中具有多种生物学功能，预测它们的结合特异性可以显著提高我们对分子途径的理解。深度神经网络已经实现了很高的结构预测准确性，但没有经过训练来预测结合特异性。在这里，我们描述了一种方法来扩展这样的网络，共同预测蛋白质结构和结合特异性。我们将AlphaFold纳入这种方法，并根据肽-MHC I类和II类结构和结合数据对其参数进行微调。微调模型接近肽-MHC特异性预测的最先进的分类准确性，并推广到其他肽结合系统，如PDZ和SH 3结构域。肽结合蛋白在生物学中起着关键作用，预测其结合特异性是一个长期的挑战。虽然大量的蛋白质结构信息是可用的，但目前最成功的方法仅使用序列信息，部分原因是对伴随序列取代的微妙结构变化进行建模一直是一个挑战。蛋白质结构预测网络，如AlphaFold模型序列-结构关系非常准确，我们推断，如果有可能在结合数据上专门训练这样的网络，就可以创建更通用的模型。我们表明，将分类器放置在AlphaFold网络之上，并针对分类和结构预测准确性微调组合网络参数，可以产生一种在广泛的I类和II类肽-MHC相互作用上具有强大可推广性能的模型，该模型接近最先进的基于NetMHCpan序列的方法的整体性能。肽-MHC优化模型在区分SH 3和PDZ结构域的结合和非结合肽方面显示出优异的性能。这种泛化能力远远超过了训练集，远远超过了仅序列模型，对于实验数据较少的系统来说应该特别强大。
Peptide-binding proteins carry out a variety of biological functions in cells and predicting their binding specificity could significantly improve our understanding of molecular pathways. Deep neural networks have achieved high structure prediction accuracy, but are not trained to predict binding specificity. Here we describe an approach to extending such networks to jointly predict protein structure and binding specificity. We incorporate AlphaFold into this approach and fine-tune its parameters on peptide-MHC Class I and II structural and binding data. The fine-tuned model approaches state-of-the-art classification accuracy on peptide-MHC specificity prediction and generalizes to other peptide-binding systems such as the PDZ and SH3 domains. Peptide-binding proteins play key roles in biology, and predicting their binding specificity is a long-standing challenge. While considerable protein structural information is available, the most successful current methods use sequence information alone, in part because it has been a challenge to model the subtle structural changes accompanying sequence substitutions. Protein structure prediction networks such as AlphaFold model sequence-structure relationships very accurately, and we reasoned that if it were possible to specifically train such networks on binding data, more generalizable models could be created. We show that placing a classifier on top of the AlphaFold network and fine-tuning the combined network parameters for both classification and structure prediction accuracy leads to a model with strong generalizable performance on a wide range of Class I and Class II peptide-MHC interactions that approaches the overall performance of the state-of-the-art NetMHCpan sequence-based method. The peptide-MHC optimized model shows excellent performance in distinguishing binding and non-binding peptides to SH3 and PDZ domains. This ability to generalize well beyond the training set far exceeds that of sequence-only models and should be particularly powerful for systems where less experimental data are available.
登录
查看更多内容
影响因子：
4.6
作者：
Antunes DA;Devaurs D;Moll M;Lizée G;Kavraki LE
通讯作者：
Kavraki LE
影响因子：
64.8
作者：
Jumper J;Evans R;Pritzel A;Green T;Figurnov M;Ronneberger O;Tunyasuvunakool K;Bates R;Žídek A;Potapenko A;Bridgland A;Meyer C;Kohl SAA;Ballard AJ;Cowie A;Romera-Paredes B;Nikolov S;Jain R;Adler J;Back T;Petersen S;Reiman D;Clancy E;Zielinski M;Steinegger M;Pacholska M;Berghammer T;Bodenstein S;Silver D;Vinyals O;Senior AW;Kavukcuoglu K;Kohli P;Hassabis D
通讯作者：
Hassabis D
影响因子：
32.4
作者：
Abelin JG;Keskin DB;Sarkizova S;Hartigan CR;Zhang W;Sidney J;Stevens J;Lane W;Zhang GL;Eisenhaure TM;Clauser KR;Hacohen N;Rooney MS;Carr SA;Wu CJ
通讯作者：
Wu CJ
Biopython：用于计算分子生物学和生物信息学的免费 Python 工具。
DOI：
10.1093/bioinformatics/btp163
发表时间：
2009-06-01
期刊：
Bioinformatics (Oxford, England)
影响因子：
--
作者：
Cock PJ;Antao T;Chang JT;Chapman BA;Cox CJ;Dalke A;Friedberg I;Hamelryck T;Kauff F;Wilczynski B;de Hoon MJ
通讯作者：
de Hoon MJ
影响因子：
5.6
作者：
Smith, Colin A.;Kortemme, Tanja
通讯作者：
Kortemme, Tanja