Improving ΔΔG Predictions with a Multitask Convolutional Siamese Network.

Improving ΔΔG Predictions with a Multitask Convolutional Siamese Network.
复制标题

DOI:
10.1021/acs.jcim.1c01497
复制
发表时间:
2022-04-25
影响因子:
5.6
通讯作者:
Koes, David Ryan
Koes, David Ryan
中科院分区:
化学2区
文献类型:
--
作者:
McNutt, Andrew T.;Koes, David Ryan

文献摘要

参考文献

被引文献

相似文献

药物发现的前导优化阶段提炼出最初的受攻击分子,使其具有所需的性质,特别是效力。在这种精化过程中对微小扰动进行合成和实验测试可能是相当昂贵和耗时的。相对结合自由能(RbFE,也称为ΔΔG)方法可以估计配体支架发生微小变化后结合自由能的变化。在这里,我们提出并评估了卷积神经网络(CNN)暹罗网络用于预测两个结合配体之间的RBFE。我们表明,通过增加潜在空间的正则化,我们的多任务损失能够改进先前最先进的用于RBFE预测的暹罗网络。与基于相同数据训练的标准有线电视新闻网(皮尔逊的R分别为0.553和0.5时)相比,暹罗网络结构非常适合预测RBFE。当对遗漏的蛋白质家族进行评估时,我们的CNN暹罗网络根据被评估的蛋白质家族(皮尔逊的R从−0.44到0.97)显示出其RBFE预测性能的变异性。在泛化过程中,通过在模型训练过程中仅从评估数据集中注入几个示例(少镜头学习),可以提高RBFE预测性能。
The lead optimization phase of drug discovery refines an initial hit molecule for desired properties, especially potency. Synthesis and experimental testing of the small perturbations during this refinement can be quite costly and time consuming. Relative binding free energy (RBFE, also referred to as ΔΔG) methods allow the estimation of binding free energy changes after small changes to a ligand scaffold. Here we propose and evaluate a Convolutional Neural Network (CNN) Siamese network for the prediction of RBFE between two bound ligands. We show that our multi-task loss is able to improve on a previous state-of-the-art Siamese network for RBFE prediction via increased regularization of the latent space. The Siamese network architecture is well suited to the prediction of RBFE in comparison to a standard CNN trained on the same data (Pearson’s R of 0.553 and 0.5, respectively). When evaluated on a left-out protein family, our CNN Siamese network shows variability in its RBFE predictive performance depending on the protein family being evaluated (Pearson’s R ranging from −0.44 to 0.97). RBFE prediction performance can be improved during generalization by injecting only a few examples (few-shot learning) from the evaluation dataset during model training.
DOI: 10.1093/nar/gkaa913
发表时间: 2021-01-08
影响因子: 14.9
作者:
Mistry J;Chuguransky S;Williams L;Qureshi M;Salazar GA;Sonnhammer ELL;Tosatto SCE;Paladin L;Raj S;Richardson LJ;Finn RD;Bateman A
通讯作者: Bateman A
DOI: 10.1021/acs.jcim.0c00411
发表时间: 2020-09-28
影响因子: 5.6
作者:
Francoeur PG;Masuda T;Sunseri J;Jia A;Iovanisci RB;Snyder I;Koes DR
通讯作者: Koes DR
DOI: 10.1093/bioinformatics/bty374
发表时间: 2018-11-01
期刊: Bioinformatics (Oxford, England)
影响因子: --
作者:
Stepniewska-Dziubinska MM;Zielenkiewicz P;Siedlecki P
通讯作者: Siedlecki P
DOI: 10.1517/17460441.2015.1032936
发表时间: 2015-05
影响因子: 6.3
作者:
Genheden S;Ryde U
通讯作者: Ryde U
DOI: 10.1007/s10822-011-9533-y
发表时间: 2012-06
影响因子: 3.5
作者:
Spitzer, Russell;Jain, Ajay N.
通讯作者: Jain, Ajay N.