ComABAN: refining molecular representation with the graph attention mechanism to accelerate drug discovery

ComABAN: refining molecular representation with the graph attention mechanism to accelerate drug discovery
复制标题

ComABAN:利用图注意力机制完善分子表示以加速药物发现

DOI:
10.1093/bib/bbac350
复制
发表时间:
2022-08
影响因子:
9.5
通讯作者:
Rong Sheng
Rong Sheng
中科院分区:
生物学2区
文献类型:
--
作者:
Huihui Yan;Yuanyuan Xie;Yao Liu;Leer Yuan;Rong Sheng

文献摘要

参考文献

相似文献

摘要在发展分子表征中一个尚未解决的挑战是确定表征分子结构的最佳方法。分子内相互作用的理解是实现这一目标的最重要的。在这项研究中,ComABAN,一个新的图形注意力为基础的方法,提出了提高分子表示的准确性,同时考虑原子-原子,键-键和原子-键的相互作用。此外,我们在8个公共和680个专有工业数据集上对模型进行了广泛的基准测试,这些数据集涵盖了各种各样的化学终点。结果表明,与经典的机器学习方法和基于深度学习的方法相比,ComABAN具有更高的预测精度。此外,训练好的神经网络用于预测150万个分子的库,并挑选出分类结果为I级的化合物。随后,使用级联对接、分子动力学模拟对这些预测的分子进行评分和排名,以产生五个潜在的候选分子。这5个分子与纳摩尔级抑制HIF-1α表达的生物活性抑制剂具有高度的相似性,我们合成了3个化合物(Y-1,Y-3,Y-4),并在体外测试了它们的抑制能力。我们的研究结果表明,ComABAN是加速药物发现的有效工具。
Abstract An unsolved challenge in developing molecular representation is determining an optimal method to characterize the molecular structure. Comprehension of intramolecular interactions is paramount toward achieving this goal. In this study, ComABAN, a new graph-attention-based approach, is proposed to improve the accuracy of molecular representation by simultaneously considering atom–atom, bond–bond and atom-bond interactions. In addition, we benchmark models extensively on 8 public and 680 proprietary industrial datasets spanning a wide variety of chemical end points. The results show that ComABAN has higher prediction accuracy compared with the classical machine learning method and the deep learning-based methods. Furthermore, the trained neural network was used to predict a library of 1.5 million molecules and picked out compounds with a classification result of grade I. Subsequently, these predicted molecules were scored and ranked using cascade docking, molecular dynamics simulations to generate five potential candidates. All five molecules showed high similarity to nanomolar bioactive inhibitors suppressing the expression of HIF-1α, and we synthesized three compounds (Y-1, Y-3, Y-4) and tested their inhibitory ability in vitro. Our results indicate that ComABAN is an effective tool for accelerating drug discovery.
DOI: 10.1021/acs.biochem.5b01003
发表时间: 2016-01-19
期刊: Biochemistry
影响因子: 2.9
作者:
Taabazuing CY;Fermann J;Garman S;Knapp MJ
通讯作者: Knapp MJ
DOI: 10.1074/jbc.c200644200
发表时间: 2003-01-17
影响因子: 4.8
作者:
Elkins, JM;Hewitson, KS;Schofield, CJ
通讯作者: Schofield, CJ
DOI: 10.1093/nar/gkr988
发表时间: 2012-01
影响因子: 14.9
作者:
Kanehisa M;Goto S;Sato Y;Furumichi M;Tanabe M
通讯作者: Tanabe M
细胞类型特异性依赖性对PI3K/AKT信号传导途径,用于神经元与星形胶质细胞中的黄胶蛋白的内源性EPO和VEGF诱导。
DOI: 10.1371/journal.pone.0069019
发表时间: 2013
期刊: PloS one
影响因子: 3.7
作者:
Sun YY;Lin SH;Lin HC;Hung CC;Wang CY;Lin YC;Hung KS;Lien CC;Kuan CY;Lee YH
通讯作者: Lee YH
晶体中分子间相互作用的键序:电荷转移、离子性和对分子内键的影响。
DOI: 10.1107/s2052252518010758
发表时间: 2018-09-01
期刊: IUCrJ
影响因子: 3.9
作者:
Alhameedi K;Karton A;Jayatilaka D;Thomas SP
通讯作者: Thomas SP