Fingerprinting Interactions between Proteins and Ligands for Facilitating Machine Learning in Drug Discovery

被引:9
作者
Li, Zoe [1 ]
Huang, Ruili [2 ]
Xia, Menghang [2 ]
Patterson, Tucker A. [1 ]
Hong, Huixiao [1 ]
机构
[1] US FDA, Natl Ctr Toxicol Res, Jefferson, AR 72079 USA
[2] NIH, Natl Ctr Adv Translat Sci, Bethesda, MD 20892 USA
关键词
molecular fingerprints; 3D structural interaction fingerprints; machine learning; drug discovery; structure-activity relationships; protein-ligand interactions; predictive modeling; QSAR;
D O I
10.3390/biom14010072
中图分类号
Q5 [生物化学]; Q7 [分子生物学];
学科分类号
071010 ; 081704 ;
摘要
Molecular recognition is fundamental in biology, underpinning intricate processes through specific protein-ligand interactions. This understanding is pivotal in drug discovery, yet traditional experimental methods face limitations in exploring the vast chemical space. Computational approaches, notably quantitative structure-activity/property relationship analysis, have gained prominence. Molecular fingerprints encode molecular structures and serve as property profiles, which are essential in drug discovery. While two-dimensional (2D) fingerprints are commonly used, three-dimensional (3D) structural interaction fingerprints offer enhanced structural features specific to target proteins. Machine learning models trained on interaction fingerprints enable precise binding prediction. Recent focus has shifted to structure-based predictive modeling, with machine-learning scoring functions excelling due to feature engineering guided by key interactions. Notably, 3D interaction fingerprints are gaining ground due to their robustness. Various structural interaction fingerprints have been developed and used in drug discovery, each with unique capabilities. This review recapitulates the developed structural interaction fingerprints and provides two case studies to illustrate the power of interaction fingerprint-driven machine learning. The first elucidates structure-activity relationships in beta 2 adrenoceptor ligands, demonstrating the ability to differentiate agonists and antagonists. The second employs a retrosynthesis-based pre-trained molecular representation to predict protein-ligand dissociation rates, offering insights into binding kinetics. Despite remarkable progress, challenges persist in interpreting complex machine learning models built on 3D fingerprints, emphasizing the need for strategies to make predictions interpretable. Binding site plasticity and induced fit effects pose additional complexities. Interaction fingerprints are promising but require continued research to harness their full potential.
引用
收藏
页数:12
相关论文
共 56 条
[21]   Deep Learning Models for Predicting Gas Adsorption Capacity of Nanomaterials [J].
Guo, Wenjing ;
Liu, Jie ;
Dong, Fan ;
Chen, Ru ;
Das, Jayanti ;
Ge, Weigong ;
Xu, Xiaoming ;
Hong, Huixiao .
NANOMATERIALS, 2022, 12 (19)
[22]   ELECTROTOPOLOGICAL STATE INDEXES FOR ATOM TYPES - A NOVEL COMBINATION OF ELECTRONIC, TOPOLOGICAL, AND VALENCE STATE INFORMATION [J].
HALL, LH ;
KIER, LB .
JOURNAL OF CHEMICAL INFORMATION AND COMPUTER SCIENCES, 1995, 35 (06) :1039-1045
[23]   CORRELATION OF BIOLOGICAL ACTIVITY OF PHENOXYACETIC ACIDS WITH HAMMETT SUBSTITUENT CONSTANTS AND PARTITION COEFFICIENTS [J].
HANSCH, C ;
MALONEY, PP ;
FUJITA, T .
NATURE, 1962, 194 (4824) :178-&
[24]  
Hong H., 2023, Computational Methods in Engineering & the Sciences, P297
[25]   Mold2, molecular descriptors from 2D structures for chemoinformatics and toxicoinformatics [J].
Hong, Huixiao ;
Xie, Qian ;
Ge, Weigong ;
Qian, Feng ;
Fang, Hong ;
Shi, Leming ;
Su, Zhenqiang ;
Perkins, Roger ;
Tong, Weida .
JOURNAL OF CHEMICAL INFORMATION AND MODELING, 2008, 48 (07) :1337-1344
[26]   Quantitative Structure-Activity Relationship Models for Predicting Inflammatory Potential of Metal Oxide Nanoparticles [J].
Huang, Yang ;
Li, Xuehua ;
Xu, Shujuan ;
Zheng, Huizhen ;
Zhang, Lili ;
Chen, Jingwen ;
Hong, Huixiao ;
Kusko, Rebecca ;
Li, Ruibin .
ENVIRONMENTAL HEALTH PERSPECTIVES, 2020, 128 (06) :1-13
[27]   Structure-activity relationship-based chemical classification of highly imbalanced Tox21 datasets [J].
Idakwo, Gabriel ;
Thangapandian, Sundar ;
Luttrell, Joseph ;
Li, Yan ;
Wang, Nan ;
Zhou, Zhaoxian ;
Hong, Huixiao ;
Yang, Bei ;
Zhang, Chaoyang ;
Gong, Ping .
JOURNAL OF CHEMINFORMATICS, 2020, 12 (01)
[28]   Protein-protein recognition [J].
Janin, J .
PROGRESS IN BIOPHYSICS & MOLECULAR BIOLOGY, 1995, 64 (2-3) :145-166
[29]   KDEEP: Protein-Ligand Absolute Binding Affinity Prediction via 3D-Convolutional Neural Networks [J].
Jimenez, Jose ;
Skalic, Miha ;
Martinez-Rosell, Gerard ;
De Fabritiis, Gianni .
JOURNAL OF CHEMICAL INFORMATION AND MODELING, 2018, 58 (02) :287-296
[30]   Combined docking and machine learning identify key molecular determinants of ligand pharmacological activity on β2 adrenoceptor [J].
Jimenez-Roses, Mireia ;
Morgan, Bradley Angus ;
Sigstad, Maria Jimenez ;
Thuy Duong Zoe Tran ;
Srivastava, Rohini ;
Bunsuz, Asuman ;
Borrega-Roman, Leire ;
Hompluem, Pattarin ;
Cullum, Sean A. ;
Harwood, Clare R. ;
Koers, Eline J. ;
Sykes, David A. ;
Styles, Iain B. ;
Veprintsev, Dmitry B. .
PHARMACOLOGY RESEARCH & PERSPECTIVES, 2022, 10 (05)