Fingerprinting Interactions between Proteins and Ligands for Facilitating Machine Learning in Drug Discovery
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, notab...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-01-01
|
| Series: | Biomolecules |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2218-273X/14/1/72 |
| _version_ | 1797344576196837376 |
|---|---|
| author | Zoe Li Ruili Huang Menghang Xia Tucker A. Patterson Huixiao Hong |
| author_facet | Zoe Li Ruili Huang Menghang Xia Tucker A. Patterson Huixiao Hong |
| author_sort | Zoe Li |
| collection | DOAJ |
| description | 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 β2 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. |
| first_indexed | 2024-03-08T11:04:43Z |
| format | Article |
| id | doaj.art-bd65eaeb51bc4c1db555d6283325e56e |
| institution | Directory Open Access Journal |
| issn | 2218-273X |
| language | English |
| last_indexed | 2024-03-08T11:04:43Z |
| publishDate | 2024-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Biomolecules |
| spelling | doaj.art-bd65eaeb51bc4c1db555d6283325e56e2024-01-26T15:19:20ZengMDPI AGBiomolecules2218-273X2024-01-011417210.3390/biom14010072Fingerprinting Interactions between Proteins and Ligands for Facilitating Machine Learning in Drug DiscoveryZoe Li0Ruili Huang1Menghang Xia2Tucker A. Patterson3Huixiao Hong4National Center for Toxicological Research, US Food and Drug Administration, Jefferson, AR 72079, USANational Center for Advancing Translational Sciences, National Institutes of Health, Bethesda, MD 20892, USANational Center for Advancing Translational Sciences, National Institutes of Health, Bethesda, MD 20892, USANational Center for Toxicological Research, US Food and Drug Administration, Jefferson, AR 72079, USANational Center for Toxicological Research, US Food and Drug Administration, Jefferson, AR 72079, USAMolecular 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 β2 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.https://www.mdpi.com/2218-273X/14/1/72molecular fingerprints3D structural interaction fingerprintsmachine learningdrug discoverystructure–activity relationshipsprotein–ligand interactions |
| spellingShingle | Zoe Li Ruili Huang Menghang Xia Tucker A. Patterson Huixiao Hong Fingerprinting Interactions between Proteins and Ligands for Facilitating Machine Learning in Drug Discovery Biomolecules molecular fingerprints 3D structural interaction fingerprints machine learning drug discovery structure–activity relationships protein–ligand interactions |
| title | Fingerprinting Interactions between Proteins and Ligands for Facilitating Machine Learning in Drug Discovery |
| title_full | Fingerprinting Interactions between Proteins and Ligands for Facilitating Machine Learning in Drug Discovery |
| title_fullStr | Fingerprinting Interactions between Proteins and Ligands for Facilitating Machine Learning in Drug Discovery |
| title_full_unstemmed | Fingerprinting Interactions between Proteins and Ligands for Facilitating Machine Learning in Drug Discovery |
| title_short | Fingerprinting Interactions between Proteins and Ligands for Facilitating Machine Learning in Drug Discovery |
| title_sort | fingerprinting interactions between proteins and ligands for facilitating machine learning in drug discovery |
| topic | molecular fingerprints 3D structural interaction fingerprints machine learning drug discovery structure–activity relationships protein–ligand interactions |
| url | https://www.mdpi.com/2218-273X/14/1/72 |
| work_keys_str_mv | AT zoeli fingerprintinginteractionsbetweenproteinsandligandsforfacilitatingmachinelearningindrugdiscovery AT ruilihuang fingerprintinginteractionsbetweenproteinsandligandsforfacilitatingmachinelearningindrugdiscovery AT menghangxia fingerprintinginteractionsbetweenproteinsandligandsforfacilitatingmachinelearningindrugdiscovery AT tuckerapatterson fingerprintinginteractionsbetweenproteinsandligandsforfacilitatingmachinelearningindrugdiscovery AT huixiaohong fingerprintinginteractionsbetweenproteinsandligandsforfacilitatingmachinelearningindrugdiscovery |