Developing a Kinase-Specific Target Selection Method Using a Structure-Based Machine Learning Approach
Arina Afanasyeva,1 Chioko Nagao,1,2 Kenji Mizuguchi1,2 1Bioinformatics Project, National Institutes of Biomedical Innovation, Health and Nutrition, Osaka, Japan; 2Institute for Protein Research, Osaka University, Osaka, JapanCorrespondence: Arina Afanasyeva Email arina.afan@gmail.comIntroduction: De...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Dove Medical Press
2020-12-01
|
| Series: | Advances and Applications in Bioinformatics and Chemistry |
| Subjects: | |
| Online Access: | https://www.dovepress.com/developing-a-kinase-specific-target-selection-method-using-a-structure-peer-reviewed-article-AABC |
| _version_ | 1818864833776320512 |
|---|---|
| author | Afanasyeva A Nagao C Mizuguchi K |
| author_facet | Afanasyeva A Nagao C Mizuguchi K |
| author_sort | Afanasyeva A |
| collection | DOAJ |
| description | Arina Afanasyeva,1 Chioko Nagao,1,2 Kenji Mizuguchi1,2 1Bioinformatics Project, National Institutes of Biomedical Innovation, Health and Nutrition, Osaka, Japan; 2Institute for Protein Research, Osaka University, Osaka, JapanCorrespondence: Arina Afanasyeva Email arina.afan@gmail.comIntroduction: Despite recent advances in the drug discovery field, developing selective kinase inhibitors remains a complicated issue for a number of reasons, one of which is that there are striking structural similarities in the ATP-binding pockets of kinases.Objective: To address this problem, we have designed a machine learning model utilizing various structure-based and energy-based descriptors to better characterize protein–ligand interactions.Methods: In this work, we use a dataset of 104 human kinases with available PDB structures and experimental activity data against 1202 small-molecule compounds from the PubChem BioAssay dataset “Navigating the Kinome”. We propose structure-based interaction descriptors to build activity predicting machine learning model.Results and Discussion: We report a ligand-oriented computational method for accurate kinase target prioritizing. Our method shows high accuracy compared to similar structure-based activity prediction methods, and more importantly shows the same prediction accuracy when tested on the special set of structurally remote compounds, showing that it is unbiased to ligand structural similarity in the training set data. We hope that our approach will be useful for the development of novel highly selective kinase inhibitors.Keywords: kinase, machine learning, activity prediction, docking, interaction descriptors |
| first_indexed | 2024-12-19T10:37:56Z |
| format | Article |
| id | doaj.art-b388cec65a5e49d496ebfb10ad7de469 |
| institution | Directory Open Access Journal |
| issn | 1178-6949 |
| language | English |
| last_indexed | 2024-12-19T10:37:56Z |
| publishDate | 2020-12-01 |
| publisher | Dove Medical Press |
| record_format | Article |
| series | Advances and Applications in Bioinformatics and Chemistry |
| spelling | doaj.art-b388cec65a5e49d496ebfb10ad7de4692022-12-21T20:25:32ZengDove Medical PressAdvances and Applications in Bioinformatics and Chemistry1178-69492020-12-01Volume 13274059896Developing a Kinase-Specific Target Selection Method Using a Structure-Based Machine Learning ApproachAfanasyeva ANagao CMizuguchi KArina Afanasyeva,1 Chioko Nagao,1,2 Kenji Mizuguchi1,2 1Bioinformatics Project, National Institutes of Biomedical Innovation, Health and Nutrition, Osaka, Japan; 2Institute for Protein Research, Osaka University, Osaka, JapanCorrespondence: Arina Afanasyeva Email arina.afan@gmail.comIntroduction: Despite recent advances in the drug discovery field, developing selective kinase inhibitors remains a complicated issue for a number of reasons, one of which is that there are striking structural similarities in the ATP-binding pockets of kinases.Objective: To address this problem, we have designed a machine learning model utilizing various structure-based and energy-based descriptors to better characterize protein–ligand interactions.Methods: In this work, we use a dataset of 104 human kinases with available PDB structures and experimental activity data against 1202 small-molecule compounds from the PubChem BioAssay dataset “Navigating the Kinome”. We propose structure-based interaction descriptors to build activity predicting machine learning model.Results and Discussion: We report a ligand-oriented computational method for accurate kinase target prioritizing. Our method shows high accuracy compared to similar structure-based activity prediction methods, and more importantly shows the same prediction accuracy when tested on the special set of structurally remote compounds, showing that it is unbiased to ligand structural similarity in the training set data. We hope that our approach will be useful for the development of novel highly selective kinase inhibitors.Keywords: kinase, machine learning, activity prediction, docking, interaction descriptorshttps://www.dovepress.com/developing-a-kinase-specific-target-selection-method-using-a-structure-peer-reviewed-article-AABCkinasemachine learningactivity predictiondockinginteraction descriptors |
| spellingShingle | Afanasyeva A Nagao C Mizuguchi K Developing a Kinase-Specific Target Selection Method Using a Structure-Based Machine Learning Approach Advances and Applications in Bioinformatics and Chemistry kinase machine learning activity prediction docking interaction descriptors |
| title | Developing a Kinase-Specific Target Selection Method Using a Structure-Based Machine Learning Approach |
| title_full | Developing a Kinase-Specific Target Selection Method Using a Structure-Based Machine Learning Approach |
| title_fullStr | Developing a Kinase-Specific Target Selection Method Using a Structure-Based Machine Learning Approach |
| title_full_unstemmed | Developing a Kinase-Specific Target Selection Method Using a Structure-Based Machine Learning Approach |
| title_short | Developing a Kinase-Specific Target Selection Method Using a Structure-Based Machine Learning Approach |
| title_sort | developing a kinase specific target selection method using a structure based machine learning approach |
| topic | kinase machine learning activity prediction docking interaction descriptors |
| url | https://www.dovepress.com/developing-a-kinase-specific-target-selection-method-using-a-structure-peer-reviewed-article-AABC |
| work_keys_str_mv | AT afanasyevaa developingakinasespecifictargetselectionmethodusingastructurebasedmachinelearningapproach AT nagaoc developingakinasespecifictargetselectionmethodusingastructurebasedmachinelearningapproach AT mizuguchik developingakinasespecifictargetselectionmethodusingastructurebasedmachinelearningapproach |