Identification of pan-kinase-family inhibitors using graph convolutional networks to reveal family-sensitive pre-moieties
Abstract Background Human protein kinases, the key players in phosphoryl signal transduction, have been actively investigated as drug targets for complex diseases such as cancer, immune disorders, and Alzheimer’s disease, with more than 60 successful drugs developed in the past 30 years. However, ma...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2022-06-01
|
| Series: | BMC Bioinformatics |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s12859-022-04773-0 |
| _version_ | 1797744030059069440 |
|---|---|
| author | Xiang-Yu Lin Yu-Wei Huang You-Wei Fan Yun-Ti Chen Nikhil Pathak Yen-Chao Hsu Jinn-Moon Yang |
| author_facet | Xiang-Yu Lin Yu-Wei Huang You-Wei Fan Yun-Ti Chen Nikhil Pathak Yen-Chao Hsu Jinn-Moon Yang |
| author_sort | Xiang-Yu Lin |
| collection | DOAJ |
| description | Abstract Background Human protein kinases, the key players in phosphoryl signal transduction, have been actively investigated as drug targets for complex diseases such as cancer, immune disorders, and Alzheimer’s disease, with more than 60 successful drugs developed in the past 30 years. However, many of these single-kinase inhibitors show low efficacy and drug resistance has become an issue. Owing to the occurrence of highly conserved catalytic sites and shared signaling pathways within a kinase family, multi-target kinase inhibitors have attracted attention. Results To design and identify such pan-kinase family inhibitors (PKFIs), we proposed PKFI sets for eight families using 200,000 experimental bioactivity data points and applied a graph convolutional network (GCN) to build classification models. Furthermore, we identified and extracted family-sensitive (only present in a family) pre-moieties (parts of complete moieties) by utilizing a visualized explanation (i.e., where the model focuses on each input) method for deep learning, gradient-weighted class activation mapping (Grad-CAM). Conclusions This study is the first to propose the PKFI sets, and our results point out and validate the power of GCN models in understanding the pre-moieties of PKFIs within and across different kinase families. Moreover, we highlight the discoverability of family-sensitive pre-moieties in PKFI identification and drug design. |
| first_indexed | 2024-03-12T15:03:52Z |
| format | Article |
| id | doaj.art-df07a3ef2fba48bf879af1badd82978c |
| institution | Directory Open Access Journal |
| issn | 1471-2105 |
| language | English |
| last_indexed | 2024-03-12T15:03:52Z |
| publishDate | 2022-06-01 |
| publisher | BMC |
| record_format | Article |
| series | BMC Bioinformatics |
| spelling | doaj.art-df07a3ef2fba48bf879af1badd82978c2023-08-13T11:24:52ZengBMCBMC Bioinformatics1471-21052022-06-0123S411310.1186/s12859-022-04773-0Identification of pan-kinase-family inhibitors using graph convolutional networks to reveal family-sensitive pre-moietiesXiang-Yu Lin0Yu-Wei Huang1You-Wei Fan2Yun-Ti Chen3Nikhil Pathak4Yen-Chao Hsu5Jinn-Moon Yang6Institute of Bioinformatics and Systems Biology, National Yang Ming Chiao Tung UniversityInstitute of Biomedical Engineering, National Yang Ming Chiao Tung UniversityInstitute of Molecular Medicine and Bioengineering, National Yang Ming Chiao Tung UniversityInstitute of Bioinformatics and Systems Biology, National Yang Ming Chiao Tung UniversityInstitute of Bioinformatics and Structural Biology, National Tsing Hua UniversityInstitute of Bioinformatics and Systems Biology, National Yang Ming Chiao Tung UniversityInstitute of Bioinformatics and Systems Biology, National Yang Ming Chiao Tung UniversityAbstract Background Human protein kinases, the key players in phosphoryl signal transduction, have been actively investigated as drug targets for complex diseases such as cancer, immune disorders, and Alzheimer’s disease, with more than 60 successful drugs developed in the past 30 years. However, many of these single-kinase inhibitors show low efficacy and drug resistance has become an issue. Owing to the occurrence of highly conserved catalytic sites and shared signaling pathways within a kinase family, multi-target kinase inhibitors have attracted attention. Results To design and identify such pan-kinase family inhibitors (PKFIs), we proposed PKFI sets for eight families using 200,000 experimental bioactivity data points and applied a graph convolutional network (GCN) to build classification models. Furthermore, we identified and extracted family-sensitive (only present in a family) pre-moieties (parts of complete moieties) by utilizing a visualized explanation (i.e., where the model focuses on each input) method for deep learning, gradient-weighted class activation mapping (Grad-CAM). Conclusions This study is the first to propose the PKFI sets, and our results point out and validate the power of GCN models in understanding the pre-moieties of PKFIs within and across different kinase families. Moreover, we highlight the discoverability of family-sensitive pre-moieties in PKFI identification and drug design.https://doi.org/10.1186/s12859-022-04773-0Pan-kinase family inhibitorGraph convolutional networkVisualized explanationGradient-weighted class activation mappingFamily-sensitive pre-moiety |
| spellingShingle | Xiang-Yu Lin Yu-Wei Huang You-Wei Fan Yun-Ti Chen Nikhil Pathak Yen-Chao Hsu Jinn-Moon Yang Identification of pan-kinase-family inhibitors using graph convolutional networks to reveal family-sensitive pre-moieties BMC Bioinformatics Pan-kinase family inhibitor Graph convolutional network Visualized explanation Gradient-weighted class activation mapping Family-sensitive pre-moiety |
| title | Identification of pan-kinase-family inhibitors using graph convolutional networks to reveal family-sensitive pre-moieties |
| title_full | Identification of pan-kinase-family inhibitors using graph convolutional networks to reveal family-sensitive pre-moieties |
| title_fullStr | Identification of pan-kinase-family inhibitors using graph convolutional networks to reveal family-sensitive pre-moieties |
| title_full_unstemmed | Identification of pan-kinase-family inhibitors using graph convolutional networks to reveal family-sensitive pre-moieties |
| title_short | Identification of pan-kinase-family inhibitors using graph convolutional networks to reveal family-sensitive pre-moieties |
| title_sort | identification of pan kinase family inhibitors using graph convolutional networks to reveal family sensitive pre moieties |
| topic | Pan-kinase family inhibitor Graph convolutional network Visualized explanation Gradient-weighted class activation mapping Family-sensitive pre-moiety |
| url | https://doi.org/10.1186/s12859-022-04773-0 |
| work_keys_str_mv | AT xiangyulin identificationofpankinasefamilyinhibitorsusinggraphconvolutionalnetworkstorevealfamilysensitivepremoieties AT yuweihuang identificationofpankinasefamilyinhibitorsusinggraphconvolutionalnetworkstorevealfamilysensitivepremoieties AT youweifan identificationofpankinasefamilyinhibitorsusinggraphconvolutionalnetworkstorevealfamilysensitivepremoieties AT yuntichen identificationofpankinasefamilyinhibitorsusinggraphconvolutionalnetworkstorevealfamilysensitivepremoieties AT nikhilpathak identificationofpankinasefamilyinhibitorsusinggraphconvolutionalnetworkstorevealfamilysensitivepremoieties AT yenchaohsu identificationofpankinasefamilyinhibitorsusinggraphconvolutionalnetworkstorevealfamilysensitivepremoieties AT jinnmoonyang identificationofpankinasefamilyinhibitorsusinggraphconvolutionalnetworkstorevealfamilysensitivepremoieties |