Deep learning applications for the accurate identification of low-transcriptional activity drugs and their mechanism of actions
Analysis of drug-induced expression profiles facilitated comprehensive understanding of drug properties. However, many compounds exhibit weak transcription responses though they mostly possess definite pharmacological effects. Actually, as a representative example, over 66.4% of 312,438 molecular si...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2022-06-01
|
| Series: | Pharmacological Research |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S1043661822001700 |
| _version_ | 1797366702106738688 |
|---|---|
| author | Shengqiao Gao Lu Han Dan Luo Zhiyong Xiao Gang Liu Yongxiang Zhang Wenxia Zhou |
| author_facet | Shengqiao Gao Lu Han Dan Luo Zhiyong Xiao Gang Liu Yongxiang Zhang Wenxia Zhou |
| author_sort | Shengqiao Gao |
| collection | DOAJ |
| description | Analysis of drug-induced expression profiles facilitated comprehensive understanding of drug properties. However, many compounds exhibit weak transcription responses though they mostly possess definite pharmacological effects. Actually, as a representative example, over 66.4% of 312,438 molecular signatures in the Library of Integrated Cellular Signatures (LINCS) database exhibit low-transcriptional activities (i.e. TAS-low signatures). When computing the association between TAS-low signatures with shared mechanism of actions (MOAs), commonly used algorithms showed inadequate performance with an average area under receiver operating characteristic curve (AUROC) of 0.55, but the computation accuracy of the same task can be improved by our developed tool Genetic profile activity relationship (GPAR) with an average AUROC of 0.68. Up to 36 out of 74 TAS-low MOAs were well trained with AUROC ≥ 0.7 by GPAR, higher than those by other approaches. Further studies showed that GPAR benefited from the size of training samples more significantly than other approaches. Lastly, in biological validation of the MOA prediction for a TAS-low drug Tropisetron, we found an unreported mechanism that Tropisetron can bind to the glucocorticoid receptor. This study indicated that GPAR can serve as an effective approach for the accurate identification of low-transcriptional activity drugs and their MOAs, thus providing a good tool for drug repurposing with both TAS-low and TAS-high signatures. |
| first_indexed | 2024-03-08T17:07:29Z |
| format | Article |
| id | doaj.art-ffd49555dbab4d00a4ab821e7c04da72 |
| institution | Directory Open Access Journal |
| issn | 1096-1186 |
| language | English |
| last_indexed | 2024-03-08T17:07:29Z |
| publishDate | 2022-06-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Pharmacological Research |
| spelling | doaj.art-ffd49555dbab4d00a4ab821e7c04da722024-01-04T04:37:17ZengElsevierPharmacological Research1096-11862022-06-01180106225Deep learning applications for the accurate identification of low-transcriptional activity drugs and their mechanism of actionsShengqiao Gao0Lu Han1Dan Luo2Zhiyong Xiao3Gang Liu4Yongxiang Zhang5Wenxia Zhou6Beijing Institute of Pharmacology and Toxicology, State Key Laboratory of Toxicology and Medical Countermeasures, Beijing 100850, ChinaBeijing Institute of Pharmacology and Toxicology, State Key Laboratory of Toxicology and Medical Countermeasures, Beijing 100850, ChinaBeijing Institute of Pharmacology and Toxicology, State Key Laboratory of Toxicology and Medical Countermeasures, Beijing 100850, ChinaBeijing Institute of Pharmacology and Toxicology, State Key Laboratory of Toxicology and Medical Countermeasures, Beijing 100850, ChinaBeijing Institute of Pharmacology and Toxicology, State Key Laboratory of Toxicology and Medical Countermeasures, Beijing 100850, ChinaBeijing Institute of Pharmacology and Toxicology, State Key Laboratory of Toxicology and Medical Countermeasures, Beijing 100850, ChinaCorresponding author.; Beijing Institute of Pharmacology and Toxicology, State Key Laboratory of Toxicology and Medical Countermeasures, Beijing 100850, ChinaAnalysis of drug-induced expression profiles facilitated comprehensive understanding of drug properties. However, many compounds exhibit weak transcription responses though they mostly possess definite pharmacological effects. Actually, as a representative example, over 66.4% of 312,438 molecular signatures in the Library of Integrated Cellular Signatures (LINCS) database exhibit low-transcriptional activities (i.e. TAS-low signatures). When computing the association between TAS-low signatures with shared mechanism of actions (MOAs), commonly used algorithms showed inadequate performance with an average area under receiver operating characteristic curve (AUROC) of 0.55, but the computation accuracy of the same task can be improved by our developed tool Genetic profile activity relationship (GPAR) with an average AUROC of 0.68. Up to 36 out of 74 TAS-low MOAs were well trained with AUROC ≥ 0.7 by GPAR, higher than those by other approaches. Further studies showed that GPAR benefited from the size of training samples more significantly than other approaches. Lastly, in biological validation of the MOA prediction for a TAS-low drug Tropisetron, we found an unreported mechanism that Tropisetron can bind to the glucocorticoid receptor. This study indicated that GPAR can serve as an effective approach for the accurate identification of low-transcriptional activity drugs and their MOAs, thus providing a good tool for drug repurposing with both TAS-low and TAS-high signatures.http://www.sciencedirect.com/science/article/pii/S1043661822001700Gene expression profilesDeep learningLow transcriptional activity |
| spellingShingle | Shengqiao Gao Lu Han Dan Luo Zhiyong Xiao Gang Liu Yongxiang Zhang Wenxia Zhou Deep learning applications for the accurate identification of low-transcriptional activity drugs and their mechanism of actions Pharmacological Research Gene expression profiles Deep learning Low transcriptional activity |
| title | Deep learning applications for the accurate identification of low-transcriptional activity drugs and their mechanism of actions |
| title_full | Deep learning applications for the accurate identification of low-transcriptional activity drugs and their mechanism of actions |
| title_fullStr | Deep learning applications for the accurate identification of low-transcriptional activity drugs and their mechanism of actions |
| title_full_unstemmed | Deep learning applications for the accurate identification of low-transcriptional activity drugs and their mechanism of actions |
| title_short | Deep learning applications for the accurate identification of low-transcriptional activity drugs and their mechanism of actions |
| title_sort | deep learning applications for the accurate identification of low transcriptional activity drugs and their mechanism of actions |
| topic | Gene expression profiles Deep learning Low transcriptional activity |
| url | http://www.sciencedirect.com/science/article/pii/S1043661822001700 |
| work_keys_str_mv | AT shengqiaogao deeplearningapplicationsfortheaccurateidentificationoflowtranscriptionalactivitydrugsandtheirmechanismofactions AT luhan deeplearningapplicationsfortheaccurateidentificationoflowtranscriptionalactivitydrugsandtheirmechanismofactions AT danluo deeplearningapplicationsfortheaccurateidentificationoflowtranscriptionalactivitydrugsandtheirmechanismofactions AT zhiyongxiao deeplearningapplicationsfortheaccurateidentificationoflowtranscriptionalactivitydrugsandtheirmechanismofactions AT gangliu deeplearningapplicationsfortheaccurateidentificationoflowtranscriptionalactivitydrugsandtheirmechanismofactions AT yongxiangzhang deeplearningapplicationsfortheaccurateidentificationoflowtranscriptionalactivitydrugsandtheirmechanismofactions AT wenxiazhou deeplearningapplicationsfortheaccurateidentificationoflowtranscriptionalactivitydrugsandtheirmechanismofactions |