Improved compound–protein interaction site and binding affinity prediction using self-supervised protein embeddings
Abstract Background Compound–protein interaction site and binding affinity predictions are crucial for drug discovery and drug design. In recent years, many deep learning-based methods have been proposed for predications related to compound–protein interaction. For protein inputs, how to make use of...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2022-12-01
|
| Series: | BMC Bioinformatics |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s12859-022-05107-w |
| _version_ | 1811291594297442304 |
|---|---|
| author | Jialin Wu Zhe Liu Xiaofeng Yang Zhanglin Lin |
| author_facet | Jialin Wu Zhe Liu Xiaofeng Yang Zhanglin Lin |
| author_sort | Jialin Wu |
| collection | DOAJ |
| description | Abstract Background Compound–protein interaction site and binding affinity predictions are crucial for drug discovery and drug design. In recent years, many deep learning-based methods have been proposed for predications related to compound–protein interaction. For protein inputs, how to make use of protein primary sequence and tertiary structure information has impact on prediction results. Results In this study, we propose a deep learning model based on a multi-objective neural network, which involves a multi-objective neural network for compound–protein interaction site and binding affinity prediction. We used several kinds of self-supervised protein embeddings to enrich our protein inputs and used convolutional neural networks to extract features from them. Our results demonstrate that our model had improvements in terms of interaction site prediction and affinity prediction compared to previous models. In a case study, our model could better predict binding sites, which also showed its effectiveness. Conclusion These results suggest that our model could be a helpful tool for compound–protein related predictions. |
| first_indexed | 2024-04-13T04:31:48Z |
| format | Article |
| id | doaj.art-dc75f63efeef4c97aed711213f30196c |
| institution | Directory Open Access Journal |
| issn | 1471-2105 |
| language | English |
| last_indexed | 2024-04-13T04:31:48Z |
| publishDate | 2022-12-01 |
| publisher | BMC |
| record_format | Article |
| series | BMC Bioinformatics |
| spelling | doaj.art-dc75f63efeef4c97aed711213f30196c2022-12-22T03:02:18ZengBMCBMC Bioinformatics1471-21052022-12-0123111210.1186/s12859-022-05107-wImproved compound–protein interaction site and binding affinity prediction using self-supervised protein embeddingsJialin Wu0Zhe Liu1Xiaofeng Yang2Zhanglin Lin3School of Biology and Biological Engineering, South China University of TechnologySchool of Biology and Biological Engineering, South China University of TechnologySchool of Biology and Biological Engineering, South China University of TechnologySchool of Biology and Biological Engineering, South China University of TechnologyAbstract Background Compound–protein interaction site and binding affinity predictions are crucial for drug discovery and drug design. In recent years, many deep learning-based methods have been proposed for predications related to compound–protein interaction. For protein inputs, how to make use of protein primary sequence and tertiary structure information has impact on prediction results. Results In this study, we propose a deep learning model based on a multi-objective neural network, which involves a multi-objective neural network for compound–protein interaction site and binding affinity prediction. We used several kinds of self-supervised protein embeddings to enrich our protein inputs and used convolutional neural networks to extract features from them. Our results demonstrate that our model had improvements in terms of interaction site prediction and affinity prediction compared to previous models. In a case study, our model could better predict binding sites, which also showed its effectiveness. Conclusion These results suggest that our model could be a helpful tool for compound–protein related predictions.https://doi.org/10.1186/s12859-022-05107-wCompound–protein interactionBinding affinityDeep learningSelf-supervised protein embedding |
| spellingShingle | Jialin Wu Zhe Liu Xiaofeng Yang Zhanglin Lin Improved compound–protein interaction site and binding affinity prediction using self-supervised protein embeddings BMC Bioinformatics Compound–protein interaction Binding affinity Deep learning Self-supervised protein embedding |
| title | Improved compound–protein interaction site and binding affinity prediction using self-supervised protein embeddings |
| title_full | Improved compound–protein interaction site and binding affinity prediction using self-supervised protein embeddings |
| title_fullStr | Improved compound–protein interaction site and binding affinity prediction using self-supervised protein embeddings |
| title_full_unstemmed | Improved compound–protein interaction site and binding affinity prediction using self-supervised protein embeddings |
| title_short | Improved compound–protein interaction site and binding affinity prediction using self-supervised protein embeddings |
| title_sort | improved compound protein interaction site and binding affinity prediction using self supervised protein embeddings |
| topic | Compound–protein interaction Binding affinity Deep learning Self-supervised protein embedding |
| url | https://doi.org/10.1186/s12859-022-05107-w |
| work_keys_str_mv | AT jialinwu improvedcompoundproteininteractionsiteandbindingaffinitypredictionusingselfsupervisedproteinembeddings AT zheliu improvedcompoundproteininteractionsiteandbindingaffinitypredictionusingselfsupervisedproteinembeddings AT xiaofengyang improvedcompoundproteininteractionsiteandbindingaffinitypredictionusingselfsupervisedproteinembeddings AT zhanglinlin improvedcompoundproteininteractionsiteandbindingaffinitypredictionusingselfsupervisedproteinembeddings |