DeepDISE: DNA Binding Site Prediction Using a Deep Learning Method
It is essential for future research to develop a new, reliable prediction method of DNA binding sites because DNA binding sites on DNA-binding proteins provide critical clues about protein function and drug discovery. However, the current prediction methods of DNA binding sites have relatively poor...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-05-01
|
| Series: | International Journal of Molecular Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1422-0067/22/11/5510 |
| _version_ | 1797532967890845696 |
|---|---|
| author | Samuel Godfrey Hendrix Kuan Y. Chang Zeezoo Ryu Zhong-Ru Xie |
| author_facet | Samuel Godfrey Hendrix Kuan Y. Chang Zeezoo Ryu Zhong-Ru Xie |
| author_sort | Samuel Godfrey Hendrix |
| collection | DOAJ |
| description | It is essential for future research to develop a new, reliable prediction method of DNA binding sites because DNA binding sites on DNA-binding proteins provide critical clues about protein function and drug discovery. However, the current prediction methods of DNA binding sites have relatively poor accuracy. Using 3D coordinates and the atom-type of surface protein atom as the input, we trained and tested a deep learning model to predict how likely a voxel on the protein surface is to be a DNA-binding site. Based on three different evaluation datasets, the results show that our model not only outperforms several previous methods on two commonly used datasets, but also demonstrates its robust performance to be consistent among the three datasets. The visualized prediction outcomes show that the binding sites are also mostly located in correct regions. We successfully built a deep learning model to predict the DNA binding sites on target proteins. It demonstrates that 3D protein structures plus atom-type information on protein surfaces can be used to predict the potential binding sites on a protein. This approach should be further extended to develop the binding sites of other important biological molecules. |
| first_indexed | 2024-03-10T11:07:55Z |
| format | Article |
| id | doaj.art-f734cd8456c945f6bbe9bce474025b60 |
| institution | Directory Open Access Journal |
| issn | 1661-6596 1422-0067 |
| language | English |
| last_indexed | 2024-03-10T11:07:55Z |
| publishDate | 2021-05-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | International Journal of Molecular Sciences |
| spelling | doaj.art-f734cd8456c945f6bbe9bce474025b602023-11-21T21:04:01ZengMDPI AGInternational Journal of Molecular Sciences1661-65961422-00672021-05-012211551010.3390/ijms22115510DeepDISE: DNA Binding Site Prediction Using a Deep Learning MethodSamuel Godfrey Hendrix0Kuan Y. Chang1Zeezoo Ryu2Zhong-Ru Xie3Computational Drug Discovery Laboratory, School of Electrical and Computer Engineering, College of Engineering, University of Georgia, Athens, GA 30602, USADepartment of Computer Science and Engineering, National Taiwan Ocean University, Keelung 202, TaiwanComputational Drug Discovery Laboratory, School of Electrical and Computer Engineering, College of Engineering, University of Georgia, Athens, GA 30602, USAComputational Drug Discovery Laboratory, School of Electrical and Computer Engineering, College of Engineering, University of Georgia, Athens, GA 30602, USAIt is essential for future research to develop a new, reliable prediction method of DNA binding sites because DNA binding sites on DNA-binding proteins provide critical clues about protein function and drug discovery. However, the current prediction methods of DNA binding sites have relatively poor accuracy. Using 3D coordinates and the atom-type of surface protein atom as the input, we trained and tested a deep learning model to predict how likely a voxel on the protein surface is to be a DNA-binding site. Based on three different evaluation datasets, the results show that our model not only outperforms several previous methods on two commonly used datasets, but also demonstrates its robust performance to be consistent among the three datasets. The visualized prediction outcomes show that the binding sites are also mostly located in correct regions. We successfully built a deep learning model to predict the DNA binding sites on target proteins. It demonstrates that 3D protein structures plus atom-type information on protein surfaces can be used to predict the potential binding sites on a protein. This approach should be further extended to develop the binding sites of other important biological molecules.https://www.mdpi.com/1422-0067/22/11/5510deep learningprotein–DNA interactionbinding site predictiondrug designconvolutional neural networkproteome |
| spellingShingle | Samuel Godfrey Hendrix Kuan Y. Chang Zeezoo Ryu Zhong-Ru Xie DeepDISE: DNA Binding Site Prediction Using a Deep Learning Method International Journal of Molecular Sciences deep learning protein–DNA interaction binding site prediction drug design convolutional neural network proteome |
| title | DeepDISE: DNA Binding Site Prediction Using a Deep Learning Method |
| title_full | DeepDISE: DNA Binding Site Prediction Using a Deep Learning Method |
| title_fullStr | DeepDISE: DNA Binding Site Prediction Using a Deep Learning Method |
| title_full_unstemmed | DeepDISE: DNA Binding Site Prediction Using a Deep Learning Method |
| title_short | DeepDISE: DNA Binding Site Prediction Using a Deep Learning Method |
| title_sort | deepdise dna binding site prediction using a deep learning method |
| topic | deep learning protein–DNA interaction binding site prediction drug design convolutional neural network proteome |
| url | https://www.mdpi.com/1422-0067/22/11/5510 |
| work_keys_str_mv | AT samuelgodfreyhendrix deepdisednabindingsitepredictionusingadeeplearningmethod AT kuanychang deepdisednabindingsitepredictionusingadeeplearningmethod AT zeezooryu deepdisednabindingsitepredictionusingadeeplearningmethod AT zhongruxie deepdisednabindingsitepredictionusingadeeplearningmethod |