Naive Prediction of Protein Backbone Phi and Psi Dihedral Angles Using Deep Learning
Protein structure prediction represents a significant challenge in the field of bioinformatics, with the prediction of protein structures using backbone dihedral angles recently achieving significant progress due to the rise of deep neural network research. However, there is a trend in protein struc...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-10-01
|
| Series: | Molecules |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1420-3049/28/20/7046 |
| _version_ | 1797572785090854912 |
|---|---|
| author | Matic Broz Marko Jukič Urban Bren |
| author_facet | Matic Broz Marko Jukič Urban Bren |
| author_sort | Matic Broz |
| collection | DOAJ |
| description | Protein structure prediction represents a significant challenge in the field of bioinformatics, with the prediction of protein structures using backbone dihedral angles recently achieving significant progress due to the rise of deep neural network research. However, there is a trend in protein structure prediction research to employ increasingly complex neural networks and contributions from multiple models. This study, on the other hand, explores how a single model transparently behaves using sequence data only and what can be expected from the predicted angles. To this end, the current paper presents data acquisition, deep learning model definition, and training toward the final protein backbone angle prediction. The method applies a simple fully connected neural network (FCNN) model that takes only the primary structure of the protein with a sliding window of size 21 as input to predict protein backbone ϕ and ψ dihedral angles. Despite its simplicity, the model shows surprising accuracy for the ϕ angle prediction and somewhat lower accuracy for the ψ angle prediction. Moreover, this study demonstrates that protein secondary structure prediction is also possible with simple neural networks that take in only the protein amino-acid residue sequence, but more complex models are required for higher accuracies. |
| first_indexed | 2024-03-10T21:00:49Z |
| format | Article |
| id | doaj.art-916096f831164eb6932fbfbc076979b6 |
| institution | Directory Open Access Journal |
| issn | 1420-3049 |
| language | English |
| last_indexed | 2024-03-10T21:00:49Z |
| publishDate | 2023-10-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Molecules |
| spelling | doaj.art-916096f831164eb6932fbfbc076979b62023-11-19T17:31:58ZengMDPI AGMolecules1420-30492023-10-012820704610.3390/molecules28207046Naive Prediction of Protein Backbone Phi and Psi Dihedral Angles Using Deep LearningMatic Broz0Marko Jukič1Urban Bren2Faculty of Chemistry and Chemical Engineering, University of Maribor, Smetanova ulica 17, SI-2000 Maribor, SloveniaFaculty of Chemistry and Chemical Engineering, University of Maribor, Smetanova ulica 17, SI-2000 Maribor, SloveniaFaculty of Chemistry and Chemical Engineering, University of Maribor, Smetanova ulica 17, SI-2000 Maribor, SloveniaProtein structure prediction represents a significant challenge in the field of bioinformatics, with the prediction of protein structures using backbone dihedral angles recently achieving significant progress due to the rise of deep neural network research. However, there is a trend in protein structure prediction research to employ increasingly complex neural networks and contributions from multiple models. This study, on the other hand, explores how a single model transparently behaves using sequence data only and what can be expected from the predicted angles. To this end, the current paper presents data acquisition, deep learning model definition, and training toward the final protein backbone angle prediction. The method applies a simple fully connected neural network (FCNN) model that takes only the primary structure of the protein with a sliding window of size 21 as input to predict protein backbone ϕ and ψ dihedral angles. Despite its simplicity, the model shows surprising accuracy for the ϕ angle prediction and somewhat lower accuracy for the ψ angle prediction. Moreover, this study demonstrates that protein secondary structure prediction is also possible with simple neural networks that take in only the protein amino-acid residue sequence, but more complex models are required for higher accuracies.https://www.mdpi.com/1420-3049/28/20/7046protein structure predictionbackbone dihedral anglesdeep neural networkfully connected neural network (FCNN)ϕ and ψ angle predictionprotein secondary structure prediction |
| spellingShingle | Matic Broz Marko Jukič Urban Bren Naive Prediction of Protein Backbone Phi and Psi Dihedral Angles Using Deep Learning Molecules protein structure prediction backbone dihedral angles deep neural network fully connected neural network (FCNN) ϕ and ψ angle prediction protein secondary structure prediction |
| title | Naive Prediction of Protein Backbone Phi and Psi Dihedral Angles Using Deep Learning |
| title_full | Naive Prediction of Protein Backbone Phi and Psi Dihedral Angles Using Deep Learning |
| title_fullStr | Naive Prediction of Protein Backbone Phi and Psi Dihedral Angles Using Deep Learning |
| title_full_unstemmed | Naive Prediction of Protein Backbone Phi and Psi Dihedral Angles Using Deep Learning |
| title_short | Naive Prediction of Protein Backbone Phi and Psi Dihedral Angles Using Deep Learning |
| title_sort | naive prediction of protein backbone phi and psi dihedral angles using deep learning |
| topic | protein structure prediction backbone dihedral angles deep neural network fully connected neural network (FCNN) ϕ and ψ angle prediction protein secondary structure prediction |
| url | https://www.mdpi.com/1420-3049/28/20/7046 |
| work_keys_str_mv | AT maticbroz naivepredictionofproteinbackbonephiandpsidihedralanglesusingdeeplearning AT markojukic naivepredictionofproteinbackbonephiandpsidihedralanglesusingdeeplearning AT urbanbren naivepredictionofproteinbackbonephiandpsidihedralanglesusingdeeplearning |