IGPRED-MultiTask: A Deep Learning Model to Predict Protein Secondary Structure, Torsion Angles and Solvent Accessibility

GÖRMEZ Y., AYDIN Z.

IEEE/ACM Transactions on Computational Biology and Bioinformatics, cilt.20, sa.2, ss.1104-1113, 2023 (SCI-Expanded) identifier identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 20 Sayı: 2
  • Basım Tarihi: 2023
  • Doi Numarası: 10.1109/tcbb.2022.3191395
  • Dergi Adı: IEEE/ACM Transactions on Computational Biology and Bioinformatics
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Aerospace Database, BIOSIS, Biotechnology Research Abstracts, Communication Abstracts, Compendex, EMBASE, INSPEC, MEDLINE, Metadex, Civil Engineering Abstracts
  • Sayfa Sayıları: ss.1104-1113
  • Anahtar Kelimeler: bioinformatics, deep learning, Feature extraction or construction, machine learning, protein structure predicition
  • Abdullah Gül Üniversitesi Adresli: Evet

Özet

Protein secondary structure, solvent accessibility and torsion angle predictions are preliminary steps to predict 3D structure of a protein. Deep learning approaches have achieved significant improvements in predicting various features of protein structure. In this study, IGPRED-Multitask, a deep learning model with multi task learning architecture based on deep inception network, graph convolutional network and a bidirectional long short-Term memory is proposed. Moreover, hyper-parameters of the model are fine-Tuned using Bayesian optimization, which is faster and more effective than grid search. The same benchmark test data sets as in the OPUS-TASS paper including TEST2016, TEST2018, CASP12, CASP13, CASPFM, HARD68, CAMEO93, CAMEO93_HARD, as well as the train and validation sets, are used for fair comparison with the literature. Statistically significant improvements are observed in secondary structure prediction on 4 datasets, in phi angle prediction on 2 datasets and in psi angel prediction on 3 datasets compared to the state-of-The-Art methods. For solvent accessibility prediction, TEST2016 and TEST2018 datasets are used only to assess the performance of the proposed model.