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

Basit Öğe Kaydı
dc.authoridgormez, yasin/0000-0001-8276-2030
dc.contributor.authorGormez, Yasin
dc.contributor.authorAydin, Zafer
dc.date.accessioned2024-10-26T18:06:00Z
dc.date.available2024-10-26T18:06:00Z
dc.date.issued2023
dc.departmentSivas Cumhuriyet Üniversitesi
dc.description.abstractProtein 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.
dc.identifier.doi10.1109/TCBB.2022.3191395
dc.identifier.endpage1113
dc.identifier.issn1545-5963
dc.identifier.issn1557-9964
dc.identifier.issue2
dc.identifier.pmid35849663
dc.identifier.scopus2-s2.0-85135223840
dc.identifier.scopusqualityQ2
dc.identifier.startpage1104
dc.identifier.urihttps://doi.org/10.1109/TCBB.2022.3191395
dc.identifier.urihttps://hdl.handle.net/20.500.12418/29284
dc.identifier.volume20
dc.identifier.wosWOS:000965674700029
dc.identifier.wosqualityQ1
dc.indekslendigikaynakWeb of Science
dc.indekslendigikaynakScopus
dc.indekslendigikaynakPubMed
dc.language.isoen
dc.publisherIEEE Computer Soc
dc.relation.ispartofIeee-Acm Transactions on Computational Biology and Bioinformatics
dc.relation.publicationcategoryMakale - Uluslararası Hakemli Dergi - Kurum Öğretim Elemanı
dc.rightsinfo:eu-repo/semantics/closedAccess
dc.subjectProteins
dc.subjectPredictive models
dc.subjectDeep learning
dc.subjectSolvents
dc.subjectAmino acids
dc.subjectRecurrent neural networks
dc.subjectFeature extraction
dc.subjectFeature extraction or construction
dc.subjectmachine learning
dc.subjectprotein structure predicition
dc.subjectbioinformatics
dc.subjectdeep learning
dc.titleIGPRED-MultiTask: A Deep Learning Model to Predict Protein Secondary Structure, Torsion Angles and Solvent Accessibility
dc.typeArticle

Dosyalar

Koleksiyon

WoS İndeksli Yayınlar Koleksiyonu
PubMed İndeksli Yayınlar Koleksyonu
Scopus İndeksli Yayınlar Koleksiyonu