MP3vec: A Reusable Machine-Constructed Feature Representation for Protein Sequences

Sanket Rajan Gupte; Dharm Skandh Jain; Ashwin Srinivasan; Raviprasad Aduri

doi:10.1109/bibm49941.2020.9313301

MP3vec: A Reusable Machine-Constructed Feature Representation for Protein Sequences

Abstrakt

—Machine Learning (ML) methods have been used with varying degrees of success on protein prediction tasks, with two inherent limitations. First, prediction performance often depends upon the features extracted from the proteins. Second, experimental data may be insufficient to construct reliable ML models. Here we introduce MP3vec, a transferable representation for protein sequences that is designed to be used specifically for sequence-to-sequence learning tasks. We use transfer learning to generate the MP3vecs by training a deep neural network on the source problem of protein secondary structure prediction, and then extracting representations learned by the trained network for use in related downstream prediction tasks. ML methods using MP3vecs perform as well as the state-of-the-art (or better) on the target problems, while being orders of magnitude faster in terms of training time. We suggest that MP3vec can act as a strong baseline for comparative work on the use of ML in protein-prediction tasks; and for future extensions with domainspecific features.

Cytowania

0

CrossRef
0

Web of Science
0

Scopus

Autorzy (4)

Sanket Rajan Gupte
Dharm Skandh Jain
Ashwin Srinivasan
Raviprasad Aduri

Cytuj jako

Pełna treść

pełna treść publikacji nie jest dostępna w portalu

pełna treść artykułu zobacz w serwisie zewnętrznym otwiera się w nowej karcie

Słowa kluczowe

—TRANSFER LEARNING, DEEP LEARNING, PROTEIN STRUCTURE, FUNCTION, AND INTERACTION PREDICTION, SEQUENCE MODELLING

Informacje szczegółowe

Kategoria:: Aktywność konferencyjna
Typ:: publikacja w wydawnictwie zbiorowym recenzowanym (także w materiałach konferencyjnych)
Tytuł wydania:: 2020 IEEE International Conference on Bioinformatics and Biomedicine (BIBM) strony 421 - 425
Język:: angielski
Rok wydania:: 2020
Opis bibliograficzny:: Gupte S. R., Jain D. S., Srinivasan A., Aduri R.: MP3vec: A Reusable Machine-Constructed Feature Representation for Protein Sequences// 2020 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)/ : , 2020, s.421-425
DOI:: Cyfrowy identyfikator dokumentu elektronicznego (otwiera się w nowej karcie) 10.1109/bibm49941.2020.9313301
Weryfikacja:: Politechnika Gdańska