Redundancy-Weighting the PDB for Detailed Secondary Structure Prediction Using Deep-Learning Models.
Researchers
Journal
Modalities
Models
Abstract
The Protein Data Bank (PDB), the ultimate source for data in structural biology, is inherently imbalanced. To alleviate biases, virtually all structural biology studies use non-redundant subsets of the PDB, which include only a fraction of the available data. An alternative approach, dubbed redundancy-weighting, down-weights redundant entries rather than discarding them. This approach may be particularly helpful for Machine Learning (ML) methods that use the PDB as their source for data.Methods for Secondary Structure Prediction (SSP) have greatly improved over the years with recent studies achieving above 70% accuracy for 8-class (DSSP) prediction. As these methods typically incorporate machine learning techniques, training on redundancy-weighted datasets might improve accuracy, as well as pave the way toward larger and more informative secondary structure alphabets.
This article compares the SSP performances of Deep Learning (DL) models trained on either redundancy-weighted or non-redundant datasets. We show that training on redundancy-weighted sets consistently results in better prediction of 3-class (HCE), 8-class (DSSP) and 13-class (STR2) secondary structures.
Data and DL models are available in http://meshi1.cs.bgu.ac.il/rw.
© The Author(s) (2020). Published by Oxford University Press. All rights reserved. For Permissions, please email: [email protected].