DeepSF: deep convolutional neural network for mapping protein sequences to folds

By Jie Hou, Badri Adhikari, Jianlin Cheng. 2017

228 citations recorded by the source corpus.

Motivation Protein fold recognition is an important problem in structural bioinformatics. Almost all traditional fold recognition methods use sequence (homology) comparison to indirectly predict the fold of a target protein based on the fold of a template protein with known structure, which cannot explain the relationship between sequence and fold. Only a few methods had been developed to classify protein sequences into a small number of folds due to methodological limitations, which are not generally useful in practice. Results We develop a deep 1D‐convolution neural network (DeepSF) to directly classify any protein sequence into one of 1195 known folds, which is useful for both fold recognition and the study of sequence‐structure relationship. Different from traditional sequence alignment (comparison) based methods, our method automatically extracts fold‐related features from a protein sequence of any length and maps it to the fold space. We train and test our method on the datasets curated from SCOP1.75, yielding an average classification accuracy of 75.3%. On the independent testing dataset curated from SCOP2.06, the classification accuracy is 73.0%. We compare our method with a top profile‐profile alignment method—HHSearch on hard template‐based and template‐free modeling targets of CASP9‐12 in terms of fold recognition accuracy. The accuracy of our method is 12.63‐26.32% higher than HHSearch on template‐free modeling targets and 3.39‐17.09% higher on hard template‐based modeling targets for top 1, 5 and 10 predicted folds. The hidden features extracted from sequence by our method is robust against sequence mutation, insertion, deletion and truncation, and can be used for other protein pattern recognition problems such as protein clustering, comparison and ranking. Availability and implementation The DeepSF server is publicly available at: http://iris.rnet.missouri.edu/DeepSF/. Supplementary information Supplementary data are available at Bioinformatics online.

Subfields: Genetics, Molecular Biology, Computational Biology & Bioinformatics

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Genetics

In Molecular Biology

In Computational Biology & Bioinformatics

Citations in the corpus

Evaluating Protein Transfer Learning with TAPE (2019, methodology): “Our evolutionary features for remote homology detection are Position Speciﬁc Scoring Matrices (PSSMs), following the recent work DeepSF [40].”
Profile Prediction: An Alignment-Based Pre-Training Task for Protein Sequence Models (2020, background): “, 2019), contact prediction (AlQuraishi, 2019), remote homology detection (Hou et al., 2018), fluorescence prediction (Sarkisyan et al.”
De Novo Protein Design for Novel Folds using Guided Conditional Wasserstein Generative Adversarial Networks (gcWGAN) (2019, methodology): “To assess the oracle as a fold predictor, we compared, in Table S4, the original DeepSF using all features or sequence only and our oracle (modiﬁedDeepSF)usingsequenceonly.”
Protein Representation Learning by Geometric Structure Pretraining (2022, background): “…Commission number prediction (Gligorijevi ´ c et al., 2021), Gene Ontology term prediction (Gligorijevi ´ c et al., 2021), fold classiﬁcation (Hou et al., 2018) and reaction classiﬁcation (Hermosilla et al., 2021) verify our Gear-Net augmented with edge message passing can consistently…”
Intrinsic-Extrinsic Convolution and Pooling for Learning on 3D Protein Structures (2021, methodology): “Lastly, we compare in the top of Tbl. 1 to the state of the art results reported in the paper (Hou et al., 2018) as to the results obtained using multiple sequence alignment, where we outperform all.”
Self-Supervised Contrastive Learning of Protein Representations By Mutual Information Maximization (2020, background): “The training, validation, and test set splits are curated in Hou et al. (2018); test sets examines three levels of distribution shift from the training dataset.”