DEEPCON: Protein Contact Prediction using Dilated Convolutional Neural Networks with Dropout
By Badri Adhikari. 2019
47 citations recorded by the source corpus.
Background Exciting new opportunities have arisen to solve the protein contact prediction problem from the progress in neural networks and the availability of a large number of homologous sequences through high-throughput sequencing. In this work, we study how deep convolutional neural network methods (ConvNets) may be best designed and developed to solve this long-standing problem. Method With publicly available datasets, we designed and trained various ConvNet architectures. We tested several recent deep learning techniques including wide residual networks, dropouts, and dilated convolutions. We studied the improvements in the precision of medium-range and long-range contacts, and compared the performance of our best architectures with the ones used in existing state-of-the-art methods. Results The proposed ConvNet architectures predict contacts with significantly more precision than the architectures used in several state-of-the-art methods. When trained using the DeepCov dataset consisting of 3,456 proteins and tested on PSICOV dataset of 150 proteins, our architectures achieve up to 15% higher precision when L/2 long-range contacts are evaluated. Similarly, when trained using the DNCON2 dataset consisting of 1,426 proteins and tested on 84 protein domains in the CASP12 dataset, our single network achieves 4.8% higher precision than the ensembled DNCON2 method when top L long-range contacts are evaluated. DEEPCON will be made publicly available at https://github.com/badriadhikari/DEEPCON/.
Subfields: Information Systems, 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 Information Systems
- Alexander Rives
- Robert Verkuil
- Sergey Ovchinnikov
- Pieter Abbeel
- John F. Canny
- Krzysztof Fidelis
- Jason Liu
- Tom Sercu
In Molecular Biology
- Andriy Kryshtafovych
- Yang Zhang
- Torsten Schwede
- Krzysztof Fidelis
- John Moult
- Sergey Ovchinnikov
- David Baker
- Bohdan Monastyrskyy
In Computational Biology & Bioinformatics
- Andriy Kryshtafovych
- Torsten Schwede
- Yang Zhang
- Krzysztof Fidelis
- David Baker
- John Moult
- Sergey Ovchinnikov
- Jinbo Xu
Citations in the corpus
- MSA Transformer (2021, methodology): “Initial models used coevolutionary features (Wang et al., 2017; Liu et al., 2018; Yang et al., 2019; Senior et al., 2019; Adhikari & Elofsson, 2020).”
- Critical Assessment of Methods of Protein Structure Prediction (CASP) – Round XIII (2019, background): “There are many potential variations on the type of residual networks currently being deployed, as well as other variables that have yet to be evaluated, such as the best use of dilation and dropout.(22) This and other aspects of the methods will likely be further developed and refined by the next CA”
- Transformer protein language models are unsupervised structure learners (2020, methodology): “Supervised contact prediction Recently, supervised methods using deep learning have resulted in breakthrough results in supervised contact prediction (Wang et al., 2017; Jones & Kandathil, 2018; Yang et al., 2019; Senior et al., 2020; Adhikari & Elofsson, 2020).”
- Single Layers of Attention Suffice to Predict Protein Contacts (2020, background): “Deep Learning has greatly advanced state of the art for supervised contact prediction (Wang et al., 2017; Jones & Kan-dathil, 2018; Senior et al., 2020; Adhikari, 2020).”