DenseCPD: Improving the Accuracy of Neural-Network-Based Computational Protein Sequence Design with DenseNet

By Yifei Qi, J. H. Zhang. 2020

85 citations recorded by the source corpus.

Computational protein design remains a challenging task despite its remarkable success in the past few decades. With the rapid progress of deep-learning techniques and the accumulation of three-dimensional protein structures, using deep neural networks to learn the relationship between protein sequences and structures and then automatically design a protein sequence for a given protein backbone structure is becoming increasingly feasible. In this study, we developed a deep neural network named DenseCPD that considers the three-dimensional density distribution of protein backbone atoms and predicts the probability of 20 natural amino acids for each residue in a protein. The accuracy of DenseCPD was 53.240.17% in a 5-fold cross validation on the training set and 55.53% and 50.71% on two independent test sets, which is more than 10% higher than those of previous state-of-the-art methods. Two approaches for using DenseCPD predictions in computational protein design were analyzed. The approach using the cutoff of accumulative probability had a smaller sequence search space compared to that of the approach that simply uses the top-k predictions and therefore enables higher sequence identity in redesigning three proteins with Rosetta. The network and the data sets are available on a web server at http://protein.org.cn/densecpd.html. The results of this study may benefit the further development of computational protein design methods.

Subfields: Computer Vision and Pattern Recognition, 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 Computer Vision and Pattern Recognition

In Molecular Biology

In Computational Biology & Bioinformatics

Citations in the corpus

Robust deep learning based protein sequence design using ProteinMPNN (2022, positive): “Recently deep learning approaches have shown considerable promise in rapidly generating plausible amino acid sequences given monomeric protein backbones without need for compute intensive explicit consideration of sidechain rotameric states (1-6).”
Learning inverse folding from millions of predicted structures (2022, methodology): “…deep learning methods have also been applied to predict amino acid propensities given the surrounding local structural environments (Anand-Achim et al., 2021; Boomsma & Frellsen, 2017; Shroff et al., 2020; Li et al., 2020; Qi & Zhang, 2020; Zhang et al., 2020; Chen et al., 2019; Wang et al., 2018).”
Learning from Protein Structure with Geometric Vector Perceptrons (2020, neutral): “GVP-GNN is the second best-performing method, behind the CNN method DenseCPD.”
AlphaDesign: A graph protein design method and benchmark on AlphaFoldDB (2022, methodology): “Moreover, DenseCPD (Qi & Zhang, 2020) takes 7 minutes to predict a 120-length protein on their server 1 .”
Using AlphaFold for Rapid and Accurate Fixed Backbone Protein Design (2021, background): “Design tasks tackled with deep learning include ﬁxed backbone design (O’Connell et al., 2018; Ingraham et al., 2019; Qi and Zhang, 2020; Norn et al., 2021), antibody design (Wang et al., 2018; Saka et al., 2021; Shin et al., 2021), de novo design (Anishchenko et al., 2020; Moffat and Jones, 2021),…”