Classification of SARS-CoV-2 Variants through The Epistatical Circos Plots with Convolutional Neural Networks

2026/01/30 by Bo Jing, Kai-Rui Zhang, Hong-Li Zeng +1 · 1 voice
Biochemistry, Genetics and Molecular Biology · #q-bio.GN

paper · pdf · doi:10.1007/s00239-026-10329-0

arxiv published 2026/01/30 · arxiv updated 2026/05/06

Abstract

The COVID-19 pandemic has profoundly affected global health, driven by the remarkable transmissibility and mutational adaptability of the SARS-CoV-2 virus. Although five variants of concern, Alpha, Beta, Gamma, Delta, and Omicron, have been identified, the classification task in this study is formulated using four classes: Alpha, Delta, Omicron, and Else, reflecting the sequence availability and temporal coverage of the dataset. Here, we develop an integrative framework that combines direct coupling analysis (DCA), Circos-based visualization, and convolutional neural networks (CNNs) to characterize lineage-specific epistatic signatures from large-scale SARS-CoV-2 genomic sequences. DCA-inferred pairwise mutational couplings were transformed into Circos images, which were then used as inputs for CNN-based classification models. The proposed framework achieved robust variant classification, with the best-performing model reaching a weighted-average F1-score of 98.68± 0.75% and an AUC close to 1.

Citations

• Two fitness inference schemes compared using allele frequencies from 1,068,391 sequences sampled in the UK during the COVID-19 pandemic
• Accurate proteome-wide missense variant effect prediction with AlphaMissense
• Association between SARS-CoV-2 and metagenomic content of samples from the Huanan Seafood Market
• Large language models generate functional protein sequences across diverse families
• Evolutionary-scale prediction of atomic-level protein structure with a language model
• ColabFold: making protein folding accessible to all
• Temporal epistasis inference from more than 3,500,000 SARS-CoV-2 Genomic Sequences
• Highly accurate protein structure prediction with AlphaFold
• SARS-CoV-2 variants, spike mutations and immune escape
• Statistical Genetics in and out of Quasi-Linkage Equilibrium (Extended)
• A Survey of Convolutional Neural Networks: Analysis, Applications, and Prospects
• Fundamental tenis : resep meraih kemenangan / Tony Mottram
• A new coronavirus associated with human respiratory disease in China
• Inferring genetic fitness from genomic data
• Improved protein structure prediction using potentials from deep learning
• EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks
• Searching for MobileNetV3
• DCA for genome-wide epistasis analysis: the statistical genetics perspective
• MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications
• Inverse Statistical Physics of Protein Sequences: A Key Issues Review
• Inverse statistical problems: from the inverse Ising problem to data science
• Densely Connected Convolutional Networks
• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Sequence co-evolution gives 3D contacts and structures of protein complexes
• Fast pseudolikelihood maximization for direct-coupling analysis of protein structure from many homologous amino-acid sequences
• ImageNet classification with deep convolutional neural networks
• Improved contact prediction in proteins: Using pseudolikelihoods to infer Potts models
• Emergence of clones in sexual populations
• Direct-coupling analysis of residue co-evolution captures native contacts across many protein families
• Statistical Genetics and Evolution of Quantitative Traits
• Inverse Ising inference using all the data
• High-dimensional Ising model selection using ℓ₁-regularized logistic regression
• Competition between recombination and epistasis can cause a transition from allele to genotype selection
• Identification of direct residue contacts in protein-protein interaction by message passing
• Statistical Analysis of Non-Lattice Data

Discussions

• Bo Jing, Yu-Han Huang, Hong-Li Zeng, Erik Aurell: Classification of SARS-CoV-2 Variants through The Epistatical Circos Plots with Convolutional Neural Networks https://arxiv.org/abs/2601.22866 https:/ [bsky, 0 points, 0 comments]

Related

• Unsupervised identification of significant lineages of SARS-CoV-2 through scalable machine learning methods
• Explainable convolutional neural network model provides an alternative genome-wide association perspective on mutations in SARS-CoV-2
• Deep learning forward and reverse primer design to detect SARS-CoV-2 emerging variants
• Using Deep Learning Sequence Models to Identify SARS-CoV-2 Divergence
• Effective and scalable clustering of SARS-CoV-2 sequences