Related Experiment Video
Updated: Jul 5, 2025

Determining the Likelihood of Variant Pathogenicity Using Amino Acid-level Signal-to-Noise Analysis of Genetic Variation
Published on: January 16, 2019
Machine learning-based approach KEVOLVE efficiently identifies SARS-CoV-2 variant-specific genomic signatures
Dylan Lebatteux1, Hugo Soudeyns2,3,4, Isabelle Boucoiran5
1Department of Computer Science, Université du Québec à Montréal, Montréal, Québec, Canada.
KEVOLVE, a machine learning tool, effectively identifies genomic signatures in SARS-CoV-2 variants. These signatures, often biologically relevant, aid in variant classification and characterization without needing sequence alignments.
Area of Science:
- Genomics
- Bioinformatics
- Machine Learning
Background:
- Genomic signatures are key for distinguishing viral species and variants.
- Identifying these signatures aids SARS-CoV-2 research, including variant recognition and functional characterization.
Purpose of the Study:
- To assess KEVOLVE, a machine learning approach, for identifying genomic signatures in SARS-CoV-2.
- To compare KEVOLVE's effectiveness against existing statistical tools.
Main Methods:
- KEVOLVE utilizes a genetic algorithm with a machine learning kernel to find minimal k-mer sets defining genomic signatures.
- KANALYZER, an extension of KEVOLVE, characterizes signature variations across different variant classes.
Main Results:
- KEVOLVE outperformed gold-standard statistical tools in identifying variant-discriminative signatures in a large SARS-CoV-2 dataset.
- Identified signatures frequently correlated with known mutations and their functional/pathological impacts.
Conclusions:
- KEVOLVE is a robust machine learning method for identifying biologically relevant genomic signatures in SARS-CoV-2 variants.
- The approach bypasses the need for multiple sequence alignments, offering a significant advantage.
Related Concept Videos
Evolutionary Relationships through Genome Comparisons
Genetic Variation
Genes exist in different versions called alleles,...
Genomics
Viral Mutations
Next-generation Sequencing
Next-Generation Sequencing Methods
Although all next-generation methods use different technologies, they all share a set of standard features....
Genetic Screens
Forward genetic screens
Forward or “classical” genetic screens involve creating random mutations in an organism’s DNA using radiation, mutagens, or insertion of additional bases, which...

