A signal processing-inspired feature extraction framework known as SMFFX can accurately classify viral genomic sequences across 30 types of viruses, achieving accuracy and F1-scores above 0.99 in binary testing. Published in Scientific Reports, the study evaluated the computational framework on a dataset of 173,228 genetic samples. The evaluation assessed the predictive performance of the approach through both binary and multi-class classification tasks to determine how effectively it identifies distinct virus categories.
The investigation addressed the increasing rate of coronavirus and other viral diseases, which have attracted significant scientific attention in recent years. Within this context, viral data can be analyzed at the intracellular level in the form of genomic sequences. To examine these genetic sequences, the researchers introduced a signal processing approach applied directly to signals corresponding to viral genomic sequences.
For the feature extraction process, the framework applies the Mel-frequency cepstral coefficients algorithm, a method based on windowing techniques that enables the examination of genetic samples. From the resulting feature space, the researchers extracted various statistical features. This process creates a structured representation that enables classification algorithms to identify and distinguish among the different virus types with high efficiency.
The framework underwent evaluation across 173,228 samples representing 30 virus types, demonstrating high predictive capability across multiple experimental setups. In binary classification tests using a decision tree classifier, the proposed approach achieved an accuracy and an F1-score of more than 0.99. In multi-class classification testing across the 30 virus types, the method achieved an accuracy of 0.99 and an F1-score exceeding 0.98. In addition, an analysis of one-versus-rest classes showed that most individual viral classes attained F1-scores and area under the curve values greater than 99 percent.
According to authors Amin Khodaei, Yalda Mahmoudi, and Aida Shahriardoust, the high classification performance indicates the reliable predictive capability of the proposed approach. Furthermore, the team conducted a feature-importance analysis to evaluate the method, concluding that the framework provides potential applicability to similar classification tasks involving other viruses or viral variants.
Subject of Research: Viral genomic sequence classification using signal processing feature extraction
Article Title: SMFFX: a signal processing-inspired feature extraction framework for viral genomic sequences classification
Article References: Khodaei, A., Mahmoudi, Y., & Shahriardoust, A. (2026). SMFFX: a signal processing-inspired feature extraction framework for viral genomic sequences classification. Scientific Reports. https://doi.org/10.1038/s41598-026-78480-8
Image Credits: AI Generated
DOI: 10.1038/s41598-026-78480-8
Keywords: signal processing, viral genomic sequences, feature extraction, SMFFX framework, Mel-frequency cepstral coefficients, MFCC algorithm, windowing techniques, statistical features, decision tree classifier, binary classification, multi-class classification, one-versus-rest analysis, virus classification, intracellular viral data, predictive capability, feature-importance analysis, viral variants
Cite Scienmag News
Kristina Jarvis. (November 19, 2026). Signal Processing Method Accurately Classifies Viral Genomic Sequences Across Thirty Types. Scienmag. https://scienmag.com/signal-processing-method-accurately-classifies-viral-genomic-sequences-across-thirty-types/
Kristina Jarvis. "Signal Processing Method Accurately Classifies Viral Genomic Sequences Across Thirty Types." Scienmag, 19 November 2026, https://scienmag.com/signal-processing-method-accurately-classifies-viral-genomic-sequences-across-thirty-types/. Accessed 19 November 2026.
Kristina Jarvis. "Signal Processing Method Accurately Classifies Viral Genomic Sequences Across Thirty Types." Scienmag. November 19, 2026. https://scienmag.com/signal-processing-method-accurately-classifies-viral-genomic-sequences-across-thirty-types/








