Codon-aware multi-scale feature fusion for metagenomic sequence classification
- 1School of Mechatronic Engineering and Automation, Shanghai University, Shanghai, People’s Republic of China
- 2Shanghai Key Laboratory of Power Station Automation Technology, Shanghai, People’s Republic of China
Abstract
Microorganisms exert profound influences on both natural ecosystems and human society. As the cornerstone of microbial research, metagenomics relies on the precise analysis of large-scale, multi-source genomic data. However, discriminating among diverse sequence types remains a formidable challenge. To address this, we present CamFi (Codon-aware Multi-scale Feature Fusion), a unified framework integrating overlapping nucleotide-triplet representations with multi-scale dilated convolutions for classifying prokaryotic chromosomes, eukaryotic chromosomes, plasmids, and viruses. Benchmark evaluations demonstrate strong mean F1 scores of 97.70% for eukaryotic chromosomes and 94.13% for plasmids. On the CAMI II marine dataset, CamFi achieved a weighted F1 of 82.65%, ranking second among six methods based on reported aggregate metrics. In a balanced four-class benchmark, CamFi attained a macro-F1 of 93.77%, compared with DeepMicroClass (90.03%) and the XGBoost stage of 4CAC (64.66%). These results establish CamFi as a competitive approach for unified metagenomic contig classification under the evaluated conditions.
Related
- Lossless compression of protein databases for efficient and accurate metagenomic sequence classification with Centrifuger
- Hyperparameter-Tuned K-Nearest Neighbor for Herbal Leaf Classification Using Shape, Color, and Texture Feature Fusion in Smart Agriculture
- BEMD-Based Color–Texture Feature Fusion with Image Sharpening for Multiclass Huanglongbing Severity Classification
- An integrated instance segmentation and deep feature fusion model for robust leaf species classification in natural environments
- Repurposing PeTriBERT for Protein Protein Interaction Prediction with Sequence Structure Fusion
BibTeX
@article{qian2026codonaware,
title = {{Codon-aware multi-scale feature fusion for metagenomic sequence classification}},
author = {Yan Qian and Quanjie Song and Xinyi Xie and Li Deng},
journal = {PLOS One},
year = {2026},
month = oct,
doi = {10.1371/journal.pone.0359090},
url = {https://doi.org/10.1371/journal.pone.0359090},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Codon-aware multi-scale feature fusion for metagenomic sequence classification
AU - Yan Qian
AU - Quanjie Song
AU - Xinyi Xie
AU - Li Deng
PY - 2026
DA - 2026/10/05
JO - PLOS One
DO - 10.1371/journal.pone.0359090
UR - https://doi.org/10.1371/journal.pone.0359090
AB - Microorganisms exert profound influences on both natural ecosystems and human society. As the cornerstone of microbial research, metagenomics relies on the precise analysis of large-scale, multi-source genomic data. However, discriminating among diverse sequence types remains a formidable challenge. To address this, we present CamFi (Codon-aware Multi-scale Feature Fusion), a unified framework integrating overlapping nucleotide-triplet representations with multi-scale dilated convolutions for classifying prokaryotic chromosomes, eukaryotic chromosomes, plasmids, and viruses. Benchmark evaluations demonstrate strong mean F1 scores of 97.70% for eukaryotic chromosomes and 94.13% for plasmids. On the CAMI II marine dataset, CamFi achieved a weighted F1 of 82.65%, ranking second among six methods based on reported aggregate metrics. In a balanced four-class benchmark, CamFi attained a macro-F1 of 93.77%, compared with DeepMicroClass (90.03%) and the XGBoost stage of 4CAC (64.66%). These results establish CamFi as a competitive approach for unified metagenomic contig classification under the evaluated conditions.
SN - 1932-6203
ER -