Related Experiment Video
Updated: Jul 17, 2026
18:10
Isolation of Fidelity Variants of RNA Viruses and Characterization of Virus Mutation Frequency
Published on: June 16, 2011
Separation of sequences from host-pathogen interface using triplet nucleotide frequencies.
Jeppe Emmersen1, Stephen Rudd, Hans-Werner Mewes
1Institut for Miljø og Bioteknologi, Aalborg Universitet, Sohngaardsholmsvej 49, 9000 Aalborg, Denmark.
Fungal Genetics and Biology : FG & B
|January 16, 2007
Summary
This study introduces a new method using triplet frequencies to classify gene origins, improving accuracy for both coding and non-coding sequences. The enhanced approach aids in understanding host-pathogen interactions and disease resistance mechanisms.
Area of Science:
- Genomics
- Bioinformatics
- Molecular Biology
Background:
- Identifying genes in host-pathogen interactions is crucial for understanding disease resistance and susceptibility.
- Traditional gene classification relies on sequence similarity, which fails for sequences without known homologues.
- Previous codon frequency methods were limited to coding regions (CDS).
Purpose of the Study:
- To develop a more versatile and accurate method for classifying gene origins from mixed cDNA pools.
- To extend gene classification to include non-coding sequences.
- To improve the prediction accuracy of gene origin classification.
Main Methods:
- Utilized sliding-window triplet frequencies for sequence analysis.
- Applied a Support Vector Machine (SVM) classifier.
- Compared the new triplet frequency method with previous codon frequency approaches.
Main Results:
- The triplet frequency method accurately classifies both coding and non-coding sequences.
- Prediction accuracy of the SVM classifier increased from 95.6+/-0.3% to 96.5+/-0.2%.
- Functional analysis identified gene families with varying classification probabilities.
Conclusions:
- Sliding-window triplet frequencies offer a robust method for classifying gene origins, overcoming limitations of previous techniques.
- The enhanced accuracy aids in a deeper understanding of host-pathogen interactions.
- A publicly available server facilitates the classification of expressed sequence tags (ESTs) using triplet frequencies.
