Related Experiment Video
Updated: Apr 1, 2026
08:27
Author Spotlight: Efficient Image Recognition Using Directional Gradient Histogram Technique and Support Vector Machines
Published on: January 5, 2024
1.8K
SAMSVM: A tool for misalignment filtration of SAM-format sequences with support vector machine.
Jianfeng Yang1, Xiaofan Ding1, Xing Sun1
11 Division of Life Science, Applied Genomics Centre and Centre for Statistical Science, Hong Kong University of Science and Technology, Clear Water Bay, Hong Kong, P. R. China.
Journal of Bioinformatics and Computational Biology
|October 1, 2015
Summary
The SAMSVM package accurately filters misaligned sequencing reads using support vector machines (SVMs). This bioinformatics tool improves variant calling accuracy by removing problematic sequence alignment/map (SAM) data.
Area of Science:
- Bioinformatics
- Computational Biology
- Genomics
Background:
- Massive parallel sequencing generates Sequence Alignment/Map (SAM) data crucial for bioinformatics.
- Misaligned reads in SAM data are a significant challenge, leading to false positives in variant calling.
- Effective filtration of misaligned reads is essential for accurate sequencing data analysis.
Purpose of the Study:
- To develop a novel package, SAMSVM, for efficient misalignment filtration of SAM-formatted sequences.
- To leverage support vector machine (SVM) algorithms for classifying and excluding misaligned reads.
- To enhance the reliability of variant calling in next-generation sequencing data.
Main Methods:
- Developed the SAMSVM package utilizing LIBSVM tools for support vector classification.
- Treated multiple features of SAM-formatted sequences as vectors in a multi-dimensional space.
- Applied cross-validation on simulated datasets with varying mutation rates (0.001 to 0.1).
Main Results:
- SAMSVM achieved high accuracies (0.89–0.97) and F-scores (0.77–0.94) in detecting misalignments across 14 simulated dataset groups.
- The developed model demonstrated robust performance in identifying and filtering misaligned reads.
- Successful application of SAMSVM to real sequencing data confirmed its effectiveness in improving variant calling.
Conclusions:
- SAMSVM provides an accurate and effective method for misalignment filtration in SAM-formatted sequencing data.
- The package enhances the quality of bioinformatics analyses by ensuring the exclusion of erroneous sequence reads.
- SAMSVM contributes to more reliable variant calling, advancing genomic data interpretation.
