Related Experiment Video
Updated: Sep 8, 2025
12:11
Simultaneous Affinity Enrichment of Two Post-Translational Modifications for Quantification and Site Localization
Published on: February 27, 2020
7.0K
Multi-metric locality sensitive hashing enhances alignment accuracy of bisulfite sequencing reads: BisHash
Hassan Nikaein1, Ali Sharifi-Zarchi1
1Department of Computer Engineering, Sharif University of Technology, Tehran, 1458889694, Iran.
Bioinformatics Advances
|August 20, 2025
Summary
Multi-Metric Locality-Sensitive Hashing (M2LSH) improves bioinformatics analysis by integrating multiple similarity metrics. This approach enhances accuracy in sequence alignment and DNA methylation analysis, particularly for complex biological data.
Area of Science:
- Bioinformatics
- Computational Biology
- Genomics
Background:
- Locality-Sensitive Hashing (LSH) is crucial for large-scale biological data analysis, including genome assembly and sequence alignment.
- Traditional single-metric LSH struggles with the diverse evolutionary and structural properties of biological data, impacting accuracy.
- Limitations are evident in sequence alignment, variant calling, and functional analysis of complex genomic regions.
Purpose of the Study:
- To introduce Multi-Metric Locality-Sensitive Hashing (M2LSH) for more accurate analysis of complex biological datasets.
- To enhance sequence alignment and similarity detection using Multi-Metric MinHash (M3Hash).
- To demonstrate the application of M2LSH in bisulfite sequencing for DNA methylation analysis via BisHash.
Main Methods:
- Development of Multi-Metric Locality-Sensitive Hashing (M2LSH) by extending LSH with multiple similarity metrics.
- Introduction of Multi-Metric MinHash (M3Hash) to improve sequence alignment and similarity detection capabilities.
- Application of M2LSH in BisHash for bisulfite sequencing data analysis, focusing on DNA methylation.
Main Results:
- M2LSH effectively captures diverse sequence and structural features, improving performance in heterogeneous biological regions.
- BisHash, an M2LSH application, shows superior accuracy in DNA methylation analysis, outperforming traditional methods in challenging scenarios like cancer studies.
- The proposed methods, M2LSH and M3Hash, demonstrate significant potential for advancing bioinformatics research.
Conclusions:
- M2LSH and M3Hash offer a more robust framework for analyzing complex biological data compared to traditional LSH.
- BisHash showcases the practical utility of M2LSH for DNA methylation analysis, providing higher accuracy in critical applications.
- The developed methodologies hold promise for improving various bioinformatics tasks, including sequence alignment and variant analysis.
