cvlr: finding heterogeneously methylated genomic regions using ONT reads.

Area of Science:

• Genomics
• Epigenetics
• Bioinformatics

Background:

• Nanopore sequencing provides rich data on DNA methylation at cytosine bases within CpG dinucleotides.
• Long Nanopore reads facilitate the analysis of multi-locus methylation patterns.

Purpose of the Study:

• To develop and apply a clustering algorithm for identifying subpopulations of molecules based on heterogeneous methylation patterns.
• To analyze imprinted genes and scan chromosome 15 for regions exhibiting variable methylation.

Main Methods:

• Utilizing long Nanopore reads to capture multi-locus methylation information.
• Implementing a clustering algorithm to define molecular subpopulations based on methylation profiles.
• Computing methylation covariance across identified regions, accounting for read mixture.

Main Results:

• Demonstrated the application of the clustering algorithm on known imprinted genes.
• Successfully scanned chromosome 15 to identify regions with heterogeneous methylation patterns.
• Developed software capable of computing methylation covariance in mixed populations.

Conclusions:

• The developed method effectively identifies subpopulations based on methylation patterns in Nanopore data.
• This approach is valuable for studying epigenetic heterogeneity, particularly in imprinted genes and specific genomic regions.
• The software provides tools for advanced analysis of DNA methylation covariance.