h5vc: scalable nucleotide tallies with HDF5

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• Genome sequencing applications are expanding, necessitating scalable analysis tools for large datasets.
• Current analysis pipelines often rely on BAM and VCF formats, which can create significant overhead for complex analyses.
• Efficient data management is crucial for handling the increasing volume and depth of genomic data.

Purpose of the Study:

• To introduce h5vc, a novel tally data structure and associated software for managing genomic data.
• To provide a more efficient and accessible approach for analyzing large-scale genome sequencing data.
• To facilitate various genomics analyses, including variant calling and copy-number estimation.

Main Methods:

• Developed h5vc, a tally data structure (nucleotides × samples × strands × genomic positions).
• Implemented software for creating tallies from BAM files using the HDF5 system.
• Integrated functionalities for data visualization, quality assessment, and statistical computation.

Main Results:

• h5vc offers a scalable solution for managing and analyzing large genomic datasets.
• The tally structure simplifies data management compared to traditional BAM/VCF workflows.
• The software provides tools for variant calling, copy-number estimation, and mutation spectrum analysis.

Conclusions:

• h5vc provides an efficient intermediate data abstraction for diverse genomics analyses.
• The software enhances scalability and accessibility for researchers working with large-scale genome sequencing data.
• h5vc aims to lower the barrier for low-level analysis of extensive genomic datasets.