2-kupl: mapping-free variant detection from DNA-seq data of matched samples.

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• Genome variant detection is crucial but computationally challenging.
• Existing mapping-based methods struggle with complex genomic regions like repeats and large structural variants.
• Comparing DNA sequencing (DNA-seq) samples requires robust variant identification.

Purpose of the Study:

• To introduce a novel mapping-free protocol for detecting variants between two DNA-seq samples.
• To overcome limitations of traditional mapping-based variant detection methods.
• To improve variant calling accuracy, especially in difficult-to-map genomic regions.

Main Methods:

• Development of 2-kupl, a k-mer based, mapping-free protocol.
• Application of 2-kupl to simulated and actual DNA-seq data.
• Utilizing prostate cancer whole exome sequencing data for validation.

Main Results:

• 2-kupl demonstrates higher accuracy compared to other mapping-free protocols on simulated and real data.
• The protocol successfully identified candidate variants in hard-to-map regions.
• Potential novel recurrent variants were proposed in prostate cancer data.

Conclusions:

• A novel mapping-free protocol, 2-kupl, has been developed for variant calling between matched DNA-seq samples.
• The protocol is effective for variant detection in unmappable genomic regions.
• It offers a viable solution for variant analysis even in the absence of a reference genome.