nPoRe: n-polymer realigner for improved pileup-based variant calling

Area of Science:

• Genomics
• Bioinformatics

Background:

• Nanopore sequencing offers rapid genomic data generation.
• Current nanopore basecalling has high accuracy for single nucleotide polymorphisms (SNPs) but struggles with insertions and deletions (INDELs).
• INDEL recall remains below 80% for standard R9.4.1 flow cells, limiting comprehensive germline variant analysis.

Purpose of the Study:

• To improve the accuracy of germline INDEL variant calling from nanopore sequencing data.
• To develop enhanced alignment algorithms for better detection of INDELs, particularly in repetitive regions.

Main Methods:

• Extension of the Needleman-Wunsch affine gap alignment algorithm.
• Introduction of novel gap penalties specifically designed for homopolymers and tandem repeats.
• Application of haplotype phasing and realignment techniques to nanopore sequencing reads.

Main Results:

• Haplotype phasing improved INDEL recall from 63.76% to [Formula: see text] at consistent precision.
• Further improvement in INDEL recall to [Formula: see text] was achieved using nPoRe realignment.
• The enhanced alignment demonstrated superior accuracy in regions with repetitive sequences.

Conclusions:

• Read phasing and realignment significantly enhance INDEL recall in nanopore sequencing data.
• Modified gap penalties in alignment are effective for accurately calling INDELs in homopolymers and tandem repeats.
• These advancements address a key limitation in nanopore-based germline variant calling, improving genomic analysis.