May 6 – 9, 2025
Abbaye de Royaumont, Asnières-sur-Oise, France
Europe/Paris timezone

VARIANT CALLING AND HAPLOTYPE RESOLUTION IN AMPLIFIED HAPLOTYPE MIXTURES

Not scheduled
20m
Abbaye de Royaumont, Asnières-sur-Oise, France

Abbaye de Royaumont, Asnières-sur-Oise, France

Abbaye de Royaumont, 95270 Asnières-sur-Oise, France
Poster Genomics & bioinformatics Virtual posters

Speaker

Zoe Vance (University of Edinburgh)

Description

Genomic surveillance of pathogens has become a critical element of modern public health. A key step in most surveillance pipelines is amplification of a target pathogen or locus through PCR, particularly in samples where higher sensitivity is desirable. Despite widespread usage of amplicon sequencing, the potential for errors introduced during amplification to mislead an analysis is underappreciated. PCR induced recombination occurs at high rates with increasing cycle number in mixed haplotype samples and can both inflate apparent diversity within a mixed sample and obscure true rare diversity. Notably this issue will only grow with increasing use of long read technologies where a single read may contain multiple breakpoints. Several tools exist for the detection and removal of chimeric reads resulting from PCR recombination, however this can lead to a large proportion of data being lost which is wasteful and likely to impact sensitivity for lower frequency haplotypes. To our knowledge, no current tools utilise all data in the estimation of haplotypes in an amplified genome mixture. The higher error rate of long reads also creates issues upstream of haplotype calling as it is challenging to call variant sites at low frequencies; most existing long read variant callers are limited to haploid or diploid calls. Additionally, many rely on variant linkage to determine true variants which is not a reliable information source in samples with extensive PCR recombination. Here we present a new pipeline for haplotyping of amplified samples sequenced with long read technologies that attempts to determine variant positions without ploidy or linkage assumptions and estimate the most likely haplotype set in the presence of PCR recombination, with performance demonstrated on highly amplified test mixtures of poliovirus serotypes. We anticipate that this will be a valuable tool for resolving variation in samples where amplification is required without loss of data.

Expedited Notification No thanks, I do not require Expedited Notification

Primary author

Zoe Vance (University of Edinburgh)

Co-authors

Dr Aine O'Toole (University of Edinburgh) Dr Rachel Colquhoun (University of Edinburgh) Prof. Andrew Rambaut (University of Edinburgh)

Presentation materials

There are no materials yet.