A Pangenomic-enabled high-capacity haplotyping objective – PANHCHO

In this project we propose to perform intensely thorough genome sequencing on 20 individuals from existing genomics initiatives in France to create a ‘pangenome’; a reference sequence with a graph structure allowing uniquely for complex genetic rearrangements to be described. We then propose to develop new statistical methods to leverage this pangenome to augment existing databases through re-analysis of 1000s of sequences and to enable statistical imputation of complex genetic variation. The new reference data and new methods will be of use in both clinical and research applications. In France, two whole-genome sequencing platforms (Auragene and Seqoia) currently process an average of 9,000 case-parent trios per year in the context of rare genetic disease diagnosis. Diagnostic yield is strong (about 30%) but not exhaustive. Concurrently, genetic epidemiologists across France design experiments involving the genotyping or sequencing of 100s if not 1000s of individuals; involving new-patient sample collection and population controls. Data-quality and completeness are key, but technical and methodological advancements in genome sequencing are rapid and France is currently somewhat behind the curve. Hence the potentially huge scope for the utility of the project where the simple goal is to provide reference data, methods, and bioinformatic infrastructure to enable a greater number of interesting genetic variants to be reliably sequenced in France. Using a pangenome can give a 100% increase in structural-variant detection as well as decreasing overall error rates for variant discovery by 34% and an overall increase of 9% of total genomic regions covered including an additional 2000 genes. The novel imputation methods we propose to develop here will crucially democratise such state-of-the-art technology.