SevenBridges Genomics: Genome graph references

Hear how SevenBridges are using genome graph references to encode multiple potential haplotypes for genomic regions at the population scale About this Event: For our last event of the year, we are excited to present Dr Richard Brown from SevenBridges to discuss their genome graph reference technologies used at the population scale. Abstract Reference genomes are an important component in genetic analyses, providing a standard for representing and comparing genomic features, as well as providing a scaffold for mapping sequencing reads in order to recover the sample genotype. The latter process is the first step in all clinical and research studies, but it is well known that it suffers from the problem of reference bias, wherein sequencing data generated from samples that evolutionarily diverged from the reference genome (i.e. non-Europeans) maps less accurately against the reference scaffold and thus leads to significantly higher undetected and artefact genotypes. Genome graph references eliminate this problem by encoding multiple potential haplotypes for any given genomic region. Seven Bridges has consistently driven the development of these next-generation references, and developed analysis pipelines to construct and utilise graph references for true population-scale data. Our team is now extending the scope of genome graph references to encode information other than DNA sequences enabling more nuanced applications that are able to combine the likely genotypes inferred for the sequencing data with the previously determined biological information to generate new insights. We will discuss this capability and present one such application, family graph references improved detection of de novo mutations in family trios by incorporating information about the phase of variant calls in individual samples.