Next-generation sequencing to identify candidate gene for hereditary mixed polyposis syndrome.

Abstract

Listen

Translate article

1515 Background: Rare monogenic disorders are often due to mutations that are highly penetrant, extremely rare, and strongly disrupt normal biology. Hereditary mixed polyposis syndrome (HMPS, OMIM ID %601228), is characterized by a mixture of atypical juvenile polyps, hyperplasic polyps, sessile serrated adenomas and an increased risk of colorectal cancer. Polyps appear to be inherited in an autosomal dominant fashion. The putative susceptibility locus initially mapped to 15q13-14, however, the genetic basis of this syndrome is not well understood, and no mutation or associated variants have been identified. Methods: Germline DNA from four individuals from a family with clinically and pathologically confirmed HMPS was subjected to massively parallel sequencing (Illumina GAII). Affected individuals had an average of 30x coverage of their exome and 10x coverage of their genome. Sequence calls were filtered using the criteria of ≥ 4x coverage, quality score over depth ≤ 50, depth of coverage ≤ 360, allele balance ≤ 0.75, number or MAPQ zero reads at locus < 4. Common variants were filtered out by excluding variants found in dbSNP (build131), 1000 Genomes Project, the Exome Variant Server or 80 unaffected controls sequenced by hybrid capture and whole exome sequencing. Polyphen2 and SIFT were used to predict pathogenicity of the novel, shared variants, and validated by Sanger sequencing. Expression levels of novel variants were examined in available tumor samples using qRT-PCR. Results: From the 32 previously unidentified nonsense, missense or splice site variants shared by the family members whose whole genomes were sequenced, only 5 (4 missense, 1 nonsense) were predicted to be damaging, leading to a small subset of novel candidate genes including ZNF426, which may be responsible for HMPS within this family. Conclusions: HMPS extremely difficult to accurately diagnosis and the genetic basis is unknown. Using next-generation sequencing we were able to detect previously unidentified low frequency allelic variants including a novel candidate locus.