High throughput germline mutation screening for hereditary breast cancer using fluidigm access arrary & 454 GS junior

Abstract

INTRODUCTION: Breast cancer is prevalent world-wide and has become the leading cause of cancer among the female population in Hong Kong. The Hong Kong Hereditary Breast Cancer Family Registry was established in 2007 and thus far 1194 patients with breast cancer were recruited from hospitals and clinics in Hong Kong. Mutational screening of susceptibility genes, BRCA1 and BRCA2, using Sanger full gene sequencing was performed and until now 706 cases have been analyzed. We sought to apply next-generation DNA sequencing to analyze the remaining samples at higher throughput. METHODS: We analyzed BRCA1, BRCA2 and also TP53 genes of 192 cases using 454 GS Junior System. Generation of barcoded amplicon libraries was streamlined by microfluidic PCR using Fluidigm Access Array System. Two Fluidigm runs can generate 41 BRAC1 and TP53 amplicons and 45 BRCA2 amplicons from each of 48 unique cases, and a total of 8 runs were performed. Each Fluidigm run was subject to one GS Junior run. Sequencing data (total length 213 Mbp) were automatically analyzed by an in-house developed bioinformatics pipeline, which mainly consists of GS Amplicon Variant Analyzer, SAMtools and Ensembl Variant Effect Predictor. To confirm the detected mutations, full BRCA1 and BRCA2 genes of 10 cases and individual deleterious mutations of remaining cases were subject to Sanger sequencing. RESULTS: On-going analysis revealed 3 novel deleterious mutations (1 BRCA1 frameshift insertion and 2 BRCA2 non-sense SNPs) among BRCA1, BRCA2 and TP53 mutations identified from 192 cases. Sensitivity of 454 sequencing approach was found to be comparable to Sanger sequencing, although detection of insertiondeletion polymorphisms around long homopolymers is challenging. Bioinformaticsbased filters were optimized based on Sanger sequencing data and were applied to minimize false positives. Signal intensity and the sequence of cyclic nucleotide flows were used in mutation prioritization to account for Carry Forward and Incomplete Extension errors around homopolymer sequences. CONCLUSIONS: Mutational screening of BRCA1, BRCA2, and TP53 genes from 192 cases were streamlined and performed at much higher throughput. The bioinformatics pipeline developed is expected to facilitate data analysis of similar diagnostic settings. In Hong Kong, our laboratory pioneered the application of both microfluidic PCR and next-generation sequencing systems for molecular diagnostics.