A Universal Analysis Pipeline of Hybrid Capture-Based Targeted Sequencing Data with Unique Molecular Indexes (UMIs)

Hybrid capture-based targeted sequencing is increasingly used for genomic variants profiling in tumor patients. Unique molecular index (UMI) technology has recently been developed and helps to increase the accuracy of variant calling by minimizing PCR biases and sequencing errors. However, UMI-adopted targeted sequencing data analysis is slightly different from the methods for other types of omics data, and its pipeline for variant calling is still being optimized in various studying groups for their own purpose. Due to this provincial usage range of tools, our group built an analysis pipeline for global application to many studies of targeted sequencing generated with different methods. First, we generated hybrid capture-based data using genomic DNA extracted from tumor tissues of colorectal cancer patients. Sequencing libraries were prepared, pooled together and an 8-plexed capture library was processed to the enrichment step before 150bp paired-end (PE) sequencing with Illumina Hiseq series. For the analysis, we evaluated several different tools published. We mainly focused on the compatibility of input and output of each tool. Finally, our laboratory built an analysis pipeline specialized for UMI-adopted data. Through this pipeline, we were able to estimate the even on-target rates and filtered consensus reads for more accurate variant calling. These results suggest the potential of our analysis pipeline in the precise examination of quality and efficiency of conducted experiments.