VCF Miner: A Platform and Tools for Collaborative Information Mining from Next-Generation Sequence Variant Data

High-throughput next-generation sequencing (NGS) technology has provided powerful tools for genetics/genomics studies. It has also imposed a great challenge on researchers to determine how to efficiently handle and quickly glean information for genotype/phenotype studies.   Quickly process NGS data for useful information screening is important before the extensive analysis may be carried out, because the latter is usually time consuming and demands large computation power. The variant call format (VCF) that stores variations against a reference genome, is one of such data formats at the heart of information crossroads of the genotype/phenotype data analysis pipelines. We propose a shared platform for collaborative VCF file storage, handling, information abstraction, querying, and data re-use. A unix-file-system-based data file repository has been built, made available Plink/SEQ, BEDOPS, and BEDtools for VCF file handling as well as a number of in-house scripts to serve various purposes during the data process. An initial collection of 15 VCF files from four laboratories were used as a test case. The goals of this project were to allow VCF data from multiple collaborative projects to be managed in a central location to facilitate pre-computed data abstractions and query.  This resource will allow data re-use in subsequent, possibly more diverse types of data analysis in an efficient manner.