Development of Gold Standard Next Generation Sequencing Data Sets

Specific Aims This research plan proposes that the development of “gold standard” data sets for various Next Generation Sequencing (NGS) studies will allow for efficient testing and benchmarking of new bioinformatics tools, algorithms, and emerging computational platforms. This project represents a first step in building NGS infrastructure for researchers and clinicians at the University of Kentucky. Aim 1. Collect and evaluate biomedical data sets Well-studied and characterized data sets will be collected for different NGS use cases, such as resequencing, RNA-seq, CHiP-seq, structural variation analysis, de novo sequencing, and metagenomics. All data sets will be of biomedical interest. Aim 2. Collect and evaluate analysis tools and methods A compressive list of the most current and widely used analyses tools, algorithms, and emerging computational platforms, relevant to all of the NGS data sets, will be collected. The literature and University of Kentucky researchers will be resources for building this list. Aim 3. Select performance metrics In order to quantitatively assess different methods and platforms, metrics will be selected for comparison of the tools, algorithms, and emerging computational platforms. Potential metrics include accuracy and speed of code, rates of data transfer, time-to-completion of analysis from start to delivery of results, and ease of tool use. Aim 4. Benchmark tools and computational platforms The list of tools will be installed on multiple platforms, including the University of Kentucky cluster (DLX) and through Globus Genomics1. Some tools may also be tested on national computational resources through XSEDE proposals.