CHURP

Abstract

Progressive decreases in the cost of DNA sequencing have contributed to a decades-long exponential increase in the production of new sequencing datasets. The processing of these datasets has in turn led biology, a field that has traditionally relied on local "lab" servers to address its computational needs, to become increasingly reliant on High Performance Computing (HPC) resources. Though many operations on sequencing datasets are trivially parallelizable on multiple levels, the lack of an HPC tradition in biological research has hampered fully parallelized deployments.