Bootstrapping with R and SPRINT
Posted: 24 Feb 2014 | 12:10
EPCC and the Division of Pathway Medicine at the University of Edinburgh have made public the report from their recent study into the performance of bootstrapping within their SPRINT R software package.
Bootstrapping is a popular, computationally-demanding resampling method used for measuring the accuracy of sample estimates and assisting with statistical inference. The report briefly describes bootstrapping and its availability within R; a free software environment for statistical computing and graphics that is very popular in both academia and commerce. The report compares the performance of the parallel implementation of bootstrapping within SPRINT with its parallel implementation in other R packages. It also highlights the need for parallel implementations of bootstrapping in R to scale up to handle the extremely large volumes of data expected in next-generation sequencing.
SPRINT (Simple Parallel R INTerface) is an easy-to-use parallel version of R. SPRINT allows R users access to high-performance computing without the need to master parallel programming methods, enabling the easy exploitation of HPC systems.
Download the report: Parallel Optimisation of Bootstrapping in R