WinBioinfTools: Bioinformatics Tools for Windows High Performance Computing Server 2008

TL;DR

WinBioinfTools is an open-source collection of bioinformatics tools optimized for Windows HPC Server 2008, enabling high-performance sequence analysis and demonstrating significant performance improvements over single machines and Linux clusters.

Contribution

This work introduces WinBioinfTools, the first open-source bioinformatics toolkit for Windows HPC, including ported tools and performance evaluations.

Findings

Significant performance gains using Windows HPC over single machines.

Successful porting of bioinformatics tools from Linux to Windows.

Comparable performance of WinBioinfTools on Windows and Linux clusters.

Abstract

Open source bioinformatics tools running under MS Windows are rare to find, and those running under Windows HPC cluster are almost non-existing. This is despite the fact that the Windows is the most popular operating system used among life scientists. Therefore, we introduce in this initiative WinBioinfTools, a toolkit containing a number of bioinformatics tools running under Windows High Performance Computing Server 2008. It is an open source code package, where users and developers can share and add to. We currently start with three programs from the area of sequence analysis: 1) CoCoNUT for pairwise genome comparison, 2) parallel BLAST for biological database search, and 3) parallel global pairwise sequence alignment. In this report, we focus on technical aspects concerning how some components of these tools were ported from Linux/Unix environment to run under Windows. We also show the advantages of using the Windows HPC Cluster 2008. We demonstrate by experiments the performance gain achieved when using a computer cluster against a single machine. Furthermore, we show the results of comparing the performance of WinBioinfTools on the Windows and Linux Cluster.