Computational protocol: Reproducible Analysis of Post-Translational Modifications in Proteomes—Application to Human Mutations

Similar protocols

Protocol publication

[…] The ProteomeScoutAPI was written in Python and is available in a Mercurial repository on the ProteomeScout Assembla project page. The current stable release (v1.0b, November, 2015) of ProteomeScout mammalian PTM file was downloaded from the ProteomeScout stable release FTP site ftp://ftp.seas.wustl.edu/pub/ProteomeScout_DbF/current_stable_release/. All calculations and analyses were performed in Python, specifically using iPython [] notebooks and the following open-source projects: Pandas [], NumPy [], and Matplotlib []. Testing for enrichment was done using a one-sided Fisher’s Exact test. We counted amino acids uniquely for having either a mutation or a modification within the specified window of 0 (on the amino acid) or 8 (within +/- 8 amino acids of the residue). If more than one mutation exists on the same amino acid and at least one of the mutations was known to be pathogenic or disease-related, then we assigned that residue a pathogenic/disease phenotype during all analyses. False discovery rate [] was used as a multiple hypothesis correction technique, where denoted.In accordance with recommendations on best practices [] for developing bioinformatics software, the ProteomeScoutAPI has a full software testing suite, built using the Python unittest framework. This series of tests ensures that updates and changes to the code do not inadvertently lead to the introduction of software bugs elsewhere. Importantly, extending and growing this suite to accommodate new features is simply a few lines of additional code, ensuring that as the ProteomeScoutAPI grows and new functionality is added, a formal testing framework can be built in parallel.The analysis code is available on GitHub and can be visualized on nbviewer at: http://nbviewer.ipython.org/github/knaegle/MutationsNotebooks/tree/master/. All calculations and graphs for study bias, PTM enrichment, and resampling for charge distributions are available in the iPython notebooks. SVG exports for RAF1 studies and data from the study by Rigbolt et al. [] were taken from ProteomeScout []. […]

Pipeline specifications

Software tools Numpy, matplotlib
Databases ProteomeScout
Application Miscellaneous
Organisms Homo sapiens