rawDiag | Accessing orbitrap data on the fast lane using R

Despite the widespread usage of R for other omics data analysis, it has
been largely ignored by the proteomics community. We believe the mayor
reason is the lack of a direct and performant interface between R and
mass spectrometry raw data formats. So far this gap has been poorly
bridged by the HUPO-PSI standards (mzML, mzXML) that can be read into R
using packages like mzR. The problem to this approach is that raw files
in vendor specific binary formats first need to be converted to these
open formats and later on parsed into R objects. Both steps consume
substantial computational resources and obstruct the development of lean
data analysis workflows in R. We therefore suggest to bypass these
costly conversions and instead use vendor-released APIs to directly
access raw data from the R command line. Our previously published R
package rawDiag [10.1021/acs.jproteome.8b00173] demonstrates the power of this approach by
means of scan
metadata contained in Thermo Fisher Scientific raw files. Currently, we
are extending our philosophy to mass spectra, the core data unit in
LC-MS driven proteomics. Our initial results show that complex
analytical workflows dealing with the analysis of spectral data can be
written directly in R and executed on standard computational
infrastructures in an OS independent fashion.