MetaboAnalystR "caches" the metabolite set libraries into your cwd as "qs" files, likewise for their compound name-to-ID mapping (with some command that you would use during the enrichment analysis on console).
The qs files from the MetaboAnalyst Platform have the following fields:
id = metaboanalyst ID (in the form of "hMSEAXXXX) name = name of the pathology/pathway member = semicolon-delimited text field listing compound common names reference = pubmed anchor html tags image = SMPDB ID
from fileinfo.com Automation install script used by Qt, a cross-platform application development toolkit; contains data used to locate and install user-defined features; compressed with the 7zip format.
installMetaboAnalyst.R
- code to install required dependencies and MetaboAnalystR itself.
convertMetaboAnalystLibraries.R
- has the code that extracts the metabolite libraries into separate RDS files for hyper-db
metaboliteMappingTable.R
- contains a function, metaboliteMappingTable.R, that acquires the metabolite name mapping through the Xia Lab's Compound Map API. No longer used, since these queries take a long time, but left as a reference point in case we do want to work with the API another time.
extractMetaboAnalystLibraryReferences
- code that extracts the literature references for each pathway within each metabolite set library
qs_Files
- Contains the qs files provided by MetaboAnalystR for Enrichment Analysis
RDS_Files
- Contains the RDS files made from extracting above qs files into a hypeR-friendly format. Also contains archive files of the sets based on compound name and by HMDB_ID, for troubleshooting regarding lack of overlap of our metabolites with these sets.
Within the qs files, there's a "reference" column that contains the pubmed link to the source paper within an html anchor tag.Unclear on if these sets were manually curated or done through mining.