v2.3.0

• fixed bug in matchSpectralDB, that is only InChIKeys in entries of MoNA msp files. Unique InChIs are now in Comments field. A while loop was also added to mitigate for any errors from OpenBabel. v2.3.0

N.B. Regarding this issue msp files downloadable from MoNA no longer contain unique chemical identification codes (i.e. InChI or SMILES). This has temporarily broken the metID.matchSpectralDB function when using the latest msp files from MoNA. However, example msp files hosted on our GitHub repo mspFiles can still be used in the meantime. An issue has been opened on the MoNA database issue #194. This is a relatively recent change and OpenBabel has no functionality to convert InChIKeys to InChI or SMILES codes. This is because InChIKeys do not represent truly unique chemical structures. See:
"Note: that while a molecule with a particular InChI will always give the same InChIKey, the reverse is not true; there may exist more than one molecule which have different InChIs but yield the same InChIKey." https://openbabel.org/docs/dev/FileFormats/InChIKey.html

v2.2.8

• fixed bug in combineMS2.Spectra. Naming of metaData slot entries was incorrect in usage certain cases.

v2.2.6

1. shinyapps deprecated. Updated Rcpp tested publishApp.
2. debugged pubmed wordcloud and added modified function cleanAbs.
3. Patch required for server.R compMS2Explorer.

v2.2.3

many additions have been made including:

• consensus scoring metrics metID.buildConsensus and differential evolution based optimization metID.optimConsensus functions.
• mean maximum chemical similarity scoring using the correlation and/or spectral similarity networks metID.chemSim.
• molecular descriptor - random forest based recursive feature elimination QSRR. metID.rtPred
• generation of msp files from compMS2 objects. metID.compMS2toMsp
• many additional functions/visualization in the compMS2Explorer shiny app.
• redundant/contaminant spectrum removal. combineMS2.removeContam
• automatic true/false positive removal. falsePosIdentify

v1.2.5

bugs in compMS2explorer fixed and dbentry visualization on click.

CompMS2miner 1.2.3

• Addition of correlation (metID.corrNetwork()) and spectral similarity (metID.specSimNetwork()) networks metID methods which can be visualized in CompMS2explorer.
• added further metID() function metID.matchSpectralDB() which can be used to match msp file databases (NIST ASCII text files) to the composite spectra, head to tail plot visualization added if a spectralDB match has been made.
• publishApp() and runGitHubApp() functions for CompMS2explorer sharing and publishing.
• added dynamic noise filtration example video to tabbed panel.

Initial Release

A long-standing challenge of untargeted metabolomic profiling by liquid-chromatography - high resolution mass spectrometry analysis (LC-hrMS) is rapid, precise and automatable transition from unknown mass spectral features in the form of a peak-picking software output peak tables to full metabolite identification.

CompMS2miner a package in the R programming language was developed to facilitate rapid, comprehensive unknown feature identification using peak-picker output files and MS/MS data files as inputs. CompMS2miner matches unknown mass spectral features to precursor MS/MS scans, dynamically filters variable noise, generates composite mass spectra by multiple scan signal summation, interprets possible substructures from a literature curated database, annotates unknown masses from several metabolomic databases, performs crude prediction of mammalian biotransformation metabolites and provides wrapper functions for pre-existing insilico fragmentation software (http://msbi.ipb-halle.de/MetFrag/).

Data curation, visualization and sharing is made possible at any stage of the CompMS2miner package workflow via an application developed with the R shiny package.

Example data illustrating CompMS2miner is provided consisting of a peak-picker output table of nano-flow LC-hrMS metabolomic dataset of human blood samples and data-dependent MS/MS data files, which is also made available as external example data within the CompMS2miner package. A example workflow using this data is available in the package vignette. The CompMS2miner package is designed to offer a more complete solution to the LC-hrMS metabolite identification challenge than currently available softwares in the R language and is also complementary to other extant R packages/ workflows.