Paired mass distance (PMD) analysis proposed in Yu, Olkowicz and Pawliszyn (2018) and PMD based reactomics proposed in Yu and Petrick (2020) for gas/liquid chromatography–mass spectrometry (GC/LC-MS) based non-targeted analysis. PMD analysis including GlobalStd algorithm and structure/reaction directed analysis. GlobalStd algorithm could found independent peaks in m/z-retention time profiles based on retention time hierarchical cluster analysis and frequency analysis of paired mass distances within retention time groups. Structure directed analysis could be used to find potential relationship among those independent peaks in different retention time groups based on frequency of paired mass distances. Reactomics analysis could also be performed to build PMD network, assign sources and make biomarker reaction discovery. GUIs for PMD analysis is also included as ‘shiny’ applications.
You can install package from this GitHub repository:
{r} remotes::install_github("yufree/pmd")
Or find a stable version from CRAN:
{r} install.packages('pmd')
For MS only data such as FT-ICR or MS imaging data, check the section of “Reactomics analysis for MS only data” in Reactomics Analysis Tutorial or this blog post.
PMD analysis paper. This paper proposed structure/reaction directed analysis with PMD.
Reactomics paper. This paper contained the concepts of PMD based reactomics, applications and data mining of reaction database and compounds database.
Slides. This is the slides for ASMS 2020 Reboot and here is the video of presentation. Press “P” and you will see the notes for each slide with details. Another full version of reactomics presentation for one hour presentation could be found here. I will not update the conference presentation while I will add new contents for the full version of reactomics presentation whenever I have new results.
To perform GlobalStd algorithm, use the following code:
{r} library(pmd) data("spmeinvivo") pmd <- getpaired(spmeinvivo) std <- getstd(pmd)
To perform structure/reaction directed analysis, use the following
code:
{r} sda <- getsda(std)
To perform GlobalStd algorithem along with structure/reaction directed analysis, use the following code:
{r} result <- globalstd(spmeinvivo)
To use the shiny application within the package, use the following code:
{r} runPMD()
To find homologous series (e.g., repeated addition of CH2) or specific reaction sequences:
```{r} # Find homologous series of CH2 (14.0157) with at least 3 nodes homolog_series <- gethomolog(spmeinvivo, unit = 14.0157, min_len = 3)
seqs <- getchainseq(spmeinvivo, c(162.0528, -18.0106))
To screen the whole feature list against a curated database of known reaction chains (e.g. desaturation-elongation, hydroxylation-glucuronidation, sequential oxidation), use the built-in chain database:
```{r}
data("pmdchain")
r <- getchainseq(spmeinvivo, db = pmdchain)
# one row per database chain: which named chains are present in this sample,
# how many matched paths were found, and the chain's `specificity` (the number
# of KEGG reaction paths sharing the chain's PMD signature: lower = rarer =
# more specific, so a rare-chain match carries more weight than a common edit)
head(r$chainsearch)A match here is a relational annotation (co-varying,
chromatographically resolved features related by the chain’s mass
differences), not compound identification. Both shiny applications
(runPMD() and runPMDnet()) also include a
one-click “Screen for known reaction chains” panel for this database
screening.
To check the pmd reaction database:
{r} # all reaction data("omics") View(omics) # kegg reaction data("keggrall") View(keggrall) # literature reaction for mass spectrometry data("sda") View(sda) # curated known reaction-chain database for screening data("pmdchain") View(pmdchain)
To check the HMDB pmd database:
{r} data("hmdb") View(hmdb)
To cite related papers:
{r} citation('pmd')