CD-MRG-metabolomics_summerschool_2023
This repository contains teaching material for the 3rd International Summer School on Non-Targeted Metabolomics Data Mining for Biomedical Research.
Exercise
Non-invasive drug monitoring is desirable as it improves patient experience. Instead of relying on invasive blood draws drug pharmacokinetics and metabolism could for example be monitored through the skin. During the summer school, you will be working with a subset of the data published by Panitchpakdi and collaborators (2022). The datasets consists of plasma and skin swabs (forearm, forehead and upper back) collected from healthy volunteers (n=7) over the course of 24 hours, who have been administered the antihistaminic diphenhydramine. Your task will be to investigate whether:
- Diphenhydramine and its metabolites can be detected in the skin and whether it exhibits similar pharmacokinetics as in plasma?
- You can observe other metabolites that exhibit interesting time trends in plasma and whether those metabolites can also be detected in the skin?
Data
Metabolomics mass spectrometry data are publicly available in the open .mzML format at: MSV000092389.
As a student you will be divided into two groups. Group 1 will be working with skin samples collected from the forehead, whereas Group 2 will work on skin samples collected from the forearm. Both groups will be investigating whether their respective skin samples could be suitable for non-invasive drug monitoring by comparing them to the plasma samples. The instructor team will be be working with skin samples collected from the upper back.
MZmine 3
Preprocessed data are available in the subfolders data/mzmine_output. [We added some placeholders for now, MZmine 3 team, please replace with the corresponding outputs for downstream analysis]
GNPS
Mass spectral molecular networks of the data are publicly accessible through the links below.
Example: https://gnps.ucsd.edu/ProteoSAFe/status.jsp?task=6c9c4b22970f448d933a6fc7fa13227e
Group 1: https://gnps.ucsd.edu/ProteoSAFe/status.jsp?task=a2baed56d46648898aad480f9e484160
Group 2: https://gnps.ucsd.edu/ProteoSAFe/status.jsp?task=ca64bf41771b4152a3f94cc09b93c546
Sirius+CSI:FingerID
Results are available as complete SIRIUS project spaces and via corresponding summary files.
Summary files for molecular formula annotations (formula_identifications.tsv), molecular structure annotations (compound_identifications.tsv) and compound class annotations (canopus_compound_summary.tsv)
can be found in subfolders of
data/sirius_output.
Complete SIRIUS project spaces are available through the links below.
Example: https://upload.uni-jena.de/data/64cb6d9e183353.74747773/example-sirius-projectspace.zip
Group1: https://upload.uni-jena.de/data/64cb6b5f245b16.51592951/group1-sirius-projectspace.zip
Group2: https://upload.uni-jena.de/data/64cb6cbadf6b89.77294609/group2-sirius-projectspace.zip
MS2LDA
Results are available through the provided GNPS and MS2LDA.org links below. Briefly, the GNPS-MS2LDA links provide quick insights in especially mass spectra that are linked to annotated Mass2Motifs from selected MotifSets from MotifDB. If there are un-annotated Mass2Motifs that require further investigation, the .dict file (downloaded from GNPS) can be uploaded into MS2LDA.org to explore the mass fragment and/or neutral loss features that are associated to the Mass2Motif. Further information such as statistics on how many mass spectra are connected to Mass2Motifs are also available.
GNPS-MS2LDA results are available through the links below.
Example: https://gnps.ucsd.edu/ProteoSAFe/status.jsp?task=15fbb620a6f24582a8d1b25e54ecc2f5
Group1: https://gnps.ucsd.edu/ProteoSAFe/status.jsp?task=396fbe18da6d4da8aadf08afa8037614
Group2: https://gnps.ucsd.edu/ProteoSAFe/status.jsp?task=ab533f6174d84759af509792570814b7
MS2LDA.org results are available through the links below.
Example: https://ms2lda.org/basicviz/summary/2988/
Group1: https://ms2lda.org/basicviz/summary/2990/
Group2: https://ms2lda.org/basicviz/summary/2991/
matchms
Statistics
Example jupyter notebooks used for data analysis are available in the subfolder notebooks.
Interactive jupyter notebooks can be launched in Google Colaboratory here:
