DownloadTo download please go to Browse All Files for the complete pack!
In mass spectrometry the detector responses are not always directly proportional to the quantity of compounds. Large peaks in LCMS traces are not equal to major components but their ionizability. The effectiveness of a medicinal plant, poisons in a mushroom, resistance and susceptibility to pest and diseases are not necessarily because of the compounds that give large peaks. These traits are determined by the different levels of specific metabolites which may show up only as very small peaks.
How would it be possible to spot an important metabolite or two, from 1000’s that occur in cells and tissues? If those truly relevant metabolites were missed during peak-picking then the following stats, however sophisticated they may be, would fail. And how many peaks are there in a single LCMS run? How can the important metabolites be represented in the top 100 or even 10000 peaks? This project tries to address these issues.
Features
- Data mining from LC-MS and GC-MS files (in NetCDF Classic format)
- Quasi-data point-wise comparison between groups of samples for detecting group-discriminating ion signals
- Univariate statistics (screening window coefficient (Z-factor), Student's t-test, and z-test)
Project Samples
