Metabolomics_ML

This project is for a metaanalysis of published, untargeted metabolomics datasets from various human fluid (serum / plasma / CSF etc) for performing various diagnostics and predictive modeling

1. Scripts to download data are found in the 1_* directory
2. These data then need to be reprocessed with XCMS but fit need a common format (mzML/mzMXL or mzData) for which the 2_* directory helps with
3. Perform XCMS processing to extract features in 3_* 3.5) use extracting_features.ipynb to package the data from the outcome of the XCMS processing / author data into a form suitable for the next steps! This is key!
4. Build various predictive modelings with scripts in 4_*