Software

MeSH recommend

his is a supplementary software for "Automated recommendation of research keywords from PubMed database that connect between metabolome data and researcher's knowledge" by Kanazawa et al. submitted. This is an output of the collaboration between Osaka University and Shimadzu Corporation.

Required Python modules: Required Python modules: Python 3.8.8, Numpy 1.20.1, Pandas 1.2.4, Scipy 1.6.2, Statsmodels 0.12.2

MeSH recommend Python classes and exemplary scripts of automated recommendation.

How to use:
1. Put json files (e.g. myfile1.json) at data/json. The list of metabolites and the known keyword are described in the json file and given to the program.
2. Run ./recommendation/main.py (e.g. python ./recommendation/main.py ./data/json/myfile1.json) Optional arguments in "./recommendation/main.py" allow you to modify parameters. You can check it in "python ./recommendation/main.py --help".
3. The script creates folder (e.g. myfile1) that includes the recommendation results (recommendation.html) and the over-representation analysis results (enrichment.html).

AI peak picker

This is a supplementary software for "Fake metabolomics chromatogram generation for facilitating deep learning of peak-picking neural networks" by Kanazawa et al. submitted. This is an output of the collaboration between Osaka University and Shimadzu Corporation.

Required Python modules: Python 3.6.8, Chainer 6.1.0, Numpy 1.15.3, Pandas 0.24.2, Scipy 0.19.1, matplotlib 3.2.0

Fake chromatogram generator Python classes and exemplary scripts of the fake chromatogram generator (Fake_chromatogram_generator).

How to use:
1. Run ./Fake_chromatogram_generator/fake_chromatogram_generator.py.
2. The script creates “output” folder that includes two text files of chromatogram (input.txt) and exact start and end points of peaks (output.txt).

U-Net based peak picker U-net based peak picking neural network (U-Net)

How to use:
1. Put folder including chromatogram data files at Test/u-net/.
2. Run ./U-Net /PeakPicking_unet.py.
3. The script creates “result” folder that includes peak-pickin results.