Repository companion for BCB 501 project paper by Kai Davidson, David Guevara, and Kasish Gupta.
Navigate to ibdmdb.org and download the aggregated taxonomic abundance as a tsv file from the files tab.
To download the HMP files, you must access the AWS bucket they are stored on. The correct command to download the specific file used in this project is:
aws s3 ls s3://hmpdcc/hmp1/hhs/microbiome/wms/analysis/metabolic_prof/humann2_2017/hmp1-II_humann2_pathabundance-nrm-mtd-qcd.pcl.bz2 . --no-sign-request
Afterwards, use the csv file in the hmp_manifests folder to filter out only healthy patients using the create_dataset.py functions within the file.
To use the models specify the correct path for the normalized and aggregated genus data in the data folder. Then run python <model.py> to run the model and generate plots. Change the dataset within the model files to run on different datasets.