Usage:

You need to first create two folders.
put clean.sh and process.sh in strains2_training_genomes folder after download 
put all the genomes (fna) file into strains2_training_genomes folder
put all the metagenomes (fastq) file into strains2_training_metagenomes folder

process.sh script will assume strains2_training_genomes and strains2_training_metagenomes lie in the same folder.
./process.sh to run the shell script
./clean.sh if you want to redo the process

change certain line if necessary (if you only want to run certain column of data)

run  `python gidx.py` parses the genomes into a long string, parse metagenomes into a list of strings, and generates the features in csv format.
run `python model.py` to train sigmoid or neural network on generated features and labels data




Data can be accessed via: https://platform.mosaicbiome.com/workspaces/219/files