The req.txt file has all the python packages needed to run the scripts

sudo pip install -r req.txt

Usage: ./1_filter_table_by_gene_ct_value.py <table> <ct> > 1_output_filename.txt 2> filter.log (log will have status updates on where it is in the file) EXAMPLE:

./1_filter_table_by_gene_ct_value.py dropseq_all.txt 900 > dropseq_training.txt 2> filter.log

Arguments: <table> table with molecule counts, samples in head, genes as row labels <ct> min gene count threshold for sample inclusion

Usage: ./2_calc_mean_variance_bin.py <table> <out> <hflag> 2> metrics.log (log will have progress of script) EXAMPLE:

./2_calc_mean_variance_bin.py dropseq_training.txt dropseq_metrics.txt 1 2> metric.log

Arguments: <table> table with molecule counts, samples in head, genes as row labels (table from step one) <out> output file name rather than stdout <hflag> 0 if input table has a header row, 0 if not

Usage: ./3_bin_norm_sel.py <table> <score> 2> bin.log (log will have progress and brief description of statistical output) EXAMPLE:

./3_bin_norm_sel.py dropseq_metrics.txt 1.7 2> select.log

Arguments: <table> table with gene metrics mean, variance and disperion <score> Z score cutoff for selection

Options: -h

This script is forthcoming, will use samples from first output, genes from last output to create table for PCA