TaxCo computes correlations between taxonomic abundances and numeric metadata and presents the results in tabular and graphic form.
usage: TaxCo.py [-h]
(-qiime <L2> <L3> <L4> <L5> <L6> | -mothur <tax.summary file> | -megan <L2> <L3> <L4> <L5> <L6>)
[-out <output directory>] [-pval <p-value>]
[-cutoffs <decimal number> [<decimal number> ...]]
[-coefficients <correlation coefficients> [<correlation coefficients> ...]]
[-metadata <metadata>]
optional arguments:
-h, --help show this help message and exit
-out <output directory>
Path to output directory
-pval <p-value> P-value cutoff
-cutoffs <decimal number> [<decimal number> ...]
Cutoffs for correlation strength, with a dot as
decimal separator
-coefficients <correlation coefficients> [<correlation coefficients> ...]
One or more correlation coefficients (kendall,
spearman, pearson)
-metadata <metadata> Metadata in the format MEGAN exports it in (tab
separated, samples in rows, data in columns)
Input files (mutually exclusive):
-qiime <L2> <L3> <L4> <L5> <L6>
Paths to QIIME taxon summaries, sorted from L2 to L6.
Samples have to be in the same order in each file.
-mothur <tax.summary file>
Path to mothur tax.summary file (including all
samples)
-megan <L2> <L3> <L4> <L5> <L6>
Paths to MEGAN CSV files, sorted from L2 to L6.
Collapse tree to matching level and export the
summarized counts.