CAPalyzer provides two utilities connected to the MetaSUB Core Analysis Pipeline. It condenses the results of the CAP into summary data tables and it provides utilities to parse those tables.
From PyPi
pip install capalyzer
From source
git clone [email protected]:dcdanko/capalyzer.git
cd capalyzer
python setup.py install
From the command line
capalyzer make-tables <cap result dir> <data packet dir>
In Code
#! /bin/env python3
from capalyzer.packet_builder import make_all_tables
filenames = make_all_tables(<cap result dir>, <data packet dir>)
for filename in filenames:
print(filename)
#! /bin/env python3
from capalyzer.packet_parser import DataTableFactory
table_factory = DataTableFactory(<data packet dir>)
taxa_tbl = table_factory.taxonomy()
amr_tbl = table_factory.amrs()
hmp-tbl = table_factory.hmp()
You can generate alpha diversity metrics using this package.
Available metrics
- shannon entropy
- richness
- chao1 richness
All metrics can be rarefied to a certain number of reads with the rarefy parameter.
from capalyzer.packet_parser import DataTableFactory
table_factory = DataTableFactory(<data packet dir>)
krakenhll_richness = table_factory.taxa_alpha_diversity(metric='richness', rarefy=1000000) # krakenhll is the default tool
metaphlan2_entropy = table_factory.taxa_alpha_diversity(tool='metaphlan2') # entropy is the default metric
Generally krakenhll is prefereable to metaphlan2 as it captures more diversity.
beta diversity scores are also supported
from capalyzer.packet_parser import DataTableFactory
table_factory = DataTableFactory(<data packet dir>)
jensen_shannon = table_factory.taxa_beta_diversity(metric='jsd')
rho_proportionality = table_factory.taxa_beta_diversity(metric='rho')
Diversity metrics may also be generated using the CLI.
capalyzer diversity alpha --help
capalyzer diversity beta --help
This package is written and maintained by David C. Danko