Large scale Markov clustering (MCL) via subgraph extraction
Project description
bigmcl
Large scale Markov clustering (MCL) via subgraph extraction
bigmcl
will isolate disconnected subgraphs from a large graph file and execute
MCL on the subgraphs. bigmcl enables MCL on large, highly disconnected graphs,
such as those used in orthogroup inference. Not recommended for graphs that are
manageable with typical MCL.
Important to note that the inflation parameter is affected by this approach - I have noted clusters are more granular if anything. In the future, I plan on implementing a systematic approach option for identifying ideal inflations for each subgraph.
Please cite this git repository and MCL when this software contributes to your analysis.
INSTALL
pip install bigmcl
Clone mcl
from here, compile, and add to your path.
USE
Input and go:
bigmcl -i <GRAPH.imx> -I 1.5
More elaborate options:
usage: bigmcl.py [-h] -i INPUT -I INFLATION [-s] [-r ROW_FILE] [-m] [-o OUTPUT] [-c CORES]
[-v]
Isolates disconnected graphs and runs MCL on the subgraphs. Input data must be numerical.
optional arguments:
-h, --help show this help message and exit
-i INPUT, --input INPUT
MCL graph file in imx format
-I INFLATION, --inflation INFLATION
-s, --symmetric Matrix is symmetric (throughput increase)
-r ROW_FILE, --row_file ROW_FILE
Continue from finished row.txt
-m, --mcl_format Output clusters in MCL format
-o OUTPUT, --output OUTPUT
Alternative output directory
-c CORES, --cores CORES
-v, --verbose
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.