anvi-script-gen-alluvial

#!/usr/bin/env python

# lousy script that works with the outputs of `anvi-export-collection`
# to reconcstruct the fate of contigs for a given algorithm and bin and
# spit out some text to be visualized on https://app.rawgraphs.io/

import sys

from collections import OrderedDict

import anvio.utils as u

from anvio.errors import ConfigError

G = lambda x: u.get_TAB_delimited_file_as_dictionary(x, no_header=True, column_names=["split_name", "bin_name"])

# crappy way to do this, indeed, but it will suffice for today.
# you need to put the output of anvi-export-collection for each
# binning algorithm into this dict with the matching filename:
algorithms = OrderedDict({'CONCOCT': G('EP-CONCOCT_1500.txt'),
                          'MAXBIN2': G('EP-MAXBIN2_1500.txt'),
                          'METABAT2': G('EP-METABAT2_1500.txt'),
                          'DASTOOL': G('EP-DASTOOL_1500.txt')})

def main(args):
    algorithm = args.algorithm
    bin_name = args.bin

    if algorithm not in algorithms:
        raise ConfigError("That algorithm we don't know. This is what we know: %s." % ', '.join(algorithms))
        sys.exit(-1)

    split_names = set()
    for split_name in algorithms[algorithm]:
        if algorithms[algorithm][split_name]['bin_name'] == bin_name:
            split_names.add(split_name)

    print(','.join(['split'] + list(algorithms.keys())))
    for split_name in split_names:
        print(','.join([split_name] + [algorithms[algorithm][split_name]['bin_name'] for algorithm in algorithms]))


if __name__ == '__main__':
    import argparse

    parser = argparse.ArgumentParser("Give this guy one of the algorithms (it knows all about '%s') and a bin name,\
                                      get back an input data for an alluvial diagram that shows which bins contain\
                                      the splits in that bin..")

    parser.add_argument('--algorithm', default=None, help="Which algorithm?")
    parser.add_argument('--bin', default=None, help="Which bin name to start with?")

    args = parser.parse_args()

    try:
        main(args)
    except ConfigError as e:
        print(e)
        sys.exit(-1)

Hi meren,

Really appreciate your prompt reply.

I changed the names in the script, like this:
algorithms = OrderedDict({'CONCOCT': G('concoct.txt'),
'MAXBIN2': G('maxbin.txt'),
'METABAT2': G('metabat.txt'),
'METAWRAP': G('metawrap.txt')})

But I still get 'ConfigError'. Should I add directory of the corresponding files? Right now .txt files and the script are in the same folder directory.

Thank you!

If you can send me your files I can find out what is wrong with them. I'm sorry it's been frustrating.

Hi meren,

I finally figured out the correct format for alluvial illustrations (on "https://app.rawgraphs.io/").

Just need to generate a table with contig names on first column and bin names on the rest of columns. Something like the following:
Contigs | maxbin | metabat | metawrap | concoct
c_000000000502_split_00001 | maxbin_bin32 | metabat_bin63 | mv_bin31 | concoct_bin95
c_000000000512_split_00001 | maxbin_bin32 | metabat_bin63 | mv_bin31 | concoct_bin95
c_000000000571_split_00001 | maxbin_bin32 | metabat_bin63 | mv_bin31 | concoct_bin95
c_000000001297_split_00001 | maxbin_bin32 | metabat_bin63 | mv_bin31 | concoct_bin95
c_000000001297_split_00002 | maxbin_bin32 | metabat_bin63 | mv_bin31 | concoct_bin95
.......
There are actually multiple ways for doing this job.

Appreciate this idea for using alluvial figures to compare binning results.

Thanks!

#Hello,

#I am having some trouble getting this script to run.

#I have modified the algorithms sections as follows

algorithms = OrderedDict({'ANVIO_CONCOCT': G('collection-Concoct.txt'),
'ANVIO_METABAT': G('collection-bbmap_Metabat2.txt'),
'MANUAL_METABAT': G('collection-manual_metabat_bins.txt'),
'DASTOOL': G('collection-dastool_bins.txt')})

#My files are located in the same folder as the script and were exported from anvio through anvi-export-collections.
collection-bbmap_Metabat2-info.txt collection-Concoct.txt collection-manual_metabat_bins-info.txt
collection-bbmap_Metabat2.txt collection-dastool_bins-info.txt collection-manual_metabat_bins.txt
collection-Concoct-info.txt collection-dastool_bins.txt

#I have been running the script as follows
python anvi-script-gen-alluvial.py

#I also had to add a line that imports anvio at the start of the script otherwise it was unable to find a module named anvio.utils.
import anvio

#Now the issue that I am running into is this:
Config Error: That algorithm we don't know. This is what we know: ANVIO_CONCOCT,
ANVIO_METABAT, MANUAL_METABAT, DASTOOL.

#So i guess it is getting stuck as that first if statement. I am not super familiar with python so any help would be greatly appreciated.

#Thanks !
#Claudia

#FIGURED IT OUT!
need to have --algorithm and --bin flags to run

for example
python anvi-script-gen-alluvial.py --algorithm ANVIO_CONCOCT --bin Bin_27