This script generates a phylogenetic tree, as well as a matrix showing absence/presence of biosynthetic gene clusters alongside this tree, for a set of genomes. It takes a list of accession numbers of genomes, predicts biosynthetic gene clusters for them using antiSMASH, groups these BGCs into families with BiG-SCAPE and NetworkX, and visualizes absence/presence of these families for each organism alongside a phylogenetic tree with autoMLST.

The following programs need to be installed:

• ncbi-acc-download
• BiG-SCAPE
• the cluster-splitting fork of antiSMASH
• the simplified wrapper fork of autoMLST

Additionally, several Python virtual environments are required. Requirements for three of these are given in requirements-ete-env.txt, requirements-general-env.txt, and requirements-networkx-env.txt. Beyond this, virtual environments for autoMLST and antiSMASH are also required; requirements for these can be found in their respective repositories.

Before running mibig-gbk-to-trees.sh for the first time, several placeholders need to be replaced. These are:

• /path/to/.virtualenvs/antismash-dmz_markers/bin: replace with path to antiSMASH virtualenv
• /path/to/.virtualenvs/automlst_env/bin: replace with path to autoMLST virtualenv
• /path/to/.virtualenvs/ete3_env/bin: replace with path to virtualenv specified in requirements-ete-env.txt
• /path/to/.virtualenvs/base_env/bin: replace with path to virtualenv specified in requirements-general-env.txt
• /path/to/.virtualenvs/networkx_env/bin/: replace with path to virtualenv specified in requirements-networkx-env.txt

• /path/to/BiG-SCAPE-master: replace with path to dir containing BiG-SCAPE
• /path/to/Pfam-A: replace with path to PFAM directory used for BiG-SCAPE (normally, but not always, in BiG-SCAPE dir)
• /path/to/autoMLST/ziemertlab-automlst-7b2b5a9a8961: replace with path to dir containing autoMLST fork
• /path/to/antismash-dmz_markers: replace with path to dir containing antiSMASH fork
• /path/to/absprestree: replace with path to dir containing this script

• /path/to/default/dir: replace with path to the directory you wish to run the script in by default If desired, you can also change curated-accessions-tiny to a list of accessions you wish to run by default.

To run the script with default settings, simply run bash mibig-gbk-to-trees.sh. All that is required is a plain text file listing accession numbers of the genomes you wish to generate an absence/presence matrix and tree for, one per line, under the name set as a default in the default base directory. Additional arguments are:

• -b: Base directory to run script in; all files and directories are normally generated here
• -a: plain text file containing accession numbers of your desired genomes, one per line
• -t: name (not path!) of the final tree image. Formats supported by ete3 are .png, .pdf and .svg.
• -g: name (not accession!) of outgroup(s) to be used. Replace spaces with _ and remove all non-alphanumeric characters except - and _. If you wish to use more than one outgroup, separate by spaces and wrap the entire list in double quotes (e.g. "Bacillus_cereus_ATCC_14579 Bacillus_megaterium_NBRC_15308__ATCC_14581")