Orthologs

Orthologs are optional

OpenGenomeBrowser does not require the following steps to be done. If orthologs and OrthoFinder-based trees are not required, simply turn the feature off in template.env with ORTHOFINDER_ENABLED=false.

A shortcut: opengenomebrowser-tools

Use the commands init_orthofinder and import_orthofinder from the opengenomebrowser-tools library!

export FOLDER_STRUCTURE=/path/to/folder_structure

# collect .faa files
init_orthofinder

# run OrthoFinder
orthofinder -f /path/to/folder_structure/OrthoFinder/fastas ...

# parse the output files into an OpenGenomeBrowser-compatible format
import_orthofinder

Introduction

Ortholog annotations are handled differently than regular annotations. The reason is that ortholog information for all genomes (result of orthofinder) is stored in one file which links orthologs to individual genes (orthologs/orthologs.tsv):

N0.HOG0000000\tgenome_1_123, genome_2_234
N0.HOG0000001\tgenome_2_345, genome_3_456, genome_4_567
N0.HOG0000002\tgenome_6_789, genome_7_890

Descriptions to orthologs may be provided in an additional file (annotation-descriptions/OL.tsv) with the following format:

N0.HOG0000000\tamino acid ABC transporter ATP-binding protein
N0.HOG0000001\twinged helix-turn-helix transcriptional regulator
N0.HOG0000002\tDNA topoisomerase (ATP-hydrolyzing) subunit B

Import orthologs

1. Open a terminal in the container
2. run python db_setup/manage_ogb.py import-orthologs

OrthoFinder (recommended, but optional)

If you want to use OrthoFinder, set the ORTHOFINDER_ENABLED environment variable to true ( see template.env).

OpenGenomeBrowser uses OrthoFinder to calculate core genome dendrograms. To shorten computation time and effort, OpenGenomeBrowser expects that the pairwise DIAMOND searches have already been run for all genomes for which this functionality should be available.

If you choose to add OrthoFinder, do as follows (can be done automatically using db_setup/init_orthofinder.py):

├── organisms
    └── ...
└── OrthoFinder
     └── fastas
         ├── <genome-identifier>.faa -> ../../organisms/.../genomes/<genome-identifier>/protein.faa
         ├── <genome-identifier>.faa -> ../../organisms/.../genomes/<genome-identifier>/protein.faa
         └── ...

1. Link (or copy) all protein fastas to OrthoFinder/fastas
   1. If your database is very big, you could exclude non-representative and contaminated genomes, for example
   2. The fastas must be named as follows: <genome-identifier>.<suffix>
   3. Tell OpenGenomeBrowser what your suffix is ( see template.env)
2. Run OrthoFinder (orthofinder -f /path/to/OrthoFinder/fastas)
   1. Because previously performed pairwise comparisons are stored and can be re-used OrthoFinder analyses can be updated very efficiently by simply adding or removing protein files. See OrthoFinder docs for more information.

When OrthoFinder is done, the folder structure should look like this:

└── organisms
    └── ...
└── OrthoFinder
     └── fastas
         ├── EXAMPLE1234-2-1.1.faa
         ├── ...
         └── OrthoFinder
             └── ...