SODA is a tool for species delimitation using only gene tree topologies. It's a fast way of doing species delimitation and relies on a polytomy test developed by Sayyari et. al implemented inside ASTRAL package. SODA has been in our tests slightly less accurate than BPP, but it is far faster.

• You can find the details of the algorithm and benchmarks in the following paper:

  • Maryam Rabiee, Siavash Mirarab, SODA: multi-locus species delimitation using quartet frequencies, Bioinformatics, Volume 36, Issue 24, 15 December 2020, Pages 5623–5631, https://doi.org/10.1093/bioinformatics/btaa1010

• Java, version 1.6 or later
• Dendropy, version > 4.0.0; Dendropy package which is easy to install with pip. You can find installation instructions here.

1. Clone the code from this github page or download the zip file.
2. Unzip the package if you downloaded the zip file.
3. cd into the directory where SODA code is placed.

No further installation step is required for the current version of SODA.

For running SODA you need your gene trees in newick format in one file and your guide tree in another file. The guide tree is an option. If you haven't obtained the guide tree, SODA uses ASTRAL to compute one.

For running SODA, given only input gene trees, you should run the following command:

python3 run_delimitation.py -i [Gene Tree File] -d [Out directory] -o [Output Name]  

You can see the list of options by running:

python3 run_delimitation.py 

• You can provide a custom guide tree using -t option.
  • If you know the correct rooting of the guide tree, use the -r option ("rooted") so that SODA avoids rerooting the tree.
• The p-value cutoff value is optional (by default, it has been set to 0.05) but can be changed and it will affect the number of species. Check out the experiments in the paper.
• Use -a option to tell SODA about the population groupings known a priori. SODA will not break these populations.
  • When this option is provided, the guide tree (if provided by the user) should have the population names not the individual names.

With the data given in the samle_data directory, you can test SODA on your machine with the following commands. To run SODA just with gene trees and a mapping of individuals to populations, you should set options as:

python3 run_delimitation.py  -i sample_data/all-gene-trees.tre.pruned -d output/ -o delim.cl  -a sample_data/mapping.txt   

The delimitation of the population can be found in the delim.cl file. To run SODA with a rooted guide tree and a specific cut-off value you can run this command:

python3 run_delimitation.py -i sample_data/all-gene-trees.tre.pruned -d output/ -o delim.cl  -t sample_data/astral.out.rt -r

The output of SODA is the names of individuals and the corresponding number per each individual. Individuals that have the same number assigned to them are in the same group and comprise a distinct species.

A Sample output would look like this:

P_min12908      0
P_min39119      0
P_min13868      0
P_min18421      0
P_min13898      0
P_min5292       0
P_v_meri16      1
P_v_cos199      1
P_v_vers_1      2
P_v_vir379      3
P_x_xant_1      3
...

Here, four different species have been identified.

In addition to the output file, inside the directory given using -d, SODA saves several files. These include the guide tree, before rooting, after rooting, and after annotation with p-values per each branch.

Please read the paper carefully for the caveats of SODA. In particular, note that SODA only works properly if you have multiple individuals from each species. This is true not just for your ingroup taxa but also for outgroups. When you have a single individual from a species, the behavior is a bit unpredictable.

• In the particular case of an outgroup from which you have only one individual and you know the outgroup is different from your species, you can simply ignore it if SODA puts the outgroup with the ingroup. The grouping of the ingroup is still valid (if it has multiple individuals).