SpecKS

Genome-level forward simulation of polyploid evolution. Produces Ks data suitable for generating simulated Ks histograms.

Example command

python3 SpecKS.py config.xml

Simulation overview

SpecKS is a pipeline application, with the following steps:

1. Make species trees for polyploids
2. Make gene trees
3. Run gene-birth-and-death model
4. Run ohnolog-shedding model
5. Evolve gene sequences along gene trees (PAML: evolver)
6. Calculate Ks for polyploid (PAML: codeml)
7. Plot simple histograms*
8. Prepare the final results (.csv files with Ks data)

• Note, SpecKS is not intended to produce beautiful histograms. It is left to the user to upload the Ks datafiles generated by SpecKS to the visualization tools of choice.

Prerequisites

• Run on an environment with PAML installed. We suggest v4.10.7 but have found that other versions also work.
• We give directions for configuring an appropriate conda environment below.

Installation

git clone https://github.com/tamsen/SpecKS

Configuration

To automatically configure a conda environment suitable for SpecKS, use the "specks_environment.yml" included in the "sample_configs" folder. Example commands to install and do a short run are below. "my-test-run.xml" should be a copy of "short-run.xml" with the paths modifed as needed.

conda env create -f /PathToSpecKS/sample_configs/SpecKS_environment.yml

conda activate specks_env

python3 /PathToSpecKS/SpecKS.py my-test-run.xml

Help

• There are example config files and an overview of the parameters in the "sample_configs" folder.

OS

• Tested on a mac and linux (ubuntu)

Referencing SpecKS

• Publication forthcoming. Please reference "Accurate Inference of the Polyploid Continuum using Forward-time Simulations" by T. Dunn and A. Sethuraman