Skip to content

everhartlab/sclerotinia-review-2017

Repository files navigation

Analysis for review of S. sclerotiorum mycelial compatibility groups

Licence

This repository contains code for a meta analysis of mycelial compatibility groups in the Sclerotinia sclerotiorum literature.

A preprint version of the manuscript submitted to Tropical Plant Pathology can be found here:

Kamvar ZN, Everhart SE. (2018) Something in the agar does not compute: On the discriminatory power of mycelial compatibility in Sclerotinia sclerotiorum. PeerJ Preprints 6:e26670v1 doi: 10.7287/peerj.preprints.26670v1

Manuscript

The manuscript is built with pandoc and xelatex and controlled with GNU make. If you have those installed, you can rebuild the PDF and word version of the manuscript with:

make -C manuscript

Details of how manuscript was written and generated into pdf/word

The manuscript was originally written in Google Docs with the Paperpile plugin for citations for the sake of an easy collaboration tool. Because there were some formatting issues with the citations in Paperpile, we had downloaded the citations in BibTeX format and the manuscript as a word doc with BibTeX citation codes. This was then converted to markdown with pandoc with the BibTeX citation codes translated to pandoc citation codes, yaml headers added, and tables and figures interspersed within the markdown manuscript.

From there, the manuscript could be built into pdf and word formats. However, because of some unknown (and quite frustrating) errors in the bibliography not parsing some italicized names, we manually copied and fixed the bibilographic entries from the intermediate tex file into a new tex file of correctly formatted citations. The citations in the intermediate tex file are then replaced using the script replace-citatons.sh. Because pandoc's institute yaml header wasn't giving us an affiliation for Dr. Everhart and myself, we replaced the line in the intermediate tex file with the corrected affiliations.

Simulation code

While this is a review paper, we simulated 20 populations to generate figure 2:

Figure 2 showing twenty simulated clonal populations (large shaded circles) with no migration between them

The results of these simulations can be found in the output markdown file. If you would like to run these simulations yourself, you will need R and GNU make. Once you have these tools, you can type:

make

or, if you would rather have an html version, you can type:

make results/kamvar2017populations.html

This will install the internal simulation package kop (with dependencies), install the code dependencies and then run the simulation Rmarkdown file.

These simulations are individual-based forward-time simulations which will use 4 processing cores. It's not uncommon for the process to take 20 minutes to complete. For convenience, the results will be cached so subsequent runs will take considerably less time.