data and scripts

This repository contains the data and scripts to reproduce the results from:

Rzezniczak et al. (2023). Regulation at Drosophila’s Malic Enzyme highlights the complexity of transvection and its sensitivity to genetic background. Genetics 223(2). https://doi.org/10.1093/genetics/iyac181

Authors: Teresa Z. Rzezniczak, Mark T. Rzezniczak, Bruce H. Reed, Ian Dworkin, Thomas J.S. Merritt

data

transvection_rawdata.csv raw data file used for the QG analysis (see details below and in associated script).

transvection_rawdata_other.csv largely the same as above but with additional variables not used in the QG analysis (so here for posteriority).

pheno_men76.txt contains the conditional means for MEN activity for each DGRP line in the Men76 dataset. Generated from the mixed model analysis described below.

pheno_men86.txt contains the conditional means for MEN activity for each DGRP line in the Men86 dataset. Generated from the mixed model analysis described below.

covar_mengwas.txt contains covariates for both the Men76 and Men86 datasets and was used in the GWAS for both datasets:

• In this file, missing data points are represented by “-9”
• Wolbachia – Wolbachia presence/absence is indicated using 1 and 2, respectively.
• MEN_SNP1 indicates which Men nonsynonymous SNP at position is present in each DGRP line
• MEN_SNP2 indicated which Men nonsynonymous SNP at position is present in each DGRP line
• MenEx3 is the conditional mean of each DGRP line when crossed to MenEx3 line
• Inversion PC1-Inversion PC5: The top 5 principal components when PCA analysis was run using annotated inversions in the DGRP
• EigVal1-Eigval5 – The top 5 principal components for PCA analysis of population structure

analysis

GWAS

GWAS analysis was done in plink (see manuscript). Plink code can be found in the file PLINK_script.md in the scripts folder.

script_qqman.R contains the scripts used to create the Manhattan plot and QQ plots in the paper.

The simulation for shared sites and genes across the two data sets is in ..\scripts\HowManySNPsInCommon.R

Quantitative Genetic analyses

All other information on the Quantitative Genetic analysis is in the TransvectionQG_ID.Rmd script. All QG analysis stems from the dataset in ../data/transvection_rawdata.csv

ID made a small edit to the raw data, removing "NA" so it can be added in more easily in R.

Misc folder

Contains intermediate files for plotting and GWAS (see TransvectionQG_ID.Rmd):

• MenEx_DGRP_conditionalMeans.csv (conditional means/BLUPs from mixed model analysis)

• Men_76_ratio_post.csv (posterior mean and 95% Credible intervals for "ratio plots")

• Men_86_ratio_post.csv (posterior mean and 95% Credible intervals for "ratio plots")

• AssortedValuesForPaperFromID.xlsx (exactly what it sounds like).

outputs

All QG figures can be generated from R scripts.

note these two files had to be removed from the github repo as they were too big. They can be regenerated using PLINK.

mengwas76.assoc.linear PLINK GWAS output for Men76

mengwas86.assoc.linear PLINK GWAS output for Men86