Skip to main content
Dryad logo

Data for: Testing for fitness epistasis in a transplant experiment identifies a candidate adaptive locus in Timema stick insects

Citation

Villoutreix, Romain et al. (2022), Data for: Testing for fitness epistasis in a transplant experiment identifies a candidate adaptive locus in Timema stick insects, Dryad, Dataset, https://doi.org/10.5061/dryad.dbrv15f2w

Abstract

Identifying the genetic basis of adaptation is a central goal of evolutionary biology. However, identifying genes and mutations affecting fitness remains challenging because a large number of traits and variants can influence fitness. Selected phenotypes can also be difficult to know a priori, complicating top-down genetic approaches for trait mapping that involve crosses or genome-wide association studies. In such cases, experimental genetic approaches, where one maps fitness directly and attempts to infer the traits involved afterward, can be valuable. Here, we re-analyse data from a transplant experiment involving Timema stick insects, where five physically clustered SNPs associated with cryptic body colouration were shown to interact to affect survival. Our analysis covers a larger genomic region than past work and revealed a locus previously not identified as associated with survival. This locus resides near a gene, Punch (Pu), involved in pteridine pigments production, implying that it could be associated with an unmeasured colouration trait. However, by combining previous and newly obtained phenotypic data, we show that this trait is not eye or body colouration. We discuss the implications of our results for the discovery of traits, genes, and mutations associated with fitness in other systems, as well as for supergene evolution.

Methods

Over 700 insects were collected from a single natural population (Angeles National Forest, CA, HF5 34° 15.584′ N, 118° 6.254′ W), on the host plant Mountain Mahogany (Cercocarpus sp.) from which we selected 437 healthy adults for use in the transplant experiment. We gave all selected individuals a unique id number, photographed them, gave them an individual mark on the ventral side using Sharpie pens (i.e., dots of different colour combinations) and released them back into the area they were collected from in one of two host-plants treatments (i.e., different host plant species dominating the vegetation in this population, details below). Before release, we took a leg (i.e., tissue sample) from each transplanted individual for DNA sequencing purposes (ddRAD). In the first treatment, we released 219 individuals onto isolated vegetation patches composed of intertwined plant individuals, one of each of two plant species (Ceanothus sp. and Adenostoma sp.; referred to as AC treatment hereafter). In the second treatment, we released 218 individuals onto an isolated Mountain Mahogany host plant (Cercocarpus sp.; referred to as MM treatment hereafter). We recaptured surviving individuals ~72h after release.

Following past work where we measured body colouration from photographs of insects used in the transplant experiment, we corrected raw photographs (i.e. .NEF format) taken during the experiment for temperature (set at 6150 °K) in the software RawTherapee (version 5.8; https://www.rawtherapee.com/) and exported them as JPEG images. We scored eye colouration from these JPEG images using ImageJ (version 1.52r; https://imagej.nih.gov/ij/) circling the right eye (when not possible we measured the left eye) with the polygon tools and using the Color Histogram add-on. Following past work, we measured the RGB colour channels (red, green and blue) and processed them to obtain RG and GB estimates (the ratio of red over green and the ratio of green over blue, respectively). As for body colouration, we therefore studied two eye colouration traits: the RG and GB estimates we described above.

Usage Notes

Data

timema_cristinae_1.3c2_braker_interproscan_predgene_and_funcann.wseq.gff3.bz2: annotation file for Timema cristinae reference genome 1.3c2.

mod_g_tchum_1.3c2.lgNA.excluded.geno: genotype file for all the individuals in the transplant experiment (both AC and MM treatments) containing genome wide information. Used in the body and eye colouration GWA analyses.

mod_g_tchum_AC_clean_LGNA_excluded.MelStripe.dsv: genotype file for individuals in the AC treatment (transplanted from MM to AC) containing information for the MelStripe locus only. Used in the gemma and LT-MAPIT analyses on survival.

mod_g_tchum_AC_clean_LGNA_excluded.geno: genotype file for individuals in the AC treatment (transplanted from MM to AC) containing genome wide information. Not used in any analysis but given for readers convenience.

lm.data.txt: genotype file for LT-MPPIT SNP outlier 1, 2, PCA 1 axis and their interactions. Used in the prediction of survival analysis.

pntest_fha2013.txt: genotype file for the linkage desiquilibrium analysis.

2019_Tchumash_epis_2022-01-12_Table4paper.xlsx: phenotype data. Survival, body and eye colouration data. Genomic prediction data for body and eye colouration.

Functions and scripts

01_gemma_bslmm.pl: function running the gemma BSLMM models

02_gemma_summary.pl: function summarizing the results accross MCMC chains.

03_gemma_sparse_formatting.pl: function formatting the output file for easy plotting.

LT-MAPIT-output_formatting.pl: function formatting the output files from LT-MAPPIT analyses.

select-MelStripe_geno_file.sh: selects SNPs within the MelStripe region from the file containing genome wide information.

generating_prediction_input file.R: generates the input file for the survival prediction analysis.

eye-body-colour-phenotypic-correlation_and_plots.R: computes the phenotypic correlations and associated plots.

genomic-correlations-script.R: computes the genomic correlations and associated plots.

gemma_bslmm_array.slurm: runs the gemma BSLMM analyses.

scripts/GWAs/GEMMA-BSLMM/colouration-traits/gemma_bslmm_relatedness-Matrix.sh: generates the relatedness matrix prior to GEMMA BSLMM analyses.

graphics_gemma_colour.R: generates the graphics for gemma BSLMM analyses on colouration traits.

graphics_gemma_survival.R: generates the graphics for gemma BSLMM analyses on survival.

gemma_predictions_bslmm_array.slurm: runs the gemma BSLMM analysis for genomic prediction of the colouration traits.

pheno.subsets.sh: subsets phenotypes prior to gemma prediction analyses.

calcLD.R: runs the linkage disequilibrium analysis in T. cristinae.

graphics_LT-MAPPIT.R: generates the graphics for LT-MAPPIT results.

survival-LT-mappit.MelStripe.Zach-filter.R: runs the LT-MAPPIT analysis on survival.

survival-LT-mappit.MelStripe.Zach-filter.slurm: slurm 'wrapper' script to run LT-MAPPIT R script.

epiAnalysis.R: runs the prediction of survival analysis.

Funding

European Research Council, Award: 770826