GithubHelp home page GithubHelp logo

malballinger / ballinger_amnat_2022 Goto Github PK

View Code? Open in Web Editor NEW
0.0 1.0 1.0 52.11 MB

Data, code, and analyses associated with Ballinger & Nachman (2022).

Home Page: https://www.journals.uchicago.edu/doi/10.1086/719028

License: Other

R 65.19% Makefile 1.48% TeX 33.33%

ballinger_amnat_2022's Introduction

The contribution of genetic and environmental effects to Bergmann’s rule and Allen’s rule in house mice

Link to our paper published in The American Naturalist: DOI: 10.1086/719028

See our bioRxiv preprint here: DOI:10.1101/2021.06.14.448454

Data and code have been deposited on Zenodo: DOI

Abstract:

Distinguishing between genetic, environmental, and genotype-by-environment effects is central to understanding geographic variation in phenotypic clines. Two of the best-documented phenotypic clines are Bergmann’s rule and Allen’s rule, which describe larger body sizes and shortened extremities in colder climates, respectively. Although numerous studies have found inter- and intraspecific evidence for both ecogeographic patterns, we still have a poor understanding of the extent to which these patterns are driven by genetics, environment, or both. Here, we measured the genetic and environmental contributions to Bergmann’s rule and Allen’s rule across introduced populations of house mice (Mus musculus domesticus) in the Americas. First, we documented clines for body mass, tail length, and ear length in natural populations, and found that these conform to both Bergmann’s rule and Allen’s rule. We then raised descendants of wild-caught mice in the lab and showed that these differences persisted in a common environment and are heritable, indicating that they have a genetic basis. Finally, using a full-sib design, we reared mice under warm and cold conditions. We found very little plasticity associated with body size, suggesting that Bergmann’s rule has been shaped by strong directional selection in house mice. However, extremities showed considerable plasticity, as both tails and ears grew shorter in cold environments. These results indicate that adaptive phenotypic plasticity as well as genetic changes underlie major patterns of clinal variation in house mice and likely facilitated their rapid expansion into new environments across the Americas.

Overview:

Ballinger_allenbergmann_AmNat_2021
|- README          # top level description of content
|
|- data            # raw and processed data, are not changed once created
|  |- raw/         # raw data, will not be altered
|  |- processed/   # cleaned data, will not be altered once created
|
|- code/           # programmatic code for cleaning, modeling, and plotting
|
|- results/        # all output from workflows and analyses
|  |- tables/      # tables, designated for supplementary material
|  |- figures/     # graphs, designated for manuscript figures
|
|- submission      # files for manuscript
|
|- Makefile        # executable Makefile for this study

Dependencies:

  • R version 4.1.1 (2021-08-10)
    • tidyverse (v. 1.3.1)
    • rmarkdown (v. 2.11)
    • here (v. 1.0.1)
    • readxl (v. 1.3.1)
    • cowplot (v. 1.1.1)
    • performance (v. 0.8.0)
    • see (v. 0.6.8)
    • car (v. 3.0.12)
    • tinytex (v. 0.35)

My computer

## R version 4.1.1 (2021-08-10)
## Platform: x86_64-apple-darwin17.0 (64-bit)
## Running under: macOS Big Sur 10.16
## 
## Matrix products: default
## BLAS:   /Library/Frameworks/R.framework/Versions/4.1/Resources/lib/libRblas.0.dylib
## LAPACK: /Library/Frameworks/R.framework/Versions/4.1/Resources/lib/libRlapack.dylib
## 
## locale:
## [1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8
## 
## attached base packages:
## [1] stats     graphics  grDevices utils     datasets  methods   base     
## 
## other attached packages:
##  [1] here_1.0.1      rmarkdown_2.11  forcats_0.5.1   stringr_1.4.0  
##  [5] dplyr_1.0.7     purrr_0.3.4     readr_2.1.1     tidyr_1.1.4    
##  [9] tibble_3.1.6    ggplot2_3.3.5   tidyverse_1.3.1
## 
## loaded via a namespace (and not attached):
##  [1] tidyselect_1.1.1 xfun_0.28        haven_2.4.3      colorspace_2.0-2
##  [5] vctrs_0.3.8      generics_0.1.1   htmltools_0.5.2  yaml_2.2.1      
##  [9] utf8_1.2.2       rlang_0.4.12     pillar_1.6.4     glue_1.5.1      
## [13] withr_2.4.3      DBI_1.1.1        dbplyr_2.1.1     modelr_0.1.8    
## [17] readxl_1.3.1     lifecycle_1.0.1  munsell_0.5.0    gtable_0.3.0    
## [21] cellranger_1.1.0 rvest_1.0.2      evaluate_0.14    knitr_1.36      
## [25] tzdb_0.2.0       fastmap_1.1.0    fansi_0.5.0      broom_0.7.10    
## [29] Rcpp_1.0.7       scales_1.1.1     backports_1.4.0  jsonlite_1.7.2  
## [33] fs_1.5.2         hms_1.1.1        digest_0.6.29    stringi_1.7.6   
## [37] rprojroot_2.0.2  grid_4.1.1       cli_3.1.0        tools_4.1.1     
## [41] magrittr_2.0.1   crayon_1.4.2     pkgconfig_2.0.3  ellipsis_0.3.2  
## [45] xml2_1.3.3       reprex_2.0.1     lubridate_1.8.0  rstudioapi_0.13 
## [49] assertthat_0.2.1 httr_1.4.2       R6_2.5.1         compiler_4.1.1

ballinger_amnat_2022's People

Contributors

malballinger avatar malloryballinger avatar

Watchers

 avatar

Forkers

nachmanlab

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.