expdean

The experiment design and analysis package is meant for dealing with data analysis of plant science (Agriculture and Genetics) and allied field. This is born out of personal use necessicity, mainly.

Installation

You can install development version of the expdean package from Github with:

devtools::install_github("DeependraD/expdean")
library(expdean)

Agricultural research experiments analysis workflow

1. lm_list: Fit linear models for multiple response variables in a dataframe
2. corstars: Correlation coefficient table with significance annotation; Carried over with slight modification from: http://myowelt.blogspot.com/2008/04/beautiful-correlation-tables-in-r.html
3. grid_arrange_share_legend: Share legend and plot multiple ggplot grobs; Copied from: http://rpubs.com/sjackman/grid_arrange_shared_legend
4. mean_differences_lm, mean_differences_emmeans: Mean separation techniques for multiple. Some visualization techniques are also included as comment.
5. p_annotator: Model summary table for various model types. This implements tidy techniques for regression model summary table generation (Looks alike stargazer summary table).
6. long_stargazer: Model summary table with modified stargazer and some useful defaults.
7. Mixed modeling when analyzing common experimental designs require estimation of heritability, BLUEs, phenotypic correlation and genotypic correlation. A solution for tidy calculation and representation of such results is implemented in trait_genetics_tidy.R.

Checklist

1. Resolve covariate specification in trait_genetics_tidy.R.

• heritability_n_blues() function should accept NULL as well as arbitrary number of covariates.
• confirm the validity of heritability_n_blues() function in estimating heritability.

2. Currently only 3 models are tidied (in join step) due to column name mismatch in model_summary_kabletype.R's p_annotator() function.

• Use metaprogramming to solve it.

Building my first package

• When hosting a package remotely as a source repo we can confirm that the name we choose for the package is infact available by:

library(available)
available("expdean")

• usethis::use_roxygen_md() to use
• usethis::use_data(some_object_usually_df) to use object in current session as exported data.
• usethis::use_tidy_eval() to use rlang and NSE features without explicitly exporting each function
• devtools::document() to automatically generate roxygen documents and namespace file
• usethis::use_mit_license()