Merge pull request #87 from egouldo/80-parameter-lookup-table

Close #80 param lookup tables for constructed variables

NAMESPACE

@@ -104,6 +104,7 @@ export(summarise_variable_counts)
 export(validate_predictions)
 export(validate_predictions_df_blue_tit)
 export(validate_predictions_df_euc)
+import(cli)
 import(dplyr)
 import(ggbeeswarm)
 import(ggplot2)
@@ -134,12 +135,11 @@ importFrom(dplyr,rename)
 importFrom(dplyr,right_join)
 importFrom(dplyr,select)
 importFrom(dplyr,summarise)
-importFrom(pointblank,col_vals_not_null)
-importFrom(pointblank,stop_if_not)
 importFrom(forcats,fct_relevel)
 importFrom(glue,glue)
 importFrom(magrittr,"%>%")
 importFrom(pointblank,col_vals_not_null)
+importFrom(pointblank,stop_if_not)
 importFrom(purrr,keep)
 importFrom(purrr,list_flatten)
 importFrom(purrr,list_rbind)

R/prepare_response_variables.R

@@ -5,8 +5,8 @@
 #' @param param_table A table of parameters \(mean, sd\) for *most* response variables used by analysts. This tibble is pulled from the named object exported by `ManyEcoEvo::`. but can be overwritten with the users's own `param_table` dataset.
 #'
 #' @return A tibble of nested list-columns
-#' @details Operates on nested list-columns of data
-#' @family targets-pipeline functions
+#' @details Operates on nested list-columns of dataframes, where each dataframe contains the response variable data for a single analysis. The function standardises the response variable data for each analysis, and returns the modified dataset to the `data` list-column. 
+#' @family targets-pipeline functions. 
 #' @family Multi-dataset Wrapper Functions
 #' @export 
 prepare_response_variables <- function(ManyEcoEvo, 

R/prepare_response_variables_yi.R

@@ -5,7 +5,7 @@
 #' @param param_table A table of parameters \(mean, sd\) for *most* response variables used by analysts. This tibble is pulled from the named object exported by `ManyEcoEvo::`. but can be overwritten with the users's own `param_table` dataset.
 #'
 #' @return A tibble of nested list-columns
-#' @details Operates on nested list-columns of data
+#' @details Operates on nested list-columns of data. The function back-transforms the response variables from the link to the response scale for each dataset in the ManyEcoEvo dataset. The back-transformed data is stored in a list-column called `back_transformed_data`. It is useful for when wanting to conduct a meta-analysis on the response scale, e.g. for the *Eucalyptus* count data.
 #' @family targets-pipeline functions
 #' @family Multi-dataset Wrapper Functions
 #' @export 

R/standardise_response.R

@@ -6,6 +6,9 @@
 #' @param dataset One of either "blue tit" or "eucalyptus"
 #'
 #' @return A tibble of analyst data with standardised values contained in a list-column called 'back_transformed_data'
+#' @details
+#' Note that for $y_i$ or out of sample predictions that are standardised, if param_table is `NA` or `NULL` for a given variable, then the response variable will not be standardised, and NA will be returned for that entry in `back_transformed_data`.
+#' 
 #' @export 
 #' @family analyst-data
 standardise_response <-  function(dat, 
@@ -78,6 +81,7 @@ standardise_response <-  function(dat,
                                   }else{NA}))
   }
 
+  #TODO for any analyses implicitly excluded, return a message to the user
   return(dat)
 
 }

README.md

@@ -64,6 +64,9 @@ targets::tar_read("ManyEcoEvo")
     1 blue tit   <tibble [174 × 38]> <tibble [174 × 54]> Zr           
     2 eucalyptus <tibble [128 × 38]> <tibble [128 × 61]> Zr           
 
+The script that generates the ManyEcoEvo package datasets is located in
+[`ManyEcoEvo/data-raw/tar_make.R`](https://github.com/egouldo/ManyEcoEvo/blob/main/data-raw/tar_make.R).
+
 ## License
 
 This software is licensed with the GNU GPL 3 license.

README.qmd

@@ -12,6 +12,8 @@ The `ManyEcoEvo` package provides a suite of functions for:
 -   Tidying and cleaning many-analyst style data for further analysis
 -   Reproducing the analysis in *Gould et al.* using your own many-analyst style data
 
+[^1]: Gould, E., Fraser, H., Parker, T. *et al.* (2023). Same data, different analysts: Variation in effect sizes due to analytical decisions in ecology and evolutionary biology \[Preprint\]. Ecology and Evolutionary Biology. https://doi.org/10.32942/X2GG62
+
 > Note that the manuscript source-code for *Gould et al.* is located in a separate repository at <https://github.com/egouldo/ManyAnalysts>, which can be viewed at <https://egouldo.github.io/ManyAnalysts/>.
 
 ## Installation
@@ -24,8 +26,7 @@ devtools::install_github("egouldo/ManyEcoEvo")
 
 # Regenerating the ManyEcoEvo dataset
 
-The data processing and analysis can be freely reproduced with the help of the [`targets::`](https://github.com/ropensci/targets) package.
-Please see the documentation at <https://docs.ropensci.org/targets/> for further detail.
+The data processing and analysis can be freely reproduced with the help of the [`targets::`](https://github.com/ropensci/targets) package. Please see the documentation at <https://docs.ropensci.org/targets/> for further detail.
 
 Should you wish to completely reproduce the dataset generation and analysis in *Gould et al.*, complete the following steps:
 
@@ -44,6 +45,8 @@ suppressPackageStartupMessages(library(targets))
 targets::tar_read("ManyEcoEvo")
 ```
 
+The script that generates the ManyEcoEvo package datasets is located in [`ManyEcoEvo/data-raw/tar_make.R`](https://github.com/egouldo/ManyEcoEvo/blob/main/data-raw/tar_make.R).
+
 ## License
 
 This software is licensed with the GNU GPL 3 license.
@@ -71,9 +74,3 @@ A BibTeX entry for LaTeX users is
     url = {https://github.com/egouldo/ManyEcoEvo},
   }
 ```
-
-[^1]: Gould, E., Fraser, H., Parker, T.
-    *et al.* (2023).
-    Same data, different analysts: Variation in effect sizes due to analytical decisions in ecology and evolutionary biology \[Preprint\].
-    Ecology and Evolutionary Biology.
-    https://doi.org/10.32942/X2GG62

_targets.R

@@ -242,6 +242,7 @@ list(tarchetypes::tar_file_read(name = euc_reviews,
                                                                               expertise_subset == "All",
                                                                               collinearity_subset == "All"))),
      tarchetypes::tar_quarto(name = README,
-                             path = "README.qmd")
-     # tarchetypes::tar_quarto(full_analysis, "analysis/analysis_revised_data.qmd")
-)
+                             path = "README.qmd"),
+     tarchetypes::tar_quarto(name = README_data_raw,
+                             path = here::here("data-raw/analysis_datasets/", "README.qmd"))
+)