FielDHub

FielDHub: A Shiny App for Design of Experiments in Life Sciences - Published in JOSS (2021)

https://github.com/didiermurillof/fieldhub

Keywords

Repository

FielDHub is an R Shiny design of experiments (DOE) app that aids in the creation of traditional, unreplicated, augmented and partially replicated (p-rep) designs applied to agriculture, plant breeding, forestry, animal and biological sciences.

README.Rmd

---
output: github_document
---



```{r, include = FALSE}
library(FielDHub)
knitr::opts_chunk$set(
  collapse = TRUE,
  comment = NA,
  fig.path = "man/figures/README-",
  out.width = "100%"
)
```




# FielDHub 


[![CRAN status](https://www.r-pkg.org/badges/version/FielDHub)](https://cran.r-project.org/web//packages/FielDHub/)
[![R-CMD-check](https://github.com/DidierMurilloF/FielDHub/actions/workflows/R-CMD-check.yaml/badge.svg)](https://github.com/DidierMurilloF/FielDHub/actions/workflows/R-CMD-check.yaml)
[![Lifecycle: experimental](https://img.shields.io/badge/lifecycle-stable-brightgreen.svg)](https://lifecycle.r-lib.org/articles/stages.html)
[![cranlogs](https://cranlogs.r-pkg.org/badges/FielDHub)](https://cranlogs.r-pkg.org/badges/FielDHub)
[![cranlogs](https://cranlogs.r-pkg.org/badges/grand-total/FielDHub)](https://cranlogs.r-pkg.org/badges/grand-total/FielDHub)



## A Shiny App for Design of Experiments in Life Sciences

## Installation

### From CRAN

``` r
install.packages("FielDHub")
```

### From GitHub

``` r
remotes::install_github("DidierMurilloF/FielDHub")
```

## FielDHub Paper

[![DOI](https://joss.theoj.org/papers/10.21105/joss.03122/status.svg)](https://doi.org/10.21105/joss.03122)

## Overview

A shiny design of experiments (DOE) app that aids in the creation of traditional, un-replicated, augmented and partially-replicated designs applied to agriculture, plant breeding, forestry, animal and biological sciences. 

For more details and examples of all functions present in the FielDHub package. Please, go to .



## Usage

This is a basic example which shows you how to launch the app:

``` r
library(FielDHub)
run_app()
```
### Diagonal Arrangement Example

A project needs to test 280 genotypes in a field containing 16 rows and 20 columns of plots. In this example, these 280 genotypes are divided among three different experiments. In addition, four checks are included in a systematic diagonal arrangement across experiments to fill 40 plots representing 12.5% of the total number of experimental plots. An option to include filler plots is also available for fields where the number of experimental plots does not equal the number of available field plots.



The figure above shows a map of an experiment randomized along with multiple experiments (three) and checks on diagonals. Distinctively colored check plots are replicated throughout the field in a systematic diagonal arrangement.



The figure above shows the layout for the three experiments in the field. 

### Using the FielDHub function `diagonal_arrangement()`

To illustrate using FielDHub to build experimental designs through R code, the design produced in the R Shiny interface described above can also be created using the function `diagonal_arrangement()` in the R script below. Note, that to obtain identical results, users must include the same random seed in the script as was used in the Shiny app. In this case, the random seed is 1249. 

```{r}
diagonal <- diagonal_arrangement(
  nrows = 16, 
  ncols = 20, 
  lines = 280, 
  checks = 4, 
  plotNumber = 101, 
  splitBy = "row", 
  seed = 1249, 
  kindExpt = "DBUDC", 
  blocks = c(100, 100, 80), 
  exptName = c("Expt1", "Expt2", "Expt3")
)
```

Users can print the returned values from `diagonal_arrangement()` as follow,

```{r}
print(diagonal)
```

First 12 rows of the field book,

```{r}
head(diagonal$fieldBook, 12)
```

Users can plot the layout design from `diagonal_arrangement()` using the function `plot()` as follows,

```{r, fig.align='center', fig.width=6.8, fig.height=4.7}
plot(diagonal)
```

In the figure, salmon, green, and blue shade the blocks of unreplicated experiments, while distinctively colored check plots are replicated throughout the field in a systematic diagonal arrangement.


The main difference between using the FielDHub Shiny app and using the standalone function `diagonal_arrangement()` is that the standalone function will allocate filler only if it is necessary, while in Shiny App, users can customize the number of fillers if it is needed. In cases where users include fillers, either between or after experiments, the Shiny app is preferable for filling and visualizing all field plots.


To see more examples, go to 

### Partially Replicated Design Example

Partially replicated designs are commonly employed in early generation field trials. This type of design is characterized by replication of a portion of the entries, with the remaining entries only appearing once in the experiment. As an example, considered a field trial with 288 plots containing 75 entries appearing two times each, and 138 entries only appearing once. This field trials is arranged in a field of 16 rows by 18 columns.



In the figure above, green plots contain replicated entries, and the other plots contain entries that only appear once.

### Using the FielDHub function `partially_replicated()`

Instead of using the Shiny FielDHub app, users can use the standalone FielDHub function `partially_replicated()`. The partially replicated layout described above can be produced through scripting as follows. As noted in the previous example, to obtain identical results between the script and the Shiny app, users need to use the same random seed, which, in this case, is 77.

```{r}
pREP <- partially_replicated(
  nrows = 16, 
  ncols = 18,  
  repGens = c(138,75),
  repUnits = c(1,2),
  planter = "serpentine", 
  plotNumber = 1,
  exptName = "ExptA",
  locationNames = "FARGO",
  seed = 77
)
```


Users can print returned values from `partially_replicated()` as follows,

```{r}
print(pREP)
```

First 12 rows of the fieldbook,

```{r}
head(pREP$fieldBook, 12)
```

Users can plot the layout design from `partially_replicated()` using the function `plot()` as follows,

```{r, fig.align='center', fig.width=6.8, fig.height=4.7}
plot(pREP)
```

To see more examples, please go to 

Owner

JOSS Publication

FielDHub: A Shiny App for Design of Experiments in Life Sciences

Published

May 18, 2021

DOI

10.21105/joss.03122

Volume 6, Issue 61, Page 3122

Authors

Didier A. Murillo
North Dakota State University

Salvador A. Gezan
VSN International

Ana M. Heilman
North Dakota State University

Thomas C. Walk
North Dakota State University

Johan S. Aparicio
CIAT (International Center for Tropical Agriculture)

Richard D. Horsley
North Dakota State University

Editor

Charlotte Soneson

Tags

Shiny DOE Unreplicated Designs Partially-Replicated Designs Plant Breeding

View PDF Review Thread Software Archive

GitHub Events

Committers

Last synced: 7 months ago

Issues and Pull Requests

Last synced: 6 months ago

Packages