• Copy and pasted data into Word/Excel tables?

• Entered data manually into a table?

• Used text boxes to customize plots in Word?

• Perform an analysis and create a Word table with the results at the same time?

• Avoid errors resulting from copying & pasting or manual data entry?

• Make insightful and pretty plots without a single drag-and-click?

The term reproducible research refers to the idea that the ultimate product of academic research is the paper along with the full computational environment used to produce the results in the paper such as the code, data, etc. that can be used to reproduce the results and create new work based on the research.

Source: en.wikipedia.org/wiki/Reproducibility#Reproducible_research

Today, I will introduce two packages in R to make beautiful, reproducible reports:

• odfWeave

• ggplot2

Reproducible Reports With odfWeave

odfWeave allows you to integrate text and code in a single document.

How it works:

1. Write your documents in odt or odf format.

2. Process the document with the function odfWeave

3. The output is a odf file with text, code, and output.

Open document format files are a "universal" file format which can be read by:

• Microsoft Office (proprietary, Windows and Mac OS)

• OpenOffice (free, open-source, multi-platform)

• LibreOffice (free, open-source, multi-platform)

• In fact, most any text editor

1. Install R from www.r-project.org.

2. Install package odfWeave.

3. Create odt file with text and code chunks.

4. Load odfWeave and process file with function odfWeave.

install.packages("odfWeave")  # Install Package

library(odfWeave)  # Load Package

• The source file for odfWeave can mix text and code.

• Write text as usual.

• Code chunks are denoted with a starting <<>>= and closing @.

• Inline code is enclosed in \Sexpr{}.

• Chunks and inline R care are written as usual.

In this example, we will see how to get started with odfWeave.

1. Save the file (for example, source.odt)

2. In R, run the command odfWeave, indicating the file (source.odt) and dest (e.g., output.odt)

If we save our example as iris.odt, the following command will output the result as an odt document named iris_output.odt to the working directory.

odfWeave("iris.odt", "iris_output.odt")

Now we can open iris_output.odt in our text editor.

Visualizing Data With ggplot2

• Plots begin by specifying data and aesthetics

ggplot( data, aes(y, x, ymin, ymax))

• Add layers named geoms using `+' to specify how to represent the data
• Add layers named scales using `+' to modify the cosmetics of the plotted data

1. Change symbol size (size)

2. Change group colours (scale_color_manual)

3. Use theme to modify settings for the following plot elements:

   • legend.title
   • legend.key
   • panel.background

library(RColorBrewer)  # Package with many color palettes

Blues <- brewer.pal(9, "Blues")[c(5, 7, 9)]

ggplot(data = iris, aes(y = Sepal.Length, x = Petal.Width, colour = Species)) + 
    geom_point(size = 2) + scale_colour_manual(values = Blues) + 
    theme(legend.title = element_blank(), 
    	  legend.key = element_blank(), 
    	  panel.background = element_rect(fill = brewer.pal(8, "Pastel2")[4]))

• We can shows groups in different elements of a grid (rows or columns) using facet.grid

• Almost any kind of summary of the plot's data frame can be added using stat_summary functions.

ggplot(data = iris, aes(y = Sepal.Length, x = Petal.Width, colour = Species)) + 
    facet_grid(~Species) + 
    geom_point(size = 2) + 
    stat_smooth(method = "lm") + 
    scale_colour_manual(values = Blues) + 
    theme(legend.title = element_blank(), 
    	  panel.background = element_rect(fill = brewer.pal(8, "Pastel2")[4]))

• odfWeave

• ggplot2