Use of principal curves for analysis and summary of palaeoecological stratigraphic data
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Failed to load latest commit information.

Summarising stratigraphic data using principal curves

This is a public repository for a paper I am writing on the use of principal curves for summarising stratigraphic palaeoecological data. The paper is being written in markdown with embedded R code. knitr is used to covert the embedded R code to outputs used in the paper in a markdown source file. The pandoc() knitr function is then used to render the markdown source to PDF via LaTeX. This stage uses the default Pandoc LaTeX template currently, but I plan to modify that template to give nicer rendered output.

This is a work in progress and an experiment in writing a paper in the open using Github and the ideals of reproducible research; if you have any comments or suggestions uses the issues tab to make them.


To run the R code in the Rmd file you will need the development version of my analogue package. You can get this from R-forge; anything version >= 0.11-1 should be OK. To install this version from within R use

install.packages("analogue", repos="")

You'll see some warnings about scaling not being a graphical parameter if you don't have a fairly recent development version of the vegan package. This annoyance was fixed in revision r2533, but because of the silly build process on R-Forge even though vegan builds fine for us it fails there so you'll need to compile vegan yourself. The warning is totally harmless and nothing will change if you have the right version of vegan installed; the warning in the paper just looks odd and annoying.

You'll also need

  • knitr, Yihui Xie's excellent R package for processing embedded R code in documents (on CRAN)
  • vegan, our R package for multivariate & community ecology methods --- analogue depends on this so that shouldn't be an issue
  • coenocliner, my coenocline simulator package for R. coenocliner is not on CRAN yet, but can be installed using Hadley Wickham's devtools package. See the coenocliner on github for installation instructions.
  • Pandoc the wonderful universal document converter from John MacFarlane


The manuscript

The paper (the markdown source file and the resulting PDF prcurve_palaeo.pdf) are copyright © Gavin L. Simpson 2013, some rights reserved. You are free to use the text under the terms of the Creative Commons By-Attribution 3.0 Unported licence.

Creative Commons Licence
Summarising stratigraphic data using principal curves by Gavin L. Simpson is licensed under a Creative Commons Attribution 3.0 Unported License.

R Code

In so far as I can (it is simply interpretted code calling existing R functions in existing R packages), I waive all copyright and related or neighboring rights to the R Code embedded in the file prcurve_palaeo.Rmd.

To the extent possible under law, Gavin L. Simpson has waived all copyright and related or neighboring rights to R code for the paper "Summarising stratigraphic data using principal curves". This work is published from: Canada.

Compiling the Rmd source

I am using knitr to write the manuscript with embedded R code. To process the files into the final PDF, the following R code can be used

require("knitr") ## load the knitr package

render_markdown() ## set the output render hooks, allows Pandoc-flavour markdown

opts_chunk$set(dev = "pdf") ## prefer PDF figures

## process the R code into a final markdown source file
## assumes you are in the directory where the file is located

## process the final markdown source to PDF using Pandoc
pandoc("", format = "latex")