#Morphometric analysis of Ficinia specimens

This repo contains the analysis for Tshiila et al. 2017. Two new species in the Ficinia indica complex..., which has been published in Phytotaxa: http://dx.doi.org/10.11646/phytotaxa.295.1.4

Analyst: Jack Viljoen
Date: 2016-08-08
Platform: x86_64-pc-linux-gnu

The analysis was run in the Jupyter notebook Analysis_notebook.ipynb.

The analysis produced the following results:

• ANOVA on univariate variables
  • F stats & Tukey’s HSD test
  • strip plots
• Discriminant functions analysis
  • discriminant coefficients
  • plots of specimen scores on DF axes
  • Pillai’s trace and significance of species classification on DF axes

#Procedure

If you wish to run the code yourself …

1. Install Miniconda from Continuum (http://conda.pydata.org/miniconda.html).

2. Install R and required packages, as well as Jupyter notebook and IRkernel:

   conda create -c r -n r-notebook notebook r-irkernel r-ggplot2 r-hmisc
   source activate r-notebook
   

   Alternatively, to create a Conda environment with the same versions of packages I used, edit the path in environment.yml and run:

   conda create -f environment.yml
   source activate r-notebook
   

3. Clone this repo and enter the directory.

   git clone https://git.sr.ht/~javiljoen/tshiila2017-analysis
   cd tshiila2017-analysis
   

4. Start the Jupyter server and re-run the analysis:

   jupyter notebook
   

#Dependencies

• Conda
• Jupyter notebook with IRkernel
• R v. 3.3.1 with packages:
  • reshape2
  • RColorBrewer
  • ggplot2
  • Hmisc
  • MASS
• optional: Ghostscript (for pdf2ps conversion)

Versions of analysis software used:

R version 3.3.1 (2016-06-21)
Platform: x86_64-pc-linux-gnu (64-bit)
Running under: Ubuntu 16.04.1 LTS

attached packages:

    MASS_7.3-45
    Hmisc_3.17-4
    Formula_1.2-1
    survival_2.39-4
    lattice_0.20-33
    ggplot2_2.1.0
    reshape2_1.4.1