Software

Code for most lab projects is available either on the lab GitHub page or my personal GitHub page.

Here are some R packages that are potentially of use.

abd: Analysis of Biological Data

Collaborating with Randall Pruim (Calvin College), I co-developed an R package (abd) to accompany the first edition of The Analysis of Biological Data by Michael Whitlock and Dolph Schluter. This is an excellent textbook (now in its second edition, still remarkably inexpensive) for teaching advanced undergraduate and introductory graduate statistics. The new edition of this book has associated R code, mostly making this package redundant. But I still like having the datasets from the book handy. I use them for teaching and demos really often, like the distribution of counts of desert birds or hemoglobin levels in different populations.

library(ggplot2)
library(abd)
ggplot(DesertBirds, aes(count)) +
  geom_histogram(bins = 30, fill = "darkred") +
  labs(x = "Count", y = "Number of Species") +
  cowplot::theme_cowplot()

library(ggplot2)
library(abd)
ggplot(Hemoglobin, aes(x = hemoglobin, y = relative.frequency)) +
  geom_bar(stat = "identity", fill = "darkred") +
  labs(x = "Hemoglobin", y = "Relative Frequency") +
  facet_grid(group ~ .) +
  cowplot::theme_cowplot()

Installation

The package can be installed via CRAN

install.packages("abd")

or directly from GitHub:

remotes::install_github("Middleton-Lab/abd")

We have had a lot of dependency creep over the years, so the number of packages that get installed alongside abd can be quite large.

abdData

The abd package has quite a few dependencies resulting from its use of the mosaic package.

The data from abd is arguably its most useful feature. A more streamlined version of abd with only the data is available via abdData, which can be installed with

remotes::install_github("Middleton-Lab/abdData")

The abdData user manual is available online.

jnt: Johnson-Neyman Technique

The Johnson-Neyman technique is used to determine the region of an analysis of covariance where the slopes are not significantly between the covariate. This allows the determination of bounds of a region of “equal” slope despite the presence of a “significant” slope (by whatever definition of “significant” you choose).

We used this approach in Lavin SR, Karasov WH, Ives AR, Middleton KM, Garland T Jr (2008) Morphometrics of the avian small intestine compared with that of nonflying mammals: A phylogenetic approach. Physiol Biochem Zool 81: 526-550.

References

  1. Johnson PO and Neyman J (1936) Tests of certain linear hypotheses and their application to some educational problems. Statistical Research Memoirs 1: 57-93.
  2. Hunka S and Leighton J (1997) Defining Johnson-Neyman regions of significance in three-covariate ANCOVA using Mathematica. Journal of Educational and Behavioral Statistics 22: 361-387.
  3. White CR (2003) The influence of foraging mode and arid adaptation on the basal metabolic rates of burrowing mammals. Physiol Biochem Zool 76: 122-134.
  4. White CR (2003) Allometric analysis beyond heterogenous regression slopes: Use of the Johnson-Neyman Technique in comparative biology. Physiol Biochem Zool 76: 135-140.

Demo

library(jnt)
Loading required package: smatr
set.seed(1234)

n <- 50

x1 <- rnorm(n)
y1 <- x1 + rnorm(n, sd = 0.2)

x2 <- rnorm(n)
y2 <- 1.25 * x2 + rnorm(n, sd = 0.2)

df1 <- data.frame(x = x1, y = y1)
df2 <- data.frame(x = x2, y = y2)

(jnt.out <- jnt(df1, df2))
Fitting with OLS
Assuming x variable is column 1, and y is column 2.

Johnson-Neyman Technique

Alpha =  0.05 

Data 1:
    Slope        0.9853 
    Intercept    0.02126 

Data 2:
    Slope        1.182 
    Intercept    0.01386 

Region of non-significant slope difference
    Lower: -0.5258 
    Upper: 0.8701 
plot(jnt.out)
`geom_smooth()` using formula = 'y ~ x'

Installation

jnt can be installed directly from GitHub:

remotes::install_github("Middleton-Lab/jnt")

MuscleTernary: Ternary plots for 3D data

Installation

MuscleTernary can be installed directly from GitHub:

remotes::install_github("Middleton-Lab/MuscleTernary")

binning: Binning for mouse wheel activity data

Installation

binning can be installed directly from GitHub:

remotes::install_github("Middleton-Lab/binning")