OCR-Stats

Contains functions to estimate oxygen consumption rates using the Seahorse XF Analyzer, and perform statistical testing between samples. Includes plotting functions as well. Manuscript in Plos ONE.

Installing requirements

R

install.packages("data.table")
install.packages("dplyr")
install.packages("ggplot2")
install.packages("ggthemes")
install.packages("magrittr")
install.packages("plotly")
install.packages("tidyr")

General repository organization notes

Sourcing the config.R file loads all necessary functions and variables. Go through the example.R which should give a direct idea of the functions and plots available.

There are 2 folders:

• functions: contains different functions needed for the statistical OCR-stats methods.
• plots: contains different plot functions. They are all created using ggplot, therefore they can be saved and edited.

R should be started from the repository root.

All data located in data/.

Main functions explained

• add_outlier_col(): adds 2 T/F columns (is.outw and is.out) to the given dataset indicating if the OCR value is a well level or single point outlier.
• compute_bionergetics(): computes all four OCR interval levels in natural and log scales. Also provides bioenergetics in the natural scale (eg. maximal respiration) and in the log scale (eg. M/Ei ratio)
• stat_test_OCR(): compares the bioenergetics of 2 samples providing an estimate with the difference and pvalue. Returns a list with 2 objects: dif_dt: for each pair of samples to be compared, gives the bioenergetics of each of them, and the respective difference; pv_dt: for each sample, returns one between-plates replicates aggregated difference wrt to a control and a pvalue.
• sh_plot(): plots a whole Seahorse experiment, differentiating samples by color. Can produce points, boxplots or violin plots. Returns a ggplot object that can be further edited.
• outlier_plot(): plots a single sample, highlighting outlier status. Returns a ggplot object that can be further edited.
• plot_bios(): plots the specified bioenergetics difference wrt to a control of all samples. Marks as red significant samples.
• sh_volcano(): makes a volcano plot, where the x axis is the bioenergetic difference wrt to a control and the y-axis the -log10 of the pvalue. Samples above the horizontal dotted line are significant.
• scatterplot_bios(): makes a scatterplot of 2 different bioenergetic differences.

Support

Let me know if you have any problems by creating an issue or sending me an email to yepez-at-in.tum.de.