example_data1


The spectral counts matrix has samples in the columns, and proteins in the rows.
Each sample consists in 500ng of standard yeast lisate spiked with 100, 200, 400 and 600fm of a mix of 48 equimolar human proteins (UPS1, Sigma-Aldrich). The dataset contains a different number of technical replicates of each sample.
This dataset has only 100 proteins of total 685 proteins in the original data for fast example execution.  If you want to use whole dataset, go for 'example_data2'.


datasets

Functions using Cohen's effect sizes (Cohen, Jacob. Statistical power analysis for the behavioral sciences. Academic press, 2013) are provided for reliable feature selection in biology data analysis.  In addition to Cohen's effect sizes, p-values are calculated and adjusted from quasi-Poisson GLM, negative binomial GLM and Normal distribution ANOVA.  Significant features (genes, RNAs or proteins) are selected by adjusted p-value and minimum Cohen's effect sizes, calculated to keep certain level of statistical power of biology data analysis given p-value threshold and sample size.

Lang Ho Lee

selfea

Select Features Reliably with Cohen's Effect Sizes

example_data1 function

Two data frames, df_contrast (protein expression profile) and df_group (experiment group information).

example_data1: Gregori Data: Yeast lisate samples spiked with human proteins

Description

Usage

Arguments

Format

References

Examples