Kolmogorov-Smirnov Test and Shapiro-Wilk Test statistics in Proteomics

Question

[osamashehzad.786]

I am trying to understand the application of the following two statistic methods in analyzing protein expression profile in cancer cells vs normal cells:

1.Kolmogorov-Smirnov Test

2.Shapiro-Wilk Test

While I understand the geeky mathematical definitions of the two tests, it has become challenging to understand and disseminate the meaning and application of these two tests to normal biology students.

We are measuring expression profiles of 5 different proteins of interest in 6 samples (A, B, C, D, E, F using mass-spec (Orbitrap Fusion).

A, B, C: normall cells

D, E, F: cancer (glioblastoma cells).

What do the two statistic test mean if we perform them on our data? What do they tell us, why should we use them, and how are the two different from each other?

Answer 1

[andrew.j.skelton73]

This seems like you want to understand why these "geeky" mathematical principles are relevant to statistical modelling of biological data. This is beyond proteomics, but more orientated to fundamental assumptions of modelling. I'd recommend reading Wolfgang Huber & Susan Holmes' book, Modern Statistics for Modern Biology (freely available online). To start with, I'd look up the difference between parametric and non-parametric statistics.