05 March 2014

regression and statistical inference

Next post Previous post

Intermediate-expression - biological interpretation


Completely re-working analysis, splitting thermally-responsive transcripts to examine differences among transcripts with colony x temperature interaction.

Nice tip on how to identify inflection point.

Found that my “Intermediate” expressed transcripts included those with the correct curvature, but that had maximum expression below (<10) or above (>31) the ‘intermediate’ interval. Felt rather foolish - this is exactly the issue with identifying stabilizing selection highlighted by Mitchell-Olds & Shaw (1987). At least I’m not the first (or last) to make this mistake.

Modified function to only call a transcript “Intermediate” if curvature is correct and the maximum is between 10 and 31C.

Interestly, also found that there’s an R function named after Mitchell-Olds & Shaw (1987) to check if maximum or minimum of quadratic function is within studied interval.


  • Thomas Mitchell-Olds, Ruth G. Shaw, (1987) Regression Analysis of Natural Selection: Statistical Inference And Biological Interpretation. Evolution 41 1149-NA 10.2307/2409084

Creative Commons Licence
This work is licensed under a Creative Commons Attribution 4.0 International License.