Hey there,
I am currently working on the statistics of counted and expected genotypes in mice litter of a mutant mouse model. I have two different mutated genes, both alone and in combination, one of them also heterozygously and homozygously mutated, which results in a total of 6 different possible genotypes. In three of these combinations, a lot of mice seem to die prenatally or before we can count and genotype them as the numbers of mutants counted are a lot less than expected by Mendelian ratio. I know that I can perform a Chi Square Test to test if there is a discrepancy between expected and counted values, but this only gives out one p value for the whole table with all genotypes. My question now is if there might be a way to test for each genotype individually if the significance between counted and expected is significant.
Could the Chi or Fisher's exact test be used for that in any way?
I also have expected/counted values available for each individual litter if this might be of any help. I appreciate your help and ideas!
Best,
Carolin
You can use a negative binomial model to analyze your data.
In such a model, you can include "litter" as a random intercept to account for correlations within litter.
If embryos die before you can get their genotyes, you measure the number of viable genotypes only (the proportion of animals with a genotype is also depending on the embryonic viability of that genotype). So you cannot expect to find a Mendelian pattern. If a genotype is associated with (embryonic) survival, this is not the same as the number of genotypes that are the result of parental recomnination.
Maybe you can use the Cochran-Mantel-Heanszel test, with genotype as 'strata' or controling categorial variable. See:
http://www.biostathandbook.com/cmh.html
R: https://rcompanion.org/rcompanion/b_10.html & the R script in attachment, which includes a usefull figure.
SAS:
DATA lap; INPUT location $ habitat $ allele $ count; DATALINES; Tillamook marine 94 56 Tillamook estuarine 94 69 Tillamook marine non-94 40 Tillamook estuarine non-94 77 Yaquina marine 94 61 Yaquina estuarine 94 257 Yaquina marine non-94 57 Yaquina estuarine non-94 301 Alsea marine 94 73 Alsea estuarine 94 6 Alsea marine non-94 71 Alsea estuarine non-94 79 Umpqua marine 94 71 Umpqua estuarine 94 48 Umpqua marine non-94 55 Umpqua estuarine non-94 48 ; PROC FREQ DATA=lap; WEIGHT count / ZEROS; TABLES location*habitat*allele / CMH; RUN;