Dear All,
I have data about the expression of genes (binary data being present or absent by PCR) in 500 isolate of certain bacteria from different reservoirs (e.g. humans, cattle, sheep, and wild birds) taken at the same place. I would like to use a model to inform which reservoir can be the source of human infection. in another word, I need a model that I can run to show the relationship between these reservoirs based on the expression profile of the candidate genes.
Could any one suggest ? ...
Hello,
In my opinion, a simple cross-tabulation ,assisted by a chi-square test of independence will suffice. Or you can use a logistic regression model; with a depndent (output) variable is expression ( absent=0,present=1), and the explantory variable is (reservoir).
An ANOVA test is impossible, because your dependent variable isn't normally distributed .
CHELLAI.F
Logistic regression (binary response varaible) is used typically, also Chi Square makes sense.
Thanks all. I have tried the tree model and worked. will try a logistic regression
- Badges
- Science topic
- Similar topics
- Computer Science and Engineering
- Bioinformatics
- Bioinformatic Tools