AMOVA and HOMOVA design file set-up

Commands in mothur
1

Good Day Forum,

I’m trying to make sure my design files are set-up properly. I have a Diet x Treatment x Sex factorial experiment and have come to the conclusion I have to analyze each factor separately for AMOVA and HOMOVA. Below are examples of my codes.

amova(phylip=colon.an.thetayc.0.03.lt.ave.dist, design=mouse.diet.design)
amova(phylip=colon.an.thetayc.0.03.lt.ave.dist, design=mouse.treatment.design)
homova(phylip=colon.an.thetayc.0.03.lt.ave.dist, design=mouse.diet.design)
homova(phylip=colon.an.thetayc.0.03.lt.ave.dist, design=mouse.treatment.design)

Here are examples of my set-up. Can someone confirm if I’m on the right track with my set-up in the file?

group diet
1Colon Sucrose
2Colon Sucrose
3Colon Fructose
4Colon Fructose
5Colon Malto
6Colon Malto

group treatment
1Colon PBS
2Colon PBS
3Colon Shigella
4Colon Shigella
5Colon PBS
6Colon Shigella

2

That looks right for using amova and homova in mothur. If you’re an R-user you should try adonis in the vegan package as a replacement for amova. It can handle the two-factor design. If you go that route you will need to provide a formatted distance matrix like so, otherwise Bad Things Will Happen…

dist <- read.table(file=“my_square_distance_matrix.dist”, header=F, row.names=1)
thetayc <- as.dist(dist)
thetayc.adonis <- adonis(thetayc~md$SP*md$Sex)

3

Thanks for confirming.