Problem using hierfstat with genind object about hierfstat

[suzkel]

AHA It does work with a subset.
I have some individuals who were not assigned to a population, leading to the error.

Thank you for the suggestion!!

[zkamvar]

One thing to note is that your data may be malformed.

You mention:

I have a data set of 4451 individuals and 473 SNPs.

But your data show:

4,386 individuals; 473 loci; 946 alleles

which indicates that you have lost 65 samples somewhere.

Moreover, the error here indicates that you have missing data in your ploidy definitions:

Error in if (any(object@ploidy < 1L)) { :
missing value where TRUE/FALSE needed

Also, given that you mention that specific data set, I'm assuming you are following http://popgen.nescent.org/StartSNP.html as an example?

• Do you get the same error on a sub-set of your data?
• Can you provide a reproducible example?
• What does the result of unique(ploidy(myData)) look like?

The above may better help us figure out what's going on.

[zkamvar]

Excellent! Thank you for posting a description of the solution :)

If it helps, I would also recommend to have sanity checks throughout your script so that if anything changes, you'll know when it happened:

## read in data and format it --------------------
## ...

# Number of individuals expected
stopifnot(nInd(dat) == 4451)

# Number of loci expected
stopifnot(nLoc(dat) == 473)

# Number of populations expected
stopifnot(nPop(dat) == 9) 

## run analyses here -----------------------------
## ...