There are so many papers on population genetics. And when you are talking about genetic diversity, you can use some software if you have data in your hand.

@ Shamsunnahar Mukta would you mind pointing to some papers and software that will make the search to the answer easier?

https://www.nature.com/articles/s41598-020-63973-x

https://academic.oup.com/aobpla/article/doi/10.1093/aobpla/plw002/2609507

There are so many other papers available.

Data analysis depends on the sample and the objective.

Normally test for variance is

* 1 sample case you can Chi-squre test

* for comparing variance/heterogenicity of two groups you can apply F- test.

Test for equality of means

* For checking the equality of two means you can apply t-test.

* ANOVA for means of 2 or more groups.

and MEGA or relevant software can be used for genetic diversity analysis.

https://github.com/laninsky/genetic_diversity_diffs

Thanks, Shamsunnahar Mukta for your answer! I used MEGA to align the sequences and it's helpful to construct a phylogenetic tree. Is there any way to conduct t-test for the mean genetic diversity of two populations using MEGA?

I'll check out the GitHub link that you provided.

Regards,

Linh

As far as I used MEGA, I failed to find this test. But there is other software available.

Thank Shamsunnahar Mukta for your response!

I could perform AMOVA in Arlequin with sequence data, but still wondering whether it is enough to say that my two populations were different in genetic diversity LEVEL (my objective is to test if A population is more diverse than B, for example, whether population A had haplotype diversity = 0.18 is significantly more diverse than B population that had Hd = 0.15).

I know some statistical software can perform t-test or ANOVA but don't know the way to conduct these kinds of test with sequence data. Do you have experience with this?

By the way, genetic_diversity_diffs function introduced in the GitHub link doesn't work in my case.

Regards,

Linh

Hi guys, Thanks for your concerns!

Even though this problem seems only happened to me due to my shallow knowledge in this field. But in case someone looking for the answer to a similar question. Below is my solution.

First, I calculated the haplotype diversity (and nucleotide diversity as well) for two of my populations (by using dnaSP). Then I got the mean values of Hd and their standard deviation values. Now, I can compare these two means by using tsum.test function in "BSDA" R package (t-test based on summary data). Below is the R code.

-----

install.packages("BSDA") #install package BSDA from CRAN

library(BSDA) #load package BSDA

tsum.test(mean.x = NULL, s.x = NULL, n.x = NULL, mean.y = NULL, s.y = NULL,

n.y = NULL, alternative = "two.sided", mu = 0, var.equal = FALSE,

conf.level = 0.95)

------

#Now, just put my values into the code line and RUN.

#mean.x = single number representing the mean of x population

# a single number representing the sample standard deviation for x

#n.x = a single number representing the sample size for x

# mean.y= a single number representing the sample mean of y

#s.y = a single number representing the sample standard deviation for y

#n.y = a single number representing the sample size for y

You can also do ANOVA based on summary data using (ind.oneway.second) function in ‘rpsychi’