Hi David,

It's really a balance of sample size to sequence coverage. High-throughput sequencers have a fixed number of reads they generate. So you can do some math to estimate how many reads you expect per individual depending on genome size and sequencing instrument (see Puckett 2016 Cons Gen and Andrews et al. 2016 Nature Reviews for some examples). You don't want a large sample size but then poor coverage that limits your ability to call SNPs.

It also depends on if you have an available reference genome or plan on de novo assembly of contigs. If going de novo, you need higher coverage. We will often select a few individuals, sequence them at a high depth using RADseq, assemble reference contigs, and then go back and re-sequence more individuals at a lower depth.

I am use to population genetics: you can get away with larger sample sizes and lower coverage using probabilistic genotyping approaches like ANGSD. For phylogenetics you may want higher coverage to have confidence in your haplotypes (but see Saglam et al. 2016 Molecular Ecology).

Personally, I have a hard time with studies using RADseq that have 1-2 individuals per population. I think you can get away with a minimum of ten. There are a few papers out there that talk about appropriate sample size to generate pop gen stats.

Hope this helps.

Dear David

The sample size for any study depends on the standard deviation of the variable ( from previous studies ) and the margin of error you decided. The formula:

n= ( Z^2 * S^2) / E^2

where : Z  ( 1.96 for 0.05 and 2.58 for 0.01 )

S = standard deviation from previous studies or pilot  study

E = significant level 

Also, you can use software to calculate sample size ( SPSS , Minitab , G*power )

My advice to use G*Power .

G*Power software is effective tool to calculate sample size for many ranges of experiments. Also, you can determine effect size and power of the test, G*Power is free to download and easy to use after reading the manual, the download link:

http://download.cnet.com/G-Power/3000-2054_4-10647044.html

The link to download G*Power manual:

http://www.gpower.hhu.de/fileadmin/redaktion/Fakultaeten/Mathematisch-Naturwissenschaftliche_Fakultaet/Psychologie/AAP/gpower/GPowerManual.pdf

I hope that be useful for you.

Good Luck

Hi David,

It's really a balance of sample size to sequence coverage. High-throughput sequencers have a fixed number of reads they generate. So you can do some math to estimate how many reads you expect per individual depending on genome size and sequencing instrument (see Puckett 2016 Cons Gen and Andrews et al. 2016 Nature Reviews for some examples). You don't want a large sample size but then poor coverage that limits your ability to call SNPs.

It also depends on if you have an available reference genome or plan on de novo assembly of contigs. If going de novo, you need higher coverage. We will often select a few individuals, sequence them at a high depth using RADseq, assemble reference contigs, and then go back and re-sequence more individuals at a lower depth.

I am use to population genetics: you can get away with larger sample sizes and lower coverage using probabilistic genotyping approaches like ANGSD. For phylogenetics you may want higher coverage to have confidence in your haplotypes (but see Saglam et al. 2016 Molecular Ecology).

Personally, I have a hard time with studies using RADseq that have 1-2 individuals per population. I think you can get away with a minimum of ten. There are a few papers out there that talk about appropriate sample size to generate pop gen stats.

Hope this helps.