R: sim1000G

Hi Berina,

I'm not an expert on genetics.
I haven't looked at the package.
And I've only glanced at your question.
So, this is probably not the best response.

But as no one else has responded, here's some comments:

(1)

Have you checked if there's a function in the package to do what you want?
The remainder of these questions assume that you have, and the answer is no.

(2)

I'm having some difficulty following the sample size.
The initial code sets an explicit value to 3000.
But if I'm following it correctly, the object that's returned has 2000
rows, containing two 1000 row groups.
But then your question implies you want 48?

Given that the sample already contains two groups, are they relevant
to the sample you're trying to produce?
And are you wanting to take a small sample of 48 from a larger sample
of 2000, or something else?

(3)

Are the observations (not sure if that's the correct term here) in
each 1000 row group, statistically independent?
If they are, then taking a smaller sample should be relatively simple.
If they're not, then this is a much more complex question, that's
probably off-topic.

(4)

What exactly is in the data?
i.e. Could you call the str() or head() functions on the data, and
show us the results?

(5)

Is there a boolean-style variable in the data, indicating whether each
row is casual or non-casual?

B.

On Fri, Oct 30, 2020 at 10:37 PM Berina Zametica UNI