Simulation drift from starting population

[Giacomo M]

Hello everyone,

I’m just getting started with population simulations, and after looking through different tools, SLiM 5 seems like it could be a very good fit for what I’d like to do. Before I get too deep into building a model, I wanted to ask for some guidance and to make sure SLiM is actually the right choice for the kind of scenario I’m interested in.

The basic idea is to simulate neutral evolution in a closed population, starting from the genetic diversity of a real dataset. I’m working with two empirical populations: population 1 and population 2. From the genetic patterns I see, my current hypothesis is that population 2 might simply reflect a strong founder event followed by drift acting on population 1. To explore this, I’d like to take the allele frequencies or overall diversity of population 1, generate a small set of founders from those frequencies, and then simulate a short demographic history that could plausibly lead to something like population 2. In practical terms, I’m thinking of starting with around twenty founders and letting the population grow quickly to about five hundred individuals over roughly five generations. I would then sample individuals at each generation and compare them with the real data using PCA.

For now I’m aiming for a relatively simple, neutral model so I can see whether this basic explanation is even plausible. Eventually, since I’m not working on humans, I’d like to incorporate a more realistic life history, including aspects of the mating system or birth–death processes, but that’s secondary at this stage. One complication is that I only have unphased genotype data. I can work with a VCF and use the nucleotide-based model if necessary, but I don’t have haplotypes to start from, so I’m especially interested in good practices for initializing SLiM with empirical allele frequencies or unphased genotypes.

My question is simply whether SLiM can handle this type of setup, and if so, what would be the most sensible way to structure a model like this. Any advice or pointers would be very welcome.

Thanks very much for your time and for letting me join the discussion group!

[Peter Ralph]

Hi, Giacomo! The short answer is that yes, SLiM can totally do this. Looking at the table of contents for the manual, it sounds like good places to start might be  section 19.12 "Reading VCF files to create nucleotide-based SNPs", and the non-Wright-Fisher section. This doesn't directly answer your "good practices for initializing from frequencies" question; I think what most people would do there would be to (a) compute allele frequencies, and (b) go through and randomly assign alleles with those probabilities at each variable site, for each starting individual.  However, I'd be tempted to do some kind of computational phasing, not worry too much about how correct it is, and then initialize from the VCF; that'll get at least the most obvious haplotype structure.

Happy slimulating!

Peter 

---

From: slim-d...@googlegroups.com <slim-d...@googlegroups.com> on behalf of Giacomo M <ramarrom...@gmail.com>
Sent: Friday, November 28, 2025 7:05 AM
To: slim-discuss <slim-d...@googlegroups.com>
Subject: Simulation drift from starting population

 

Hello everyone, I’m just getting started with population simulations, and after looking through different tools, SLiM 5 seems like it could be a very good fit for what I’d like to do. Before I get too deep into building a model, I wanted to

ZjQcmQRYFpfptBannerStart

You haven't previously corresponded with this sender.

Use caution with links and attachments. Learn more about this email warning tag.

Report Suspicious

 

ZjQcmQRYFpfptBannerEnd

--
SLiM forward genetic simulation: http://messerlab.org/slim/
Before posting, please read http://benhaller.com/slim/bugs_and_questions.html
---
You received this message because you are subscribed to the Google Groups "slim-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to slim-discuss...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/slim-discuss/b306e889-69d5-403f-b922-d32c4e4e40dcn%40googlegroups.com.