Simulating climate change in SLiM

[Billy Wallisch]

Hello All, 

This is my first time posting to SLiM discuss, I hope I am not misusing this group. 

I am attempting to simulate climate change in SLiM using three main phases, a burn in phase, a major disruption and a final response phase. 

Essentially, the burn in would create my base genetic diversity, phase two would be some major disruption by increasing temperature and then phase 3 temperature will be higher for all populations in the model in order to observe response to disruption. 

The main question I want to answer is: Did this major disruption compromise ability to adapt? 

I would love some help from this community if there is any literature you can steer me to in order to gain inspiration, or if you have any ideas for me as to how I could solve this issue. Maybe there is somewhere in the manual that you think could be most helpful. 

I am considering creating an optimal fitness level based on optimal climate patches, and then increasing these climate patches temperature by 1 - 1.5 degrees. 

I would love to discuss any ideas you all would have in simulating climate change in SLiM. 

Thank you all for the help, and my very best, 

Billy

[Peter Ralph]

Hello! 

 Here is a recent paper that does fairly sophisticated modeling of response to temperature that changes:

  https://www.biorxiv.org/content/10.64898/2026.02.26.708249v1.full

It is answering a different question, but gives you the tools, I think. Also note: https://www.journals.uchicago.edu/doi/full/10.1086/738891

In general, a suggestion is to pick a species - it is easier to choose parameter values (e.g., generation time; the mechanism for thermal (in)tolerance) if you have a particular species in mind.

Happy to hear what others have to say!

Peter

--
SLiM forward genetic simulation: http://messerlab.org/slim/
---
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/db1f918e-d655-43fe-a1ad-5e63c4bcee82n%40googlegroups.com.

[William Wallisch]

Hey Peter, 

Thank you for getting back to me so fast, and for providing great resources. I have really started to think more about this. 

In general I would like to keep my simulation abstract and stay away from choosing one species, but with your advice I will use an initial attempt to create a simple model by choosing one species. 

I am hoping to get to a point where I am using high-performance computing to simulate 1000s of species potentially, so I am attempting to make a simple, efficient, yet accurate model. 

I was thinking of mixing a few model types like a model with two mutation types, here is my basic outline: 

Burn in: ticks 0-10,000 (nonWF model) 

1. 4 populations with migration. 

2. Two mutation types: 

m1 being beneficial in high temperature and deleterious in low temperature , and m2 being beneficial in low temperature and deleterious in high temperature. 

3. A Static climate optimum of 0.0. 

Major Disruption: Tick 10,001

1. Optimum jumps to an extreme temperature. 

Possibly using a nonWF model I can use a selection curve to select which individuals survive after this shift in the optimum. 

Response 10,001 - 20,0000 ticks 

- Track extinction, population reduction, lag time, etc.

Output :
Pop size ~ time (generations)
Genetic diversity ~ time  (generations)

Again, I hope this is an okay use of this discussion forum, I am very new to SLiM and looking for some inspiration and guidance.

Thank you and my very best, 

Billy 

To view this discussion visit https://groups.google.com/d/msgid/slim-discuss/CAO0xWMC3v_bgpKACpLFGUi_CMv4AHqEhSiNjt%2B4jVE2rWGGOGA%40mail.gmail.com.

[Ben Haller]

Hi Billy!  It's fine to ask for general ideas on the list, but I'd suggest that if there are folks who want to have a discussion with you about this, they should probably contact you off-list.  If you get responses, you might organize a discussion off of this list among those folks, and then report back at the end with a new post, on this thread, summarizing your discussion and conclusions/decisions you reached.  This list goes to a couple hundred people, so it generates a lot of email for people who aren't interested in the thread; it's therefore best to have long discussions that are probably not of general interest elsewhere.  :->

As for your ideas, they seem fine, and honestly I'd just suggest that you jump into it and see what questions/difficulties you encounter.  If you haven't used SLiM enough to be sure where to start, then I'd suggest that you start by doing the SLiM Workshop, which can be done in person (but at present the three scheduled in-person workshops are all full and have gone to waitlist), or can be done on your own using the free workshop materials posted online.  That ought to get you pretty far towards being able to write this kind of model.  The SLiM manual is also a good resource, and has examples for things like spatial variation and temporal variation in selection, modeling polygenic/quantitative traits, etc.  There will be lots of decisions to think about along the way, but getting a good sense of the tools available in SLiM will help you to realize what the questions are and what the right answers might be.  :->  Happy modeling!

Cheers,
-B.

Benjamin C. Haller
Messer Lab
Cornell University

To view this discussion visit https://groups.google.com/d/msgid/slim-discuss/CAEcV7x_%3DXiRMVYFXT_fzOe8uXFiKFKo1WZ9x8pTgzy2M5fOP0g%40mail.gmail.com.