How to model coculture using multi-species function in COBRA?

[Yuhan]

Hi, I'm currently investigating the interaction between the microorganisms during wine production and wanting to find out their common contribution to wine aroma. I implemented FBA to my four models (Saccharomyces cerevisiae, Komagataella phaffii, Oenococcus oeni, Leuconostoc mesenteroides) separately and got the flux distribution data. What I want to do now is to find out how to put the species together and simulate with FBA. So is there a function in cobrapy allowing me to do that? Thanks.

[Greg Medlock]

Hi Yuhan,

Definitely take a look at Christian Diener's cobrapy-based micom package for your study: https://micom-dev.github.io/

But pay special attention to how assumptions you make about the behavior of your community are reflected in the simulations; Christian has a great section on the logic they use for gut microbiota in particular: https://micom-dev.github.io/micom/logic.html#Steady-states-in-the-community

Ultimately, you are going to be able to perform more realistic simulations if you have experimental data, particularly 1) relative abundance for each species, 2) approximate growth rates for the entire community (but preferably the individual members), and 3) any available community or species-specific metabolite uptake or secretion rates. The further away you can get from relying on an objective function that makes an assumption about community-level behavior, the better!

Greg

On Wed, Jun 24, 2020 at 2:25 PM Yuhan <winter...@gmail.com> wrote:

Hi, I'm currently investigating the interaction between the microorganisms during wine production and wanting to find out their common contribution to wine aroma. I implemented FBA to my four models (Saccharomyces cerevisiae, Komagataella phaffii, Oenococcus oeni, Leuconostoc mesenteroides) separately and got the flux distribution data. What I want to do now is to find out how to put the species together and simulate with FBA. So is there a function in cobrapy allowing me to do that? Thanks.

--
You received this message because you are subscribed to the Google Groups "cobra pie" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cobra-pie+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/cobra-pie/ca954de0-c113-4eff-a734-77ad2c26fc89o%40googlegroups.com.

[Yuhan]

That's brilliant! Thanks so much, Greg. It helps me a lot :)

在 2020年6月25日星期四 UTC+8下午11:36:27，Greg Medlock写道：

To unsubscribe from this group and stop receiving emails from it, send an email to cobr...@googlegroups.com.

[Yuhan]

One thing forget to mention, I found some trouble for me to open the first link and in case anyone else has the same problem, I just got another one: https://micom-dev.github.io/micom/ 

This works for me well so I thought it might be helpful.

在 2020年6月26日星期五 UTC+8下午10:36:21，Yuhan写道：

[SANKAR MAHESH R]

Hi Yuhan. Were you able to model using MICOM? I'm also interested in a case similar to yours, but I'm stuck with implementing the exchanges.

Particularly,

I'm interested in building an artificial community,

where a metabolite produced and secreted by organism 1 is then consumed by organism 2 and processed downstream.

To model this, I set an exchange reaction ('EX_') for model 2, but this way it assumes that the model 2's medium contains this metabolite.

I want this metabolite to be conditionally available only in the presence of model 1.

So do you think I should make changes to the media during community simulations or what else should I do to conditionally enforce this?