Knockout analysis without specific medium?

[PathogeNish]

Hi All,

I have just started using MICOM's knockout analysis to study the microbial interaction networks in 16S stool samples. By,

• Using the standard build function to create a MICOM community
• Using the com.knockout_taxa to perform the knockout study

It just occurred to me that nowhere in this have I yet specified the growth medium. Unlike while using the grow command to analyse metabolic fluxes where you have to add a set of input fluxes.

My questions are:

1. Doesn't the knockout analysis need a medium to perform the study (the com.medium variable looks like it contains the maximum possible flux of every metabolite involved in any reaction for a microbe in the community)?
2. If I force a medium onto the community by changing the com.medium variable to the western diet fluxes, and then run the knockout analysis, does this give me more "medium relevant" results?

[cdiener]

Hi, sorry for overlooking this. If you don't specify a medium MICOM will use whatever is predefined in the models and that tends to correspond to the richest possible medium (all metabolites with a known import reaction are available).

I agree that it is probably more relevant to run the knockout analysis with a realistic medium. What you propose in step 2 is exactly how you do it. Just be aware that com.medium expects a pandas.Series or dictionary and not a DataFrame like grow. So if your medium is a DataFrame you probably want to do something along the lines of com.medium = medium.flux.