Fatty acid synthase complex modelling

[PCarme]

In the fatty acid biosynthesis model that is currently on PomBase, I have modelled the fatty acid synthase complex as a protein complex, using the composite MF term "fatty acid synthase activity", making that part of the pathway pretty simple:

However, the full process of palmitate synthesis by the FAS is very well described, with the roles of every subunit well understood, so under the new GO-CAM guidelines, I guess we should model each activity separately.
The problem is, "well understood" isn't always the same as simple, and the FAS is a pretty complicated molecular machine: it is composed of 6 units of Fas1 and 6 units of Fas2 ; both proteins have several molecular functions that coordinate to first assemble a malonyl-coA and an acetyl-coA into an 4 carbon fatty acyl onto the ACP domain of Fas2, and then progressively elongates it by adding more malonyl-coA until reaching a 16 carbons chain.

For now, I have only modelled the first elongation cycle and it is already looking kind of mad (see the middle branch with all the fas1 and fas 2 here: http://noctua.geneontology.org/workbench/noctua-visual-pathway-editor/?model_id=gomodel%3A678073a900002931.

Should I continue, or should we try to convince GO to make this one of the "molecular machine" exceptions to the rule?

[PCarme]

Although, in all honesty, the rest of palmitate biosynthesis will be quite simpler from where the model is now: each subsequent elongation round is just adding 2 carbons from malonyl-coA onto the existing fatty acyl-ACP using Fas2's '3-oxoacyl-[acyl-carrier-protein] synthase activity' and the 3 following linear reactions, until it reaches 16 carbons.

So, it won't be too difficult to model or too complicated in terms of connections. but it would make a very long and repetitive model...

[ValWood]

I'm not sure.
Isn't this 2 different issue.
1 should we model the individual parts of the complex (I think yes here)
2. Should we model every addition cycle
It will look a little crazy but maybe we should do it.
can discuss tomorrow.
v

[PCarme]

1 should we model the individual parts of the complex (I think yes here)

At least we could, since we have all the molecular details to model it. But when researchers think about the fatty acid synthase complex function, do they think of every separate step, or do they think of the start and end products of the process? Is this level of detail useful in the model?
It might be worth having it modelled as a complex in the fatty acid biosynthesis pathway, and as distinct stepwise activities in a separate "palmitate biosynthesis" pathway, but we might not be able to join the 2 models...

2. Should we model every addition cycle

If the point of GO-CAM is to accurately model detailed biological processes, then I think it makes sense.
We could model it as an elongation cycle, with general compounds terms as the outputs of each step like it is usually represented in schematics:

But I think that would make the chain of causality less clear, which kind of defeats the point of GO-CAM.

[ValWood]

Thoughts how best to model fatty acid biosynthesis?
@vanaukenk @pgaudet

[ValWood]

I think I now agree with @PCarme suggestion to model the FAS as a single activity, but model the elongatation cycle.