feat: dead end / orphan metabolites and blocked reactions

[sulheim]

Description of the issue:

Based on the Memote report we find that the model has 645 universally blocked reactions, 137 orphan metabolites (only consumed, not produced) and 219 dead-end metabolites (only produced, not consumed).

A question is if we should keep, remove or fix these blocked reactions and the orphan / dead-end metabolites?

To do:

• Agree on a solution for handling these reactions / metabolites
• perform necessary curation (remove / fix)

I hereby confirm that:

• This problem persists in the master branch of the repository
• A similar issue does not already exist
• I've considered asking this first in the Gitter chat room

[edkerk]

I assume these are all gene associated reactions? Probably best not to remove them, as there is evidence that these enzymes are present and if anything future curation might connect them. Non-gene associated reactions in that list don't seem to have a function and are not included because of sequence proof, so can be removed from the model?

Are there any reactions that can be argued to be outside the scope of a metabolic model? For instance protein modifications is sometimes included but can be argued not to be part of metabolism.

Not sure how to efficiently fix the remaining reactions though? Could there be an (semi-)automated way, or would this require checking each blocked reaction/metabolite one-by-one?

[sulheim]

562 of the reactions have genes associated. Seems like all the reactions are within the scope of the model. To me it seems like quite a few of them are in the fatty acid / lipid biosynthesis and another big group is BioCyC-originated reactions from Sco4 not well connected to the model.

There are a few of the reactions without any gene annotation that can be removed, but we should keep the fatty-acid elongation reactions even if they don't have an associated gene.
Some of the reactions from BioCyc seems a bit far out, but I guess we can keep them

Actually, I think we should put som effort into unblocking some of the reactions.

gene_reactions_blocked.txt