fix-gene: remove non-transporter genes - updated

[haowang-bioinfo]

Main improvements in this PR:

• remove 20 non-transporter genes from Transporter reactions, as proposed in remove non-transporter genes from transport reactions #444

I hereby confirm that I have:

• Tested my code on my own computer for running the model
• Selected develop as a target branch

[mihai-sysbio]

See my comment in #444 - I'm worried that we might be moving too fast here.

[johan-gson]

It was difficult to see the exact changes from last time, but I think you thought it through - I will not go over everything again. Good job!

[haowang-bioinfo]

as suggested in #426, simulation checks with models before and after these deletions were conducted and showed exactly the same growth rates.

[JonathanRob]

simulation checks with models before and after these deletions were conducted and showed exactly the same growth rates

Why would removing genes change simulation results? Unless you're using an enzyme-constrained model, I don't quite understand how it would influence flux at all (assuming by "simulation" you mean FBA).

[haowang-bioinfo]

Why would removing genes change simulation results? Unless you're using an enzyme-constrained model, I don't quite understand how it would influence flux at all (assuming by "simulation" you mean FBA).

@JonathanRob yes, now it's just FBA that gave the same results since stoichiometric was not changed. As just commented in #426, a systematic checking code would be needed as reaction addition/removal may happen in the future.

[feiranl]

Gene essentiality analysis should performed in terms of removing genes? The check function is coming.

[haowang-bioinfo]

Gene essentiality analysis should performed in terms of removing genes?

agree

[johan-gson]

If you want I can do the gene essentiality part - I recently did this for the single-cell modeling paper, so I know how to do it and have the code on the cluster etc. set up for this. I can do it in 1-2 weeks (or later), I don't know the status of the model, when it will be ready.

[johan-gson]

But don't expect any large changes to the prediction power :)