Remove duplicates - identical except that one is unidirectional, the other reversible

[johan-gson]

Description of the issue:

The following reactions is of the type
remove: a[x] + b[x] => c[x]
keep: a[x] + b[x] <=> c[x]

It is difficult to say which is right, but at least the 'remove' reactions are currently meaningless and only makes the model larger. They also create loops, which is sometimes annoying, for example in tINIT.

Otherwise identical, although sometimes reversed perhaps.
The reactions (46 in total):

To remove:  	to keep:
{'MAR08971'}    {'MAR04306'}
{'MAR00121'}    {'MAR00607'}
{'MAR04911'}    {'MAR04411'}
{'MAR06324'}    {'MAR00079'}
{'MAR00101'}    {'MAR02199'}
{'MAR00445'}    {'MAR02339'}
{'MAR07441'}    {'MAR01728'}
{'MAR07742'}    {'MAR01827'}
{'MAR04455'}    {'MAR11237'}
{'MAR04463'}    {'MAR11251'}
{'MAR04478'}    {'MAR11252'}
{'MAR04502'}    {'MAR11253'}
{'MAR04517'}    {'MAR11254'}
{'MAR04576'}    {'MAR11247'}
{'MAR04609'}    {'MAR11249'}
{'MAR04622'}    {'MAR11248'}
{'MAR04626'}    {'MAR11255'}
{'MAR04653'}    {'MAR11256'}
{'MAR04659'}    {'MAR11257'}
{'MAR04669'}    {'MAR11258'}
{'MAR04713'}    {'MAR11260'}
{'MAR04724'}    {'MAR11261'}
{'MAR04747'}    {'MAR11262'}
{'MAR04751'}    {'MAR11263'}
{'MAR04753'}    {'MAR11264'}
{'MAR04761'}    {'MAR11265'}
{'MAR04794'}    {'MAR11238'}
{'MAR04795'}    {'MAR11266'}
{'MAR04798'}    {'MAR11267'}
{'MAR04800'}    {'MAR11239'}
{'MAR04801'}    {'MAR11268'}
{'MAR06494'}    {'MAR11240'}
{'MAR06497'}    {'MAR11241'}
{'MAR06504'}    {'MAR11242'}
{'MAR06541'}    {'MAR11243'}
{'MAR06582'}    {'MAR11244'}
{'MAR06590'}    {'MAR11246'}
{'MAR06594'}    {'MAR11271'}
{'MAR06706'}    {'MAR11245'}
{'MAR06853'}    {'MAR11280'}
{'MAR06856'}    {'MAR11278'}
{'MAR06857'}    {'MAR11279'}
{'MAR06858'}    {'MAR11259'}
{'MAR07191'}    {'MAR09057'}

Expected feature/value/output:

Current feature/value/output:

Reproducing these results:

I hereby confirm that I have:

• Tested my code on my own computer for running the model
• Done this analysis in the main branch of the repository
• Checked that a similar issue does not exist already

[johan-gson]

And, all pairs (remove, keep) have identical GPRs

[JonathanRob]

Again, nice find, and it makes things simpler since the GPRs are the same.

As for which of the reactions to keep from each pair, I wonder if we should keep the older (lower numbered) reaction and make it reversible, or just keep the reversible reaction regardless of it's relative "age"?

[mihai-sysbio]

Hmm it might be easier to keep the reversible reaction, so that deprecating the reactions implies less work in terms of carrying over reaction annotation.

[JonathanRob]

Resolved in #403 and #405.