superset-subset solution

[VasilisFlouris]

Dear Straindesign Team,

While using sd.any, I have observed that in some cases the solutions found appear to have a superset–subset relationship involving one or more reactions. To my understanding, although sd.any does not guarantee minimality of overall cardinality, the individual solutions should still be minimal with respect to their reaction sets.

Is this behavior expected, or could there be an issue with how minimality is being assessed?

Thank you for your time,

Vasilis

[VonAlphaBisZulu]

Hi Vasilis, I updated the main branch, and some of the processing for found solutions has changed. Try installing StrainDesign from main to see if the problem persists,

Thanks for reporting the issue,
Phil

[VasilisFlouris]

Thanks for the update,

I'll try with the new branch and hopefully the problem is solved ! I'll update this issue once I have an answer,

Vasilis

[VasilisFlouris]

Hello again !

after extended simulations with the newer version the problem has unfortunately not disappeared. In some cases the super-sets can be up to 50% of the total computed strains.

Vasilis

[VonAlphaBisZulu]

Would you mind sharing the code (in private, if you prefer) that produces this error. The setups that I use in my continuous integration tests and benchmarks don't produce that problem and I'd need to reproduce your problem in order to fix it.

Best,
Phil