I am trying to create a protein network which is able to reorganize itself upon environment/parameter changes.

I have several components that can interact with spatially separated domains but I found the Crosslink Restraint not suitable to define this.

Whenever a domain has multiple crosslinks given to it, in the minimal output the domain is always equally far from its interaction sites.

Is it possible to have all the possible interactions as input, but the domain chooses only one of them (the optimal one depending on the spatial configuration)?

Maybe a logarithmic approach for converging to one of them or using a different restraint.

(I am using the BuildModel macro.)

Thanks in advance,

Komjati Kornel