Hi all,
I was running the integration test of the enzyme design (main/tests/integration/tests/enzdes). According to the fifth constraint block in the cstfile at position Trp100 residues WFY with atom type aroC are allowed. However within 150 output structures I never observe a mutation away from the native Trp. In contrast, residue Ser98 is variable in the results (only has backbone constraints). Similarly, I never observe any mutation away from a native catalytic sidechain in my own project.
Is this expected behavior? I am presuming that due to the prior optimization of the catalytic interactions, the mutation of a catalytic residue is highly unfavored and therefore no mutation of these is ever observed but I'm not sure about this.
Do you know if it is by principle possible that a catalytic residue (sidechain involved in catalysis) is mutated by enzyme_design, when listing different amino acids in the cst file as possible residues?
Thanks!
I was running the integration test of the enzyme design (main/tests/integration/tests/enzdes). According to the fifth constraint block in the cstfile at position Trp100 residues WFY with atom type aroC are allowed. However within 150 output structures I never observe a mutation away from the native Trp. In contrast, residue Ser98 is variable in the results (only has backbone constraints). Similarly, I never observe any mutation away from a native catalytic sidechain in my own project.
Is this expected behavior? I am presuming that due to the prior optimization of the catalytic interactions, the mutation of a catalytic residue is highly unfavored and therefore no mutation of these is ever observed but I'm not sure about this.
Do you know if it is by principle possible that a catalytic residue (sidechain involved in catalysis) is mutated by enzyme_design, when listing different amino acids in the cst file as possible residues?
Thanks!
Category:
Post Situation:
The "design" using the enzdes pipeline is performed using the matcher app. The matcher app determines which of the allowed amino acids could satisfy the constraints and outputs a PDB with that information in REMARK lines. The sequence of the catalytic residues is specified in the REMARK lines at the top of the PDB, and it is fixed by default, I think. To accomplish screening of multiple catalytic residue identities, you would need to use several different inputs where the REMARK lines differ. (The 3-letter code for the coordinates of those residues may also need to match.)
I've been using the enzdes-style constraints more for the way they describe the system of interactions than in an actual catalytic residue redesign case, but this is my understanding of the way it works.
Hope that helps.
It depends a bit on the process you use to do the mutations, but generally with the Enzdes protocol the identities of the catalytic residues are fixed.
As mentioned, the Enzdes process typically functions in two stages - the matching stage to place the catalytic residues on the backbone, and the "design" phase to redesign the context. The catalytic residues are sampled during the matching phase, and then for the design phase the matched residues are used, and the constraint file is used solely for the geometry specification, and not for any identity sampling.
If you want to sample the catalytic residue identity, you would want to do this during the matching stage, such that the residue identities going into the design phase are what you need. (Note that, in general, a backbone which can support appropriate catalytic geometry for one AA identity (e.g. Glu) can't necessary support a different identity (e.g. Asp). That's why the residue identity is determined at the match phase, and a backbone with geometry which can support multiple residue identities will be output with multiple match results, one for each residue identity combination.)
Thanks a lot to both of you for your helpful explanations. Really clarified things for me.