On 2/14/12 12:51 PM, Johs, Alexander wrote: > I’m trying to model a protein with a cyanocobalamin cofactor. I have > included CHARMM topology and parameter files using the corresponding > .append commands, created a custom residue for the cyanocobalamin in > restyp.lib and added atomic radii. Looks like there is a limit on the > number of atoms per residue… How can I increase this limit?
Sorry, you can't - the limit of 100 atoms per residue is hard-coded into Modeller. (Even if you could, there are other limits that you'd run into, such as a maximum number of bonds per residue.) If your cofactor is present in the template, then you should be able to use a simple block residue ('.' one letter code in your alignment) to copy it rigidly from template to target instead - that shouldn't care how many atoms there are. Or you could strip out the hydrogens - that should get your atom count under 100. Or you could split the cofactor into several residues, each with less than 100 atoms, and add suitable inter-residue bonds (ugly, but it should work just fine).
I can look into removing this limit in future versions of Modeller. Can you send me your topology file so I can make sure it works?
Ben Webb, Modeller Caretaker