Fredrik wrote:
If I would like to do it the hard way, can I do it like below? I.e. will the modelling be correct? (I've done it and it seems to modell correct.)
Everything looks OK. However, since the K+ ion is just a point, this won't give you much beyond treating it as a BLK residue (I think the only difference you'll see is that it'll have a defined VDW radius this way, assuming you also alter radii.lib - Modeller doesn't do electrostatics by default, so the charge is ignored).
in the par.lib, par_all22_prot_b5.inp,
Modeller doesn't use par_all22_prot_b5.inp - that's the original CHARMM file from which par.lib is derived - so there's no point in editing that.
in the top_ca.lib in the top.lib, top-charmmH.lib in the top_allh.lib, top_allh-charmmH.lib, top_polh.lib, top_mnss.lib, top_mnch.lib, top_mcwo.lib, top_mccb.lib, top_heav.lib, top_cacb.lib, top_ca3h.lib, in the top_all22_prot_b5.inp (Guess this would be unnessesary?!)
You probably only need to change top_heav.lib; the other top_*.lib files are only used if you use non-default topologies. There is no need to change the *-charmmH.lib files, as they're not used by Modeller 7 and above (they are the old versions of top.lib, from Modeller 6). top_all22_prot_b5.inp is the original CHARMM file again, so as you suggested, changing it is redundant.
Would it be correct to change to: ------------------------------------------------------- 104 | POT | k | POT | potassium ion ------------------------------------------------------- in the restyp.lib, restyp_accelrys.lib?
Looks OK, although you don't need to change restyp_accelrys.lib - that's only used in the commercial version of Modeller.
Ben Webb, Modeller Caretaker
--
http://www.salilab.org/modeller/
Modeller mail list: http://salilab.org/mailman/listinfo/modeller_usage