Protonate

I developed the automation script after getting one successful dyana->leap->sander run completed. It should have worked correctly, however the same error occurred. FATAL: Atom .R NCYS 1.A HS 14 does not have a type. I have now discovered that I had correctly modified the nomenclature for the cysteine residue's proton on the sulfur atom. I had in fact used HSG as required. That solution was masked because protonate also changes nomenclature and over wrote my fix. Now I have added a sed line:

sed -e 's/ HG /CYS/HSG CYS/' < leap_temp.apdb > leap_temp_pro.pdb

after protonate has updated the file and before leap processes the data

Does anyone know how to make protonate produce nomenclature for leap in the first place. Or at least have tleap recognize the original HG nomenclature?

My concern is that even though my sed line patches the problem, it could cause problems in a future protein that has a CYS at some other location other than the n-terminus. I also know that a serine also labels a proton HG so global substitution will not work.

Using AMBER for Molecular Dynamics

I have a peptide tail that has a CYS residue as the n-terminous and when I run tleap:

tleap -s -f leaprc.ff99.tf using what I thought was the correct pdb syntax file

...

ATOM      1  N   CYS     1       1.325   0.000   0.000

ATOM      2  H1  CYS     1       0.858  -0.893   0.114

ATOM      3  H2  CYS     1       1.960   0.150   0.776

ATOM      4  H3  CYS     1       0.635   0.742  -0.020

ATOM      5  CA  CYS     1       2.073   0.000  -1.245

ATOM      6  HA  CYS     1       1.393  -0.209  -2.045

ATOM      7  CB  CYS     1       2.727   1.357  -1.513

ATOM      8 3HB  CYS     1       3.751   1.327  -1.194

ATOM      9 2HB  CYS     1       2.204   2.118  -0.965

ATOM     10  SG  CYS     1       2.660   1.743  -3.300

ATOM     11  HG  CYS     1       3.231   2.924  -3.520

ATOM     12  C   CYS     1       3.105  -1.128  -1.186

ATOM     13  O   CYS     1       3.825  -1.264  -0.198

ATOM     14  N   ARG     2       3.143  -1.909  -2.256

ATOM     15  H   ARG     2       2.515  -1.728  -3.026

.... etc

which has already been pre-processed using protonate I get the following error message.

Created a new atom named: HS within residue: .R

FATAL: Atom .R.A does not have a type

I have check the PDB file and it seems to have the correct number of hydrogens so why is it trying to add another? I have also look through the leap libraries and noticed a number of difference nomenclatures for the hydrogen on the sulfur (HS, HGS, HG) and have tried those permutations in the PDB file without success.

Well, I was about to send off my message to the amber@scripps.edu mirror site for help in the AMBER group and I finally discovered my error. The line in the library was

"HSG" "HS" 0 -1 0.0

Indicating that the correct nomenclature was HSG! Even though the standard PDB format would be HG and the AMBER type is HS. Anyway, I am glad to get over this small error. I was able to run a set of molecular dynamics calculations to test out sander (the main program in the AMBER suite) and a solution was computed. My next task is to streamline the whole process and run a set of 50 to 100 structures to ensure that they converge together.

# posted by xcrabturtle : 7:31 PM

Using AMBER

Well I have gotten AMBER running on one of our machines. The next key task is to get it running on a cluster of SGI's using a batch submit script. The version I got running is on an Onyx and takes a fair amount of time to run. We have got to get the performance better.

#Processors | Time
------------------------------
1 | 11 minutes / run
2 | 7.5 minutes / run
4 | 6 minutes / run
------------------------------

And these were for relatively small structures.