CHARMM Development Project Forums User Discussion & Questions Energy terms, Constraints, Restraints, and Solvation A problem induced by SHAK ANGH

Dear All CHARMMer,

I want to use shake to constrain my protein structure. So I have two questions.

1) During MD calculation, I would like to fix the shape of helix and sheet to save time, but I hope the distance between different helix and sheet would change in MD calculation. I don't know if it is appropriate to use Shak Angh. Accually after I looked for the mannual of CHARMM, I choose this. Could anybody give me any suggestion about that?

2) I have started to use shak, but the calculation stopped abnormally, the output is as following:

CHARMM> shak angh sele atom ctf_ 2 * end
SELRPN> 20 atoms have been selected out of 1004
SHKCOM> SHAKE parameters: TOL = 0.1000D-09 MXITer = 500
SHKSET: * WARNING * Atom 18 is involved in:
4 bond end, 3 angle mid, and 3 angle end constraints.
The number of degrees of freedom will probably be wrong.

SHKSET: * WARNING * Atom 25 is involved in:
3 bond end, 0 angle mid, and 4 angle end constraints.
The number of degrees of freedom will probably be wrong.


39 constraints will held by SHAKE.

***** ERROR IN SHAKEA ***** COORDINATE RESETTING WAS NOT ACCOMPLISHED IN 500 ITERATIONS

*** LEVEL -2 WARNING *** BOMLEV IS -5
SHAKEA: Using routine SHAKEA2 for shake calculation.
contraints: 39
EBONDC: Using routine EBONDFS for energy calculation.
EANGLC: Using routine EANGLFS for energy calculation.
EBONDC: Using routine EBONDFS for energy calculation.
EPHIC: Using routine EPHIFS for energy calculation.
EIMPHIC: Using routine EIPHIFS for energy calculation.
ENBOND: Using routine ENBFS8 for energy calculation.
***** ERROR IN SHAKEF ***** FORCE RESETTING WAS NOT ACCOMPLISHED IN 500 ITERATIONS

*** LEVEL -2 WARNING *** BOMLEV IS -5
***** ERROR IN SHAKEA ***** COORDINATE RESETTING WAS NOT ACCOMPLISHED IN 500 ITERATIONS

*** LEVEL -2 WARNING *** BOMLEV IS -5
SHAKEA: Using routine SHAKEA2 for shake calculation.
contraints: 39
EBONDC: Using routine EBONDFS for energy calculation.
EANGLC: Using routine EANGLFS for energy calculation.
EBONDC: Using routine EBONDFS for energy calculation.
EPHIC: Using routine EPHIFS for energy calculation.
EIMPHIC: Using routine EIPHIFS for energy calculation.
ENBOND: Using routine ENBFS8 for energy calculation.
***** ERROR IN SHAKEF ***** FORCE RESETTING WAS NOT ACCOMPLISHED IN 500 ITERATIONS

What is wrong with calculation?

I am looking forward to your answer.

Thank you very much

Holly

You cannot keep helix/sheet shapes with shake angh; shake with angles is not widely used - typically SHAKE BONH is used to constrain all X-H bonds. I do not know of any (working) way of fixing shapes of portions of your structure, yet allow them to move relative one another, to save time.

Thank you very much for your reply. To my understanding if I can fix angle of some atoms, the distance of these atoms and relative position should not change. Doesn't it mean the shape, which consisting of these atoms, wouldn't change?

And could you please tell me why the error happened in my calculation?

Again thanks a lot for your answer.

Best regards

Holly

Fixing bond angles does not hinder rotations around bonds.

I really appreciate for your reply. I see what you mean. Maybe I should just fix the bond length to save time. Is that right?

Best

Holly