You are here

how do I repeat relax runs and get scoring functions at every state

4 posts / 0 new
Last post
how do I repeat relax runs and get scoring functions at every state
#1

I have two questions:
1) To compare two different minimization algorithms, I run "-relax -cenrlx -ex1 -ex2", which, I believe, contains randomness in perturbation. How do I make each relax run the same?

2) How do I get the output of scoring functions of the chain after every move, especially during minimization?

Thank you very much.
Temsiri

Mon, 2008-11-03 19:42
Temsiri

*.sc file that I got only gives final scores. How do I get scores after every minimization step?

Thanks,
Temsiri

> I have two questions:
> 1) To compare two different minimization algorithms, I run "-relax -cenrlx -ex1 -ex2", which, I believe, contains randomness in perturbation. How do I make each relax run the same?

You can add the flag -constant_seed

> 2) How do I get the output of scoring functions of the chain after every move, especially during minimization?

the *.sc file should record the scores

>
> Thank you very much.
> Temsiri

Thu, 2008-11-06 19:04
yiliu

1)Would -constant_seed work with multiple decoys?
My command line is
../../rosetta++/rosetta.gcc aa 1d3z _ -nstruct 1 –relax -cenrlx -ex1 -ex2
-l

2)* .sc file that I got only gives final scores. How do I get scores after every minimization step?

Thanks, Temsiri

Fri, 2008-11-07 06:50
Temsiri

> 1) To compare two different minimization algorithms, I run "-relax -cenrlx -ex1 -ex2", which, I believe, contains randomness in perturbation. How do I make each relax run the same?

BTW, the ex flags apply only to full atom, not to centroid modes, so I don't think they are doing anything for you in your command line. (They refer to chi 1 and chi 2 angles of side chains).

Wed, 2008-12-10 12:30
vanita