extract 10001+ pdbs

[Anonymous]

Awesome, back-to-back posts

While I’m having trouble getting 3.1 working, I’m using my old 2.3.0 version. I ran a calculation with 20000 structures and I’m having trouble extracting them. I get a ton of parse errors in the output. I notice in the silent output file that in the SCORE line, I’ll have pdb codes such as S_13824_8376, whereas at the end of the other lines it’ll simply be S_13824_837. Is this “fixed” in 3.1? Or should I find a work-around (which I’ll do in the meantime anyways)?

As a side note … what ARE the last 4 digits of each pose, anyways?

[Anonymous]

I know little about silent files in 3.x and nothing about silent files in 2.x.

The last 4 digits are just the number of the trajectory. If you do 5000 trajectories, you’ll get my_result_0001, my_result_0002, …, my_result_4999, my_result_5000.

I believe in ++ it was hardcoded to 4 digits.

I know for large parts of 3.x it is NOT hardcoded to 4 digits. It is a minimum of 4 digits, but it resizes itself larger if you use nstruct larger (nstruct 20000, for example. I did 2000000 once, it was a dumb idea, since it never got close to that number before the job ran out of cpu time.)

[Author]

Posts