CHARMM Development Project
Posted By: shane yue MAXTimesteps in CORREL - 07/07/14 02:10 PM
Hi CHARMM users!

Hope you have enjoyed your holiday! I have a question about the "MAXTimesteps" flag in CORREL. The documentation says "The MAXTimesteps value is the largest number of steps any
time series will contain." But I still don't understand whether it corresponds to the number of snapshorts or the MD steps in the trajectory. Say I have a trajectory with 1500000 MD steps. As the printing frequency is 500 MD steps thus I have 3000 snapshorts in total. But I don't know which value should be specified for "MAXTimesteps". I tried both and they gave the same outputs. Hope someone can help. Thanks in advance!

Best
Posted By: shane yue Re: MAXTimesteps in CORREL - 07/07/14 02:30 PM
An update. Seems that "MAXTimesteps" corresponds to the number of snapshots. If I specified the flag with any number smaller than the number of snapshorts in the trajectory read in, CHARMM died due to the error:
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
***** LEVEL -2 WARNING FROM *****
***** TOO MANY TIME STEPS. REMAINDER IGNORED
******************************************
BOMLEV ( 0) IS REACHED - TERMINATING. WRNLEV IS 5
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

I adapte my script from the one wrote by Lennart Nilsson and updated by Jianhan Chen (http://mmtsb.org/workshops/mmtsb-ctbp_wo...ial.html#orient). That script wrote
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
READ TOP/PAR, PSF/PDB, DCD ...

! by default, let's look at Gly41 (the most flexible residue from nmr
! relaxation analysis; see nmr.inp and plot_nmr-s2.gnu)
if @?rid eq 0 set rid = 41

!Open the output file and write header
open write unit 11 card name phipsi-dist.dat
write title unit 11
*#phi and psi distribution of residue: @rid
*#angle p_phi p_psi
*

open write unit 12 card name phipsi.dat
write title unit 12
*#phi and psi time series of residue: @rid
*#time phi psi
*

!Invoke CORREL mode
correl maxtime ?nfile maxseries 2 maxatoms 10

COMPUT TIME SERIES AND HISTOGRAMS
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
When I directly copied the command line, the "?nfile" was substituted by 1000 and killed the CHARMM. I do not know where "?nfile" came from and what it referred to. I searched subst.doc but found nothing.

Does anyone have any idea? Thanks again.

Best
Posted By: rmv Re: MAXTimesteps in CORREL - 07/07/14 04:29 PM
See subst.doc
Posted By: shane yue Re: MAXTimesteps in CORREL - 07/07/14 04:58 PM
Thanks Rick.

I double checked subst.doc. ?nfile is the number of frames in the most recently read trajectory file. That explains my problem. Thanks again.

Best
© CHARMM forums