Dear Dr. Stránský, Thanks a lot for the scripts. These will be really helpful in speeding up the substructure solution and phasing trials. Manually changing the parameters and doing multiple trials is a pain.
Thanks again, Best Regards, Arko On Tue, Jul 29, 2014 at 4:34 PM, Jan Stránský <[email protected]> wrote: > Dear Arko, > I have done similar think recently. I have used "cat" for changing input > files. Script "run_shelxd" takes an argument, which uses as prefix and run > SHELXD with various options of SHEL and SEED. It outputs > "prefix_progress.log" for control of done processes and for pause (you can > stop the script by placing file called "term" into the running folder) and > resume. > Second handy thing is runnig various SHELXE parallely, since it is only > one-threaded application. You need to run it in BASH, because for example > tcsh handles jobs differently. > See both scripts enclosed, at least for inspiring. > > Best regards > > Jan > > ---------- Původní zpráva ---------- > Od: Arka Chakraborty <[email protected]> > Komu: [email protected] > Datum: 28. 7. 2014 14:19:08 > Předmět: Re: [ccp4bb] SCRIPT FOR RUNNING MULTIPLE RUNS OF SHELX C DE FOR > RIP > > Dear Dr. Tim Gruene, > > Thanks a lot for helping me out. My limited knowledge of shell scripting > failed me. I will try it out. > > Best Regards, > > Arko > > > On Mon, Jul 28, 2014 at 1:49 PM, Tim Gruene <[email protected]> > wrote: > > Dear Arka Chakraborty, > > such a task is best done with shell tools. Assuming you have a template > input file shelxc.inp with a line "DCSA 0.98", you can create the new > ones with > > #> for i in $(seq -w 90 99); do > dsca=$(echo "scale = 3; $i / 100" | bc); > sed "s/DSCA 0.98/DSCA $dsca/" shelxc.inp > shelxc_${i}.inp; > done > > and run shelxc with > for i in shelxc_*.inp; do > shelxc ${i%.inp} < $i | tee ${i%.inp}.log > done > > and shelxd with > for i in shelxc_*_fa.ins; do > shelxd ${i%.ins} > done > > If you have access to several machines, you can wrap the last, > time-consuming step into a shell script that changes into your working > directory to start shelxd. Then you can add the script to ssh in order > to remotely execute shelxd. > > I did not explicitly test the scripts, but except for typos which should > be easy to correct, they should work. > > Regards, > Tim > > On 07/28/2014 01:10 PM, Arka Chakraborty wrote: > > Dear CCPers, > > > > I am running multiple runs of shelx c d e for a RIP dataset. Does anyone > > happen to have a script for doing 100s of runs varying the DSCA parameter > > for downscaling the 'after' dataset?. It will be of great help to me if > you > > could kindly share it. > > Also if you have past experience solving structures with RIP in low > > symmetry sg, eg. P1211 at ~2.9 ang resolution (according to I/sigmaI =1.9 > > in the highest shell, CC half around 0.85) and would kindly share it that > > will be great. > > > > Thanks a lot, > > > > Best Regards, > > > > Arka Chakraborty > > > > -- > Dr Tim Gruene > Institut fuer anorganische Chemie > Tammannstr. 4 > D-37077 Goettingen > > GPG Key ID = A46BEE1A > > > > > -- > *Arka Chakraborty* > *ibmb (Institut de Biologia Molecular de Barcelona)* > *BARCELONA, SPAIN* > > -- *Arka Chakraborty* *ibmb (Institut de Biologia Molecular de Barcelona)* *BARCELONA, SPAIN*
