Hi, Is it possible in a modeller top file to calculate, let's say 100 models, and then tell modeller to refine the best 10 and then to add loops for the best 5 (all in one run). Thanks, Dan
-- Dan Thomas Major (at Dr. B. Fischer's lab) Bar-Ilan University Ramat-Gan, Israel
Yes, but it would require running a unix script out of the top file to do the sorting and file renaming.
Andrej
-----Original Message----- From: owner-modeller_usage@salilab.org [mailto:owner-modeller_usage@salilab.org] On Behalf Of Dan Thomas Major Sent: Thursday, January 10, 2002 6:53 AM To: modeller_usage Subject: stepwise top
Hi, Is it possible in a modeller top file to calculate, let's say 100 models, and then tell modeller to refine the best 10 and then to add loops for the best 5 (all in one run). Thanks, Dan
This might help: use this as a shell script replace $i with model name. This will find the best three models and move them into a directory best3 and superimpose them. It will also move the best model into directory best1. You will just need to replace the impose section with loop modeling and add the 4th and 5th models.
>>>>
grep OBJECTIVE $i.B9999????|cut -d':' -f1 > temp1 grep OBJECTIVE $i.B9999????|cut -d' ' -f10 > temp2
paste temp1 temp2 >temp3 sort -n +1 temp3 >temp4 sed -n '1,3w temp5' temp4 cut -f1 temp5 >temp6 sed -n '1w temp6a' temp6 sed -n '2w temp6b' temp6 sed -n '3w temp6c' temp6 read SEQ1 < temp6a read SEQ2 < temp6b read SEQ3 < temp6c echo $SEQ1 echo $SEQ2 echo $SEQ3 cp $SEQ1 best1/$i.pdb echo "Finished" rm temp1 temp2 temp3 temp4 temp5 temp6 temp6a temp6b temp6c echo "#Superimpose models
SET ATOM_FILES = '$SEQ1' '$SEQ2' '$SEQ3'
SET WRITE_FIT = ON MALIGN3D" > impose$i.top mod4 impose$i.top cp $SEQ1 $SEQ2 $SEQ3 best3 cp $i.imposed impbest3 <<<<<<
At 09:19 AM 1/10/02 -0500, Andrej Sali wrote:
> > Yes, but it would require running a unix script out of the top file to do the > sorting and file renaming. > > > > Andrej > > > > -----Original Message----- > From: owner-modeller_usage@salilab.org > [mailto:owner-modeller_usage@salilab.org] On Behalf Of Dan Thomas > Major > Sent: Thursday, January 10, 2002 6:53 AM > To: modeller_usage > Subject: stepwise top > > > > Hi, > Is it possible in a modeller top file to calculate, let's say 100 models, and > then tell modeller to refine the best 10 and then to add loops for the best 5 > (all in one run). > Thanks, > Dan > > -- Dan Thomas Major (at Dr. B. Fischer's lab) Bar-Ilan University Ramat-Gan, > Israel
************************************************ Michael Buck NCSU Genetics mjbuck@unity.ncsu.edu Phone (919)515-5759 Fax (919)515-3355 http://www4.ncsu.edu/~mjbuck *************************************************
participants (3)
-
Andrej Sali
-
Dan Thomas Major
-
Michael Buck