Grepsel - Revision history

Bell: 6 revisions

2014-03-28T03:06:30Z

6 revisions

Pyadmin: 1 revision

2014-03-28T02:16:15Z

1 revision

Speleo3: fix selection argument

2011-10-06T18:36:03Z

fix selection argument

New page

Create selections matching motifs, using python regular expression syntax.

This is very similar to the [[FindSeq]] script.

== The Code ==

<source lang="python">
#Create named selections using regular expressions for the protein sequence

import pymol
import re
cmd = pymol.cmd

aa = { 'ASP' : 'D' , 'GLU' : 'E' , 'GLN' : 'Q' , 'ASN' : 'N' , 'SER' : 'S' ,
'THR' : 'T' , 'CYS' : 'C' , 'HIS' : 'H' , 'ARG' : 'R' , 'LYS' : 'K' ,
'MET' : 'M' , 'ALA' : 'A' , 'ILE' : 'I' , 'LEU' : 'L' , 'VAL' : 'V' ,
'GLY' : 'G' , 'PRO' : 'P' , 'TRP' : 'W' , 'PHE' : 'F' , 'TYR' : 'Y' ,
'SCY' : 'U' , 'ASX' : 'B' , 'GLX' : 'Z' , 'XXX' : 'X'}

#made this before the sequence view option, probably another way to do it now

def seqoneint(model):
pymol.stored.seq = []
cmd.iterate("%s and name ca"%model,"stored.seq.append(resn)")
seq = ""
for x in pymol.stored.seq:
if aa.has_key(x):
res = aa[x]
seq = seq+res
else:
seq = seq + '-'
return seq

def grepsel(selection="(all)",stretch="",prefix="",combined="0",single="1"):
'''
DESCRIPTION

Create selections matching motifs, using python regular expression syntax.
Motif is automatically converted to uppercase. Motif selections are labelled
as "prefix_motif_###", where ### is the index for the first residue of the
match. Prefix defaults to selection name. combined = 1 creates one selection
for all occurences. single = 1 creates one selection for each occurance
(the default).

USAGE

grepsel selection, motif, [prefix, [combined, [single ]]]

EXAMPLES

Create selections for all motifs matching "ESS" (selection_ESS_###,...):
grepsel selection, ess

Create selections for the PxGY motif with prefix m (m_P.CY_###,...):
grepsel selection, p.gy, m
'''

if selection == "(all)":
selection = "all"
if prefix == "":
prefix=selection

stretch = stretch.upper()
seq = seqoneint(selection)
pymol.stored.resi = []
pymol.stored.chain = []
cmd.iterate("%s and name ca"%selection,"stored.resi.append(resi);stored.chain.append(chain)")
motif = re.compile(stretch)
occurrences = motif.finditer(seq)
stretchmod = stretch.replace("+","\+")
stretchmod = stretchmod.replace("?","\?")

print stretchmod
if combined == "1":
cmd.select("%s_%s"%(prefix,stretch), "none")

for find in occurrences:

mb = pymol.stored.resi[find.start()]
me = pymol.stored.resi[find.end()-1]

ch = pymol.stored.chain[find.start()]
cmd.select("%s_%s_%s%s"%(prefix,stretch,me,ch), "chain %s and (i; %s-%s)"%(ch,int(mb),int(me)))
if combined == "1":
cmd.select("%s_%s"%(prefix,stretch),"\"%s_%s\" | (%s and chain %s and (i; %s-%s))"%(prefix,stretchmod,selection,ch,int(mb),int(me)))

cmd.select("none")
cmd.delete("sel*")

cmd.extend("grepsel",grepsel)
</source>

== See Also ==

* [[FindSeq]]
* [[Selection Algebra|pepseq]] selection operator
* [http://pldserver1.biochem.queensu.ca/~rlc/work/pymol/ seq_select] by Robert Campbell

[[Category:Script_Library|Grep selections]]
[[Category:ObjSel_Scripts]]

Gawells: A bit cleaner, fixed problems with non-greedy metacharacters

2005-11-28T08:05:42Z

A bit cleaner, fixed problems with non-greedy metacharacters

Gawells at 08:01, 28 November 2005

2005-11-28T08:01:39Z

Gawells: fixed selection from "chainless" models

2005-09-10T10:24:37Z

fixed selection from "chainless" models

Gawells at 16:11, 20 July 2005

2005-07-20T16:11:33Z

Gawells at 16:09, 20 July 2005

2005-07-20T16:09:54Z

New page

<source lang="python">
#Create named selections using regular expressions for the protein sequence

import pymol
import re

aa = { 'ASP' : 'D' , 'GLU' : 'E' , 'GLN' : 'Q' , 'ASN' : 'N' , 'SER' : 'S' ,
'THR' : 'T' , 'CYS' : 'C' , 'HIS' : 'H' , 'ARG' : 'R' , 'LYS' : 'K' ,
'MET' : 'M' , 'ALA' : 'A' , 'ILE' : 'I' , 'LEU' : 'L' , 'VAL' : 'V' ,
'GLY' : 'G' , 'PRO' : 'P' , 'TRP' : 'W' , 'PHE' : 'F' , 'TYR' : 'Y' ,
'SCY' : 'U' , 'ASX' : 'B' , 'GLX' : 'Z' , 'XXX' : 'X'}

#made this before the sequence view option, probably another way to do it now

def seqoneint(model):
pymol.stored.seq = []
cmd.iterate("%s and name ca"%model,"stored.seq.append(resn)")
seq = ""
for x in pymol.stored.seq:
if aa.has_key(x):
res = aa[x]
seq = seq+res
else:
seq = seq + '-'
return seq

def grepsel(model="(all)",stretch="",prefix="",combined="0",single="1"):
'''
DESCRIPTION

Make selections matching regular expressions. Selections are labelled
as "prefix_expression_###", where ### is the index for the first residue
of the match. Prefix defaults to model name. combined = 1 creates one
selection for all occurences. single = 1 creates one selection for each
occurance (the default).

USAGE

grepsel selection, expression, [prefix, [combined, [single ]]]

EXAMPLES

Make selections for all motifs matching "ESS" (model_ESS_###,...):
grepsel model, ess

Make selections for the PxGY motif with prefix m (m_P.CY_###,...):
grepsel model, p.gy, m

'''

if model == "(all)":
model = "all"
if prefix == "":
prefix=model

stretch = stretch.upper()
seq = seqoneint(model)
pymol.stored.resi = []
pymol.stored.chain = []
cmd.iterate("%s and name ca"%model,"stored.resi.append(resi);stored.chain.append(chain)")
motif = re.compile(stretch)
occurrences = motif.finditer(seq)
if combined == "1":
cmd.do("select %s_%s, none"%(prefix,stretch))

for match in occurrences:
fx = match.start()
pos = 0
for fy in range(int(fx),int(fx)+len(stretch)):
ch = pymol.stored.chain[fy]
ri = pymol.stored.resi[fy]
if pos == 0:
pos = ri
if single == "1":
cmd.select("%s_%s_%s%s"%(prefix,stretch,pos,ch), "none")
if single == "1":
cmd.select("%s_%s_%s%s"%(prefix,stretch,pos,ch), "%s_%s_%s%s | %s/%s/"%(prefix,stretch,pos,ch,ch,ri))
if combined == "1":
cmd.select("%s_%s"%(prefix,stretch), "%s_%s | %s//"%(prefix,stretch,ri))
cmd.select("none")
cmd.delete("sel*")

cmd.extend("grepsel",grepsel)
</source>

[[Category:Script_Library|grep selections]]

← Older revision		Revision as of 16:11, 20 July 2005
Line 88:		Line 88:


−	[[Category:Script_Library\|~~grep~~ selections]]	+	[[Category:Script_Library\|Grep selections]]

← Older revision	Revision as of 03:06, 28 March 2014
(No difference)

← Older revision	Revision as of 02:16, 28 March 2014
(No difference)

@@ Line 30: / Line 30: @@
     '''
 DESCRIPTION
 Create selections matching motifs, using python regular expression syntax.
      Motif is automatically converted to uppercase. Motif selections are labelled
      as "prefix_motif_###", where ### is the index for the first residue of the
@@ Line 39: / Line 40: @@
 USAGE
-     grepsel selection, expression, [prefix, [combined, [single ]]]
+     grepsel selection, motif, [prefix, [combined, [single ]]]
 EXAMPLES
-     Make selections for all motifs matching "ESS" (model_ESS_###,...):
+     Create selections for all motifs matching "ESS" (selection_ESS_###,...):
-     grepsel model, ess
+     grepsel selection, ess
-     Make selections for the PxGY motif with prefix m (m_P.CY_###,...):
+     Create selections for the PxGY motif with prefix m (m_P.CY_###,...):
-     grepsel model, p.gy, m
+     grepsel selection, p.gy, m
      '''

@@ Line 30: / Line 30: @@
     '''
 DESCRIPTION
+Create selections matching motifs, using python regular expression syntax.
-    Make selections matching regular expressions. Selections are labelled
+    Motif is automatically converted to uppercase. Motif selections are labelled
-     as "prefix_expression_###", where ### is the index for the first residue
+     as "prefix_motif_###", where ### is the index for the first residue of the
-     of the match. Prefix defaults to model name. combined = 1 creates one
+     match. Prefix defaults to selection name. combined = 1 creates one selection
-     selection for all occurences. single = 1 creates one selection for each
+     for all occurences. single = 1 creates one selection for each occurance
-     occurance (the default).
+     (the default).
 USAGE
@@ Line 50: / Line 50: @@
      '''
-−
+    if selection == "(all)":
-    if model == "(all)":
+       selection = "all"
     if prefix == "":
-       prefix=model
+       prefix=selection
     stretch = stretch.upper()
-    seq = seqoneint(model)
+    seq = seqoneint(selection)
     pymol.stored.resi = []
     pymol.stored.chain = []
-    cmd.iterate("%s and name ca"%model,"stored.resi.append(resi);stored.chain.append(chain)")
+    cmd.iterate("%s and name ca"%selection,"stored.resi.append(resi);stored.chain.append(chain)")
     motif = re.compile(stretch)
     occurrences = motif.finditer(seq)
     if combined == "1":
-       cmd.do("select %s_%s, none"%(prefix,stretch))
+       cmd.select("%s_%s"%(prefix,stretch), "none")
     cmd.select("none")
     cmd.delete("sel*")
-−
 cmd.extend("grepsel",grepsel)
 </source>

@@ Line 78: / Line 78: @@
                 cmd.select("%s_%s_%s%s"%(prefix,stretch,pos,ch), "none")
           if single == "1":
-             cmd.select("%s_%s_%s%s"%(prefix,stretch,pos,ch), "%s_%s_%s%s | %s/%s/"%(prefix,stretch,pos,ch,ch,ri))
+             cmd.select("%s_%s_%s%s"%(prefix,stretch,pos,ch), "%s_%s_%s%s | ///%s/%s/"%(prefix,stretch,pos,ch,ch,ri))
           if combined == "1":
-             cmd.select("%s_%s"%(prefix,stretch), "%s_%s | %s//"%(prefix,stretch,ri))
+             cmd.select("%s_%s"%(prefix,stretch), "%s_%s | ////%s/"%(prefix,stretch,ri))
     cmd.select("none")
     cmd.delete("sel*")