def kegg_converter():
"""process list of uniprot accessions for KEGG pathway analysis"""
from root.ed.tools import prot_id_converter
protList = []
headerFlag = True
with open("../bob/processed/24h_bobprots_up_full.csv","r") as inpF:
for inpLine in inpF:
if headerFlag:
headerFlag = False
continue
inpList = inpLine.split(",")
protList.append(inpList[1])
print prot_id_converter(protList, outDB="kegggeneid")
def interactor_finder():
"""take a list of protein names and check if they are in Bob's dataset"""
from root.ed.tools import prot_id_converter
proteinList = []
with open("../datafiles/known_interactors.txt","r") as inpProt: # create list of gene names from hand-made text file with known ptp22 interactors
for protLine in inpProt:
if protLine != "\n":
curName = protLine.strip().split("\t")[0]
curName = curName[0] + curName[1:].lower()
proteinList.append(curName)
inpIdL = prot_id_converter(proteinList, "10090", "genesymbol", "uniprotaccession") # convert to uniprot accessions
print inpIdL
with open("../bob/processed/bobprots_all.csv","r") as targetF: # create list of all uniprot accessions in Bob's dataset (unique razor proteins only)
targetD = {}
for targetLine in targetF:
targetD[targetLine.split(",")[0]] = targetLine.split(",")[1].strip()
for inpIdItem in inpIdL:
for queryI in inpIdItem:
if queryI in targetD:
print targetD[queryI]
break
def intact_parser():
"""open ptpn22.txt and extract prey protein uniprot accessions.
Convert those to refseq protein accessions.
Return them as a list."""
from root.ed.tools import prot_id_converter, file_importer
relPath = "ptpn22_ppi_data/ptpn22.txt"
inpF = file_importer(relPath, "r")
headerFlag = True
preyL = []
for inpLine in inpF:
if headerFlag:
headerFlag = False
continue
inpList = inpLine.split("\t")
inpItem = inpList[1].split(":")[-1]
if inpItem not in preyL:
preyL.append(inpItem)
inpItem = inpList[0].split(":")[-1]
if inpItem not in preyL:
preyL.append(inpItem)
inpF.close()
idList = prot_id_converter(preyL, "", outDB="refseqproteingi") # convert uniprot ID to refseq accessions
return idList
def main():
from root.ed.tools import html_creator, prot_id_converter, prot_entrez_fetch
from copy import deepcopy
print "this is peptide parser"
queryS = "Orc6" # this is the search term that will be worked on. It should be a protein name like "Ptpn22"
print "working on: " + queryS
with open("../bob/peptides.txt","r") as inpF:
pepL, uniId = peptide_finder(targetFile=inpF, targetS = queryS) # find peptides from peptides.txt for the protein name
print pepL
print "peptides found"
targetL = [queryS]
idList = prot_id_converter(targetL, "10090", inpDB = "genesymbol",outDB="refseqproteingi")
seqL = prot_entrez_fetch(idList, retM="gb", retT="fasta")
for seqItem in seqL:
seqS = seqItem.split("\n")[1]
print seqS
print "protein sequence found"
annotS = deepcopy(seqS)
pStartL = []
pEndL = []
for pepItem in pepL: # locate peptides in full protein sequence and store positions for starts and ends. merge overlapping peptides.
pepStart = seqS.index(pepItem)
pepEnd = pepStart + len(pepItem)
startCount = 0 # handle starts
for startItem in pStartL:
if startItem <= pepStart:
startCount += 1
endCount = 0
for endItem in pEndL:
if endItem <= pepStart:
endCount += 1
if startCount == endCount and pepStart not in pEndL: # start new peptide
pStartL.append(pepStart)
elif startCount == endCount and pepStart in pEndL: # start a new peptide at the end of another peptide
pEndL.remove(pepStart)
overlapCount = 0
for startItem in pStartL[:]: # handle ends
if pepStart<startItem<=pepEnd:
pStartL.remove(startItem) # remove extra starts
overlapCount += 1
for endItem in pEndL[:]:
if pepStart<=endItem<=pepEnd: # remove extra ends
pEndL.remove(endItem)
overlapCount -= 1
if pepEnd not in pEndL and overlapCount <= 0: # add end
curStart = 500000
for pSI in pStartL:
if curStart > pSI > pepEnd:
curStart = pSI
curEnd = 500000
for pEI in pEndL:
if curEnd > pEI > pepEnd:
curEnd = pEI
if curStart <= curEnd: # check if next tag is start or end. if start, add end. if end, do nothing
pEndL.append(pepEnd)
print uniId
phL = []
with open("../datafiles/Phosphorylation_site_dataset") as phInp: # now for the phosphosite data
for phLine in phInp:
phList = phLine.split("\t")
try:
if uniId == phList[1]:
phL.append(int(phList[4][1:-2]) - 1)
except IndexError:
continue
print phL
fullL = pStartL + pEndL
for phItem in phL:
if phItem not in fullL: fullL.append(phItem)
fullL.sort()
offsetN = 0
for posI in fullL: # from the resulting intervals, create emphasis in html file, mark phosphosites in red
if posI in pStartL:
annotS = annotS[:posI+offsetN] + "<mark>" + annotS[posI+offsetN:]
offsetN += 6
elif posI in pEndL:
annotS = annotS[:posI+offsetN] + "</mark>" + annotS[posI+offsetN:]
offsetN += 7
if posI in phL:
annotS = annotS[:posI+offsetN] + r"""<strong style="color: red;">""" + annotS[posI+offsetN] + r"""</strong>""" + annotS[posI+offsetN + 1 :]
offsetN += 37
print annotS
html_creator(queryS + " peptides", annotS, queryS + ".html")
print "found peptides marked in the file: ",
print queryS + ".html"
def sh3_counter():
"""look up a list of uniprot IDs, download their full genbank entries from the Entrez database
and count the number of SH3 domains the interactors have. Print the results to STDout"""
from root.ed.tools import prot_entrez_fetch, prot_id_converter
from root.ed.bobscripts.bobdata_parser import protein_name_collector
# idList = intact_parser() # to use for Ptpn22 interactome
fullPreyL = protein_name_collector()
# fullPreyL = ["P20152", "Q8BFZ3", "P17182", "P17742", "P11499"]
print fullPreyL
if len(fullPreyL) > 200: # chop up very large lists of uniprot IDs to batches of 100
maxBatch = (len(fullPreyL)/200) + 1
lenCount = 0
idList = []
seqL = []
batchCount = 0
preyL = []
for listItem in fullPreyL:
preyL.append(listItem)
lenCount += 1
if lenCount == 200:
batchCount += 1
print "processing batch number %d of Uniprot IDs..." % (batchCount, )
idList = prot_id_converter(preyL, "10090", outDB="refseqproteingi")
seqL = seqL + prot_entrez_fetch(idList, retM="gb", retT="text").split("\n") # fetch the complete genbank entries from entrez using this function from tools.py
lenCount = 0
idList = []
preyL = []
print "this was batch number %d of %d" %(batchCount, maxBatch)
print ""
if lenCount != 0:
batchCount += 1
print "processing batch number %d of Uniprot IDs..." % (batchCount, )
idList = prot_id_converter(preyL, "10090", outDB="refseqproteingi")
seqL = seqL + prot_entrez_fetch(idList, retM="gb", retT="text").split("\n") # fetch the complete genbank entries from entrez using this function from tools.py
lenCount = 0
idList = []
preyL = []
print "this was batch number %d of %d" %(batchCount, maxBatch)
print ""
else:
idList = prot_id_converter(fullPreyL, "10090", outDB="refseqproteingi")
seqL = prot_entrez_fetch(idList, retM="gb", retT="text").split("\n") # fetch the complete genbank entries from entrez using this function from tools.py
"""
dumpStr = ""
for seqLine in seqL:
seqStr = "***".join(seqLine)
dumpStr = dumpStr + "xxxxx" + seqStr # this might get huge
outputDump = open("entrezdump.txt", "w")
outputDump.write(dumpStr)
"""
regionFlag = False
sHFlag = True
regionCount = 0
sHCount = 0
protCount = 0
shProtCount = 0
for flatLine in seqL:
if flatLine[:10] == "LOCUS ": # this is usually the title of an entrez flatfile and contains the protien name
protCount+= 1
newProtFlag = True
# curProt = flatLine
if flatLine[:11] == " Region": # regions like this mark domains in the flatfile. Look here for SH3 domains
regionCount += 1
regionFlag = True
continue
if regionFlag:
if flatLine[:11] == " ":
if "SH3" in flatLine and sHFlag:
sHCount += 1
sHFlag = False
if newProtFlag:
shProtCount += 1
newProtFlag = False
# print flatLine
# print curProt
else:
regionFlag = False
sHFlag = True
print "%d SH3 domains found in %d domains of %d proteins" % (sHCount, regionCount, protCount)
print "%d proteins out of %d contain SH3 domains" % (shProtCount, protCount)