-
Notifications
You must be signed in to change notification settings - Fork 1
/
craftToTab
executable file
·187 lines (164 loc) · 7.3 KB
/
craftToTab
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
#!/usr/bin/env python
import logging, sys, optparse, glob, marshal
import xml.etree.ElementTree as et
from os.path import join, basename, isfile
from collections import defaultdict
# === COMMAND LINE INTERFACE, OPTIONS AND HELP ===
parser = optparse.OptionParser("""usage: %prog [options] type inDirOrFile - convert CRAFT corpus standoff XML to tab-sep format, prints to stdout
type can be "prot" or "gene"
for entrez genes:
requires geneToSym.tab in the same dir, create it like this:
wget ftp://ftp.ncbi.nlm.nih.gov/gene/DATA/gene_info.gz -O - | gunzip | cut -f1,2,3 > geneToSym.tab
cat geneToSym.tab | gawk '($1=="9606")' > geneToSym.9606.tab
also requires homologene.data:
wget ftp://ftp.ncbi.nih.gov/pub/HomoloGene/current/homologene.data
for proteins:
requires uniprotmapping.txt:
wget ftp://ftp.pir.georgetown.edu/databases/ontology/pro_obo/PRO_mappings/uniprotmapping.txt
""")
parser.add_option("-d", "--debug", dest="debug", action="store_true", help="show debug messages")
#parser.add_option("-f", "--file", dest="file", action="store", help="run on file")
#parser.add_option("", "--test", dest="test", action="store_true", help="do something")
(options, args) = parser.parse_args()
if options.debug:
logging.basicConfig(level=logging.DEBUG)
else:
logging.basicConfig(level=logging.INFO)
# ==== FUNCTIONs =====
def convertEntrez(tree, entrezToSymTax, groupToGenes, geneToGroups):
mentions = {}
for annotEl in tree.findall("annotation"):
mentionEl = annotEl.find("mention")
instId = mentionEl.attrib["id"]
spanEl = annotEl.find("span")
start = spanEl.attrib.get("start")
end = spanEl.attrib.get("end")
spannedText = annotEl.find("spannedText").text
mentions[instId] = (start, end, spannedText)
slotToInstance = {}
for classEl in tree.findall("classMention"):
instanceId = classEl.attrib.get("id")
slotId = classEl.find("hasSlotMention").attrib.get("id")
slotToInstance[slotId] = instanceId
for slotEl in tree.findall("integerSlotMention"):
slotId = slotEl.attrib.get("id")
instId = slotToInstance[slotId]
start, end, spanText = mentions[instId]
entrezId = int(slotEl.find("integerSlotMentionValue").attrib.get("value"))
sym, taxId = entrezToSymTax.get(entrezId, ("VALID_GENE?", "VALID_GENE?"))
homoloGroup = geneToGroups.get(entrezId, None)
humanGeneIdStr, humanSymStr = "Unknown", "Unknown"
if homoloGroup!=None:
humanGeneIds = groupToGenes[homoloGroup].get(9606, ["NO_HOMOLOG"])
humanSyms = [entrezToSymTax.get(hGene, ("VALID_HUMAN_GENE?", 0))[0] for hGene in humanGeneIds]
humanGeneIds = [str(x) for x in humanGeneIds]
humanGeneIdStr = "|".join(humanGeneIds)
humanSymStr = "|".join(humanSyms)
row = [ pmid, entrezId, sym, taxId, humanGeneIdStr, humanSymStr, \
spanText.encode("utf8"), start, end]
row = [str(x) for x in row]
print "\t".join(row)
def convertProt(tree, protToUp, upToEntrezList, entrezToSymTax):
# <classMention id="PRO_Instance_30000">
# <mentionClass id="PR:000004918">(PR) carbonic anhydrase 2 ({carbonic anhydrase C, CA-II, carbonate dehydratase II, carbonic anhydrase II}{CAC, CA2, Car2})</mentionClass>
# </classMention>
classToProtOnt = {}
for classEl in tree.findall("classMention"):
classId = classEl.attrib.get("id")
protOntId = classEl.find("mentionClass").attrib.get("id")
classToProtOnt[classId] = protOntId
# <annotation>
# <mention id="PRO_Instance_30000" />
# <annotator id="PRO_Instance_400000">CCP Colorado Computational Pharmacology</annotator>
# <span start="807" end="828" />
# <spannedText>carbonic anhydrase II</spannedText>
# </annotation>
for annotEl in tree.findall("annotation"):
mentionEl = annotEl.find("mention")
mentionId = mentionEl.attrib["id"]
spanEl = annotEl.find("span")
start = spanEl.attrib.get("start")
end = spanEl.attrib.get("end")
spannedText = annotEl.find("spannedText").text.encode("utf8")
protOntId = classToProtOnt[mentionId]
#print protOntId
upIds = protToUp[protOntId]
for upId in upIds:
if not upId in upToEntrezList:
continue
entrezIds = upToEntrezList[upId]
for entrezId in entrezIds:
sym, taxId = entrezToSymTax.get(int(entrezId), ("UNKNOWN_SYMBOL", 0))
row = [pmid, entrezId, sym, protOntId, upId, spannedText, start, end]
row = [str(x) for x in row]
print "\t".join(row)
def parseEntrez(fname):
# parse entrez gene info: taxon and symbol for each gene
logging.info("Parsing entrez genes")
entrezToSymTax = {}
for line in open(fname):
if line.startswith("#"):
continue
tax, gene, sym = line.rstrip("\n").split("\t")
entrezToSymTax[int(gene)] = (sym, int(tax))
return entrezToSymTax
# ----------- MAIN --------------
if args==[]:
parser.print_help()
exit(1)
dataType, inDir = args
if dataType=="gene":
entrezToSymTax = parseEntrez("geneToSym.tab")
logging.info("Parsing homologene")
# parse homologene info
# like this:
# 3 9606 34 ACADM 187960098 NP_001120800.1
geneToGroups = dict()
groupToGenes = defaultdict(dict)
for line in open("homologene.data"):
if line.startswith("#"):
continue
group, taxonId, geneId, symbol, protGi, protAcc = line.rstrip("\n").split("\t")
if taxonId not in ['9606', '10090']:
continue
group, geneId, taxonId = int(group), int(geneId), int(taxonId)
geneToGroups[geneId] = group
groupToGenes[group].setdefault(taxonId, []).append(geneId)
headers = ["pmid", "geneId", "sym", "taxonId", "humanGenes", "humanSyms", "text", "start", "end"]
elif dataType=="prot":
# example lines:
# PR:000000049 UniProtKB:Q62312-1
# PR:000000050 UniProtKB_VAR:VAR_022351
entrezToSymTax = parseEntrez("geneToSym.9606.tab")
protToUp = defaultdict(list)
logging.info("Parsing protein ontology -> uniprot mapping")
for line in open("uniprotmapping.txt"):
protId, upId = line.rstrip("\n").split("\t")
if "_VAR" in upId:
continue
#protId = protId.split(":")[1]
upId = upId.split(":")[1]
protToUp[protId].append(upId)
# parse up -> entrez gene
_pubsDir = "/hive/data/inside/pubs"
geneDataDir = _pubsDir+"/geneData"
GENEDATADIR=geneDataDir
fname = join(GENEDATADIR, "uniprot.tab.marshal")
logging.info("Loading %s" % fname)
data = marshal.load(open(fname))[9606]
upToEntrezList = data["upToEntrez"]
headers = ["pmid", "geneId", "sym", "protOntId", "uniprotId", "text", "start", "end"]
print "\t".join(headers)
if isfile(inDir):
fnames = [inDir]
else:
fnames = glob.glob(join(inDir, "*.xml"))
for fname in fnames:
pmid = basename(fname).split(".")[0]
tree = et.parse(fname)
root = tree.getroot()
logging.debug("processing %s" % pmid)
if dataType=="gene":
convertEntrez(tree, entrezToSymTax, groupToGenes, geneToGroups)
elif dataType=="prot":
convertProt(tree, protToUp, upToEntrezList, entrezToSymTax)