-
Notifications
You must be signed in to change notification settings - Fork 0
/
ublastn.2.0.py
211 lines (146 loc) · 5.9 KB
/
ublastn.2.0.py
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
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
from Bio import SeqIO
import sys
import os
from StringIO import StringIO # Python 2
import subprocess
from random import randrange
from optparse import OptionParser
parser = OptionParser()
parser.add_option("-i", "--infile", dest="filename", help="input sequence file", metavar="FILE")
parser.add_option("-o", "--outfile", dest="outfile", default="ublast.out", help="output file name")
parser.add_option("-t", "--threads", dest="threads", default="8", help="output file name")
parser.add_option("-f", "--format", dest="infmt", default="fasta", help="input file format, e.g. fasta, fastq, genbank")
parser.add_option("-m", "--maxhits", dest="maxhits", default="10", help="max number of hits to return for each query")
parser.add_option("-d", "--identity", dest="id", default="0.9", help="percent identity threshold")
parser.add_option("-b", "--database", dest="db", default="/db/gbnt.udb", help="database, default = /db/gbnt.udb")
parser.add_option("-e", "--evalue", dest="evalue", default="1.0e-10", help="e value threshold")
parser.add_option("-c", "--cons", dest="cons", default="0.75", help="specify consensus taxon scoring threshold 0 to 1, default = 0.75")
(options, args) = parser.parse_args()
r2=str(randrange(100000))
db=options.db
dbname = db.split("/")[-1]
filesize=int(os.path.getsize(db))
if filesize < 10000000000:
if os.path.isfile('/db/%s' %dbname)== False:
print "loading db to ram"
subprocess.Popen("sudo mkdir -p /db && sudo mount -t tmpfs tmpfs /db && sudo chmod 777 /db && cp -n %s /db/%s" %(db,dbname), shell=True).wait()
db = "/db/%s" %dbname
else:
db = "/db/%s" %dbname
print "db already exists"
else:
print "db too large for memory.. usearch to use directly, will be a little slower", "size = ", str(filesize/1000000000),"Gb"
def chunkstring(string, length):
return (string[0+i:length+i] for i in range(0, len(string), length))
#count reads
inputfile = options.filename #sys.argv[1]
outputfile=options.outfile #sys.argv[2]
threads = int(options.threads)
fmt = options.infmt
maxhits= options.maxhits
id= options.id
evalue=options.evalue
cons = float(options.cons)
fetchsize = 10000
maxacc = 5
subprocess.Popen("rm -rf ~/scripts/tmp2", shell=True).wait()
subprocess.Popen("mkdir ~/scripts/tmp2", shell=True).wait()
out1=open('/OSM/HOME-MEL/all29c/scripts/tmp2/finaloutput%s.ublast' %(r2),'w')
f2=open('/OSM/HOME-MEL/all29c/scripts/tmp2/inp%s.fa' %(r2),'w')
#split large reads/contigs if >50kb
print "splitting large seq records if >50kb..."
for i in SeqIO.parse(inputfile,fmt):
if len(i.seq)>50000:
seqlist=list(chunkstring(i.seq,50000))
c=-1
for t in seqlist:
c=c+1
print i.id, c+1
f2.write(">"+i.id+"-part-"+str(c+1)+"\n"+str(t)+"\n")
else:
f2.write(">"+i.id+"\n"+str(i.seq)+"\n")
f2.close()
inputfile2=open('/OSM/HOME-MEL/all29c/scripts/tmp2/inp%s.fa' %(r2),'r')
count = SeqIO.index('/OSM/HOME-MEL/all29c/scripts/tmp2/inp%s.fa' %(r2), fmt)
c1= len(count)
if threads >c1:
chunksize = c1
threads = c1
else:
chunksize = int(c1/threads)+1
seqs=[]
print "threads=",threads
print "input file=", c1
print "Chunksize =", chunksize
for i in SeqIO.parse(inputfile2,fmt):
seqs.append(i)
print len(seqs)
if len(seqs)>chunksize:
seqlist=list(chunkstring(seqs,chunksize))
c=-1
for t in seqlist:
c=c+1
SeqIO.write(t,'/OSM/HOME-MEL/all29c/scripts/tmp2/%subin%d.fna' %(r2,c),fmt)
else:
SeqIO.write(seqs,'/OSM/HOME-MEL/all29c/scripts/tmp2/%subin0.fna' %(r2),fmt)
print "Files written"
print "Multithreading ublast"
#numt = number of times to run multithread
p1=[]
q=-1
for i in range(0,threads):
p1.append("")
for x in range(0,threads):
q=q+1 #file number
print""
p1[x] = subprocess.Popen("usearch -ublast ~/scripts/tmp2/%subin%d.fna -db %s -lopen 3 -lext 1 -strand both -maxaccepts %s -maxrejects 255 -maxhits %s -userout ~/scripts/tmp2/%subin%dout.txt -userfields query+target+id+ql+qlo+qhi+tl+tlo+thi+alnlen+evalue+bits -id %s -evalue %s" %(r2,q,db,maxacc,maxhits,r2,q,id,evalue), shell=True)
#wait for n threads to finish
for x1 in range(threads):
print "waiting for usearch to finish........."
p1[x1].wait()
for x2 in range(0,threads):
print "concatenating.....", x2
subprocess.Popen("cat ~/scripts/tmp2/%subin%dout.txt >> ~/scripts/tmp2/finaloutput%s.ublast" %(r2,x2,r2), shell=True).wait()
#subprocess.Popen("rm -rf ~/scripts/tmp2/ubin* ~/scripts/tmp2/1ubin* ~/scripts/tmp2/2ubin* ~/scripts/tmp2/3ubin* ~/scripts/tmp2/4ubin* ~/scripts/tmp2/5ubin* ", shell=True).wait()
print"adding taxonomy.."
subprocess.Popen("python ~/scripts/taxonfetch.py ~/scripts/tmp2/finaloutput%s.ublast %s p %s" %(r2,outputfile,fetchsize), shell=True).wait()
############otufreq2.py
#cons= consensus threshold, only take a species if its more than this frequent in the list of hits for a subject
f2 = open(outputfile,'r')
g2= open(outputfile+".species",'w')
# read output file species into list
species=[]
k=[]
print "Getting consensus species.."
def most_common(lst):
answer = str(max(set(lst), key=lst.count))
return answer
sp={}
for line in f2:
scaff=line.split("\t")[0]
if scaff not in sp.keys():
sp[scaff]=[]
k=line.rstrip("\n").split("\t")[-1].split(" ")
#get rid of brackets
if k[0][0]=="[":
k[0]=k[0][1:]
if k[0][-1]=="]":
k[0]=k[0][0:-1]
if len(k)>1:
tx = " ".join(str(x5) for x5 in k[0:2]) #nb makes two element list into a single string
sp[scaff].append(tx) #
if len(k)==1:
sp[scaff].append(k[0])
for i in sp.keys():
freq= float(sp[i].count(most_common(sp[i])))/float(len(sp[i]))
if freq> cons:
species.append(most_common(sp[i]))
species_set=set(species)
print "Counting species.."
for i in species_set:
g2.write(i+"\t"+str(species.count(i))+"\n")
subprocess.Popen("rm /OSM/HOME-MEL/all29c/scripts/tmp2/finaloutput%s.ublast" %(r2), shell=True).wait()
subprocess.Popen("rm /OSM/HOME-MEL/all29c/scripts/tmp2/inp%s.fa" %(r2), shell=True).wait()
subprocess.Popen("rm -r /OSM/HOME-MEL/all29c/scripts/tmp2/%subin*" %(r2), shell=True).wait()
f2.close()
g2.close()