#!/usr/bin/env python3
"""
optparse is deprecated in favor of argparse as of Python 2.7. However,
since 2.7 is not always present on many systems, at this writing,
it is safer to stick with optparse for now. It should be easy
to change later, since the syntax is very similar between argparse and optparse.
from optparse import OptionParser
"""
from optparse import OptionParser
import datetime
import getpass
import os
import re
import shutil
import stat
import subprocess
import sys
import time
'''
bl_fastq_pair.py - Set environment variables for BioLegato Helper Applications
Synopsis: bl_fastq_pair.py --tsv tsvfile [options]
@modified: April 5, 2019
@author: Brian Fristensky
@contact: Brian.Fristensky@umanitoba.ca
'''
#blib = os.environ.get("BIRCHPYLIB")
#sys.path.append(blib)
#from birchlib import Birchmod
PROGRAM = "bl_fastq_pair.py : "
USAGE = "\n\tUSAGE: bl_fastq_pair.py --tsv tsvfile [options]"
DEBUG = True
if DEBUG :
print('bl_fastq_pair: Debugging mode on')
#BM = Birchmod(PROGRAM, USAGE)
# - - - - - - - - - - - - - Utility classes - - - - - - - - - - - - - - - - -
def chmod_ar(filename):
"""
Make a file world-readable.
"""
if os.path.exists(filename):
st = os.stat(filename)
os.chmod(filename, st.st_mode | stat.S_IREAD \
| stat.S_IRGRP | stat.S_IROTH)
def chmod_arx(filename):
"""
Make a file or directory world-readable and world-executable/searchable.
"""
if os.path.exists(filename):
st = os.stat(filename)
os.chmod(filename, st.st_mode | stat.S_IEXEC | stat.S_IREAD \
| stat.S_IXGRP | stat.S_IRGRP | stat.S_IXOTH \
| stat.S_IROTH)
def LocalHostname():
"""
Return the name of the local machine. Tries a number of methods
to get a name other than 'localhost' or a null result.
"""
import socket
import platform
def CheckName(name) :
if name == None or name.startswith("localhost") or name == "" :
OKAY = False
else :
OKAY = True
return OKAY
name = os.getenv('HOSTNAME')
if not CheckName(name) :
name = platform.uname()[1]
if not CheckName(name) :
if socket.gethostname().find('.')>=0:
name=socket.gethostname()
else:
name=socket.gethostbyaddr(socket.gethostname())[0]
return name
def SendEmail(From,To,Subject,Text) :
"""
Very simple email method adapted from:
http://stackoverflow.com/questions/882712/sending-html-email-using-python
There are more elaborate examples on this site for sending
HTML messages and attachments.
"""
import smtplib
from email.mime.multipart import MIMEMultipart
from email.mime.text import MIMEText
Host = 'localhost'
msg = MIMEMultipart('alternative')
msg['Subject'] = Subject
Html = """\
%s
""" %(Text)
part1 = MIMEText(Text, 'plain')
part2 = MIMEText(Html, 'html')
msg.attach(part1)
msg.attach(part2)
try:
server = smtplib.SMTP(Host)
server.sendmail(From, To, msg.as_string())
server.quit()
print("Successfully sent email")
except :
print("Error: unable to send email")
# - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
class Parameters:
"""
Wrapper class for command line parameters
"""
def __init__(self):
"""
Initializes arguments:
TSVFILE = ""
HASHSIZE = 100000
BUCKETS = False
OUTDIR = ""
EMAIL = ""
Then calls read_args() to fill in their values from command line
"""
self.TSVFILE = "."
self.HASHSIZE = 100000
self.BUCKETS = False
self.OUTDIR = "."
self.EMAIL = ""
self.read_args()
if DEBUG :
print('------------ Parameters from command line ------')
print(' TSVFILE: ' + self.TSVFILE)
print(' HASHSIZE: ' + str(self.HASHSIZE))
print(' BUCKETS: ' + str(self.BUCKETS))
print(' OUTDIR: ' + self.OUTDIR)
print(' EMAIL: ' + self.EMAIL)
print()
def read_args(self):
"""
Read command line arguments into a Parameter object
"""
parser = OptionParser()
parser.add_option("--tsvfile", dest="tsvfile", action="store", default="",
help="list of paired/unpaired read files")
parser.add_option("--hashsize", dest="hashsize", action="store", default=100000,
help="trim N bases in addition to adapter removal")
parser.add_option("--buckets", dest="buckets", action="store_true", default=False,
help="Write number of sequences per bucket to output")
parser.add_option("--outdir", dest="outdir", action="store", default=".",
help="output directory")
parser.add_option("--email", dest="email", action="store", default="",
help="send email upon completion")
(options, args) = parser.parse_args()
self.TSVFILE = options.tsvfile
self.HASHSIZE = options.hashsize
self.BUCKETS = options.buckets
self.OUTDIR = options.outdir
self.EMAIL = options.email
# - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
class TSVFiles :
"""
Methods for reading lists of paired read TSV files, and for
writing lists to output.
"""
def __init__(self):
"""
Initializes arguments:
READPAIRS = []
"""
self.READPAIRS = []
def ReadTSVfile(self,FN) :
"""
TSV file containing names of paired-end and/or single end read files.
Paired-end files are on lines such as
leftreadfile.fq.gzrightreadfile.fq.gz
Single-end files have a each file on a separate line
reads1.fq.gz
reads2.fq.gz
reads3.fq.gz
"""
TAB = '\t'
F = open(FN,"r")
for line in F.readlines() :
line = line.strip()
if len(line) > 0 and not line.startswith('#') :
# get rid of double quotes that enclose fields when some programs write
# output, and then split by TABs.
tokens = line.replace('"','').split(TAB)
# ignore blank fields. Add either single or pair of filenames
# to list. Only process names from first two fields on a line
# and ignore other fields.
if len(tokens) > 0 :
r1 = tokens[0].strip()
if len(r1) > 0 :
fnames = [r1]
else :
fnames = []
if len(tokens) > 1 :
r2 = tokens[1].strip()
if len(r2) > 0 :
fnames.append(r2)
if len(fnames) > 0 :
self.READPAIRS.append(fnames)
if DEBUG :
print(str(self.READPAIRS))
F.close()
# - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
def RunFastq_Pair(P,PR,LOGFILE) :
def MoveOutfiles(FN,OUTDIR) :
PairedFile = FN + '.paired.fq'
Dest = os.path.join(OUTDIR,PairedFile)
shutil.move(PairedFile,Dest)
SingleFile = FN + '.single.fq'
Dest = os.path.join(OUTDIR,SingleFile)
shutil.move(SingleFile,Dest)
# Only process pairs of files
if len(PR) == 2 :
COMMAND=["fastq_pair"]
if P.BUCKETS :
COMMAND.append("-p")
COMMAND.extend(["-t", str(P.HASHSIZE)])
COMMAND.extend([PR[0],PR[1]])
# Run the command - - - - - - - - - - - - - - - - -
LOGFILE.write('======== fastq_pair ==========' + '\n')
LOGFILE.write(str(COMMAND) + '\n')
StartTime = datetime.datetime.now()
LOGFILE.write('Start time: ' + str(StartTime) + '\n')
LOGFILE.write('\n')
LOGFILE.flush()
p = subprocess.Popen(COMMAND,stdout=LOGFILE, stderr=LOGFILE)
p.wait()
FinishTime = datetime.datetime.now()
LOGFILE.write('\n')
LOGFILE.write('Finish time: ' + str(FinishTime) + '\n')
ElapsedTime = FinishTime - StartTime
LOGFILE.write('Elapsed time: ' + str(ElapsedTime) + '\n')
# Move output files to OUTDIR
MoveOutfiles(PR[0],P.OUTDIR)
MoveOutfiles(PR[1],P.OUTDIR)
else :
LOGFILE.write('Ignoring single file ' + PR[0] + '\n')
#======================== MAIN PROCEDURE ==========================
def main():
"""
Called when not in documentation mode.
"""
# Read parameters from command line
P = Parameters()
TF = TSVFiles()
if not P.TSVFILE == "" :
TF.ReadTSVfile(P.TSVFILE)
OKAY = True
else :
OKAY = False
if not os.path.exists(P.OUTDIR) :
os.mkdir(P.OUTDIR)
LOGFN = os.path.join(P.OUTDIR,'bl_fastq_pair.log')
# First create the file, then re-open it so that we can append to it.
# We first create with 'w' just to make sure that we're creating a fresh copy,
# rather than appending to an existing copy from a previous run.
LOGFILE = open(LOGFN,'w')
LOGFILE.close()
LOGFILE = open(LOGFN,'a')
LOGFILE.write('\n')
# Run fastq_pair
if OKAY :
for PR in TF.READPAIRS :
RunFastq_Pair(P,PR,LOGFILE)
LOGFILE.close()
# Notify user when job is done, if email address was
# supplied using --email
if P.EMAIL != "" :
Sender = getpass.getuser() + '@' + LocalHostname()
Subject = 'bl_fastq_pair.py completed'
Message = 'bl_fastq_pair.py: Completed
'
LOGFILE = open(os.path.join('bl_fastq_pair.log'),'r')
for line in LOGFILE.readlines() :
Message = Message + line + '
'
LOGFILE.close()
SendEmail(Sender,[P.EMAIL],Subject,Message)
if __name__ == "__main__":
main()
#else:
#used to generate documentation
# import doctest
# doctest.testmod()
#if (BM.documentor() or "-test" in sys.argv):
# pass
#else:
# main()