#!/usr/bin/env python2.7

from Bio import pairwise2, SeqIO

from Bio.Seq import reverse_complement

from Bio.pairwise2 import format_alignment

import argparse

import numpy as np

parser = argparse.ArgumentParser()

parser.add_argument('spacers', help='spacers file [.fa]')

parser.add_argument('reads', help='comma separeted reads files [1.fastq,2.fastq]') #

parser.add_argument('names', help='Names file of final result')

parser.add_argument('seq', help='.seq file of 1st step')

parser.add_argument('out', help='out name file')

parser.add_argument('-r', type=int, default=3, help="mismatch [Default: 3]")

args = parser.parse_args()

spacersDic = dict() # dict of spacers and seq (from fasta)

readsFilesList = []

readsDict = dict() # dict of reads 1 or 2 (fastq)

#readsDict1 = dict() # dict of reads temp (fastq)

readQualDic = dict() # dict of spacersName:readsQual list

readNameDic = dict() # dict of spacerName:readsName list

spacerLen = []

namesDict = dict() # Dict of names from names file {name1: [name1,name2,name3,...]}

seqDict = dict() # Dict {spacerOcu:readName}

spacerReadsDic = dict() # merge of namesDict and seqDict {spacer:[R1,R2,R3,..]}

empList = []

## Dict of results

aveQBaseDic = dict()

stdQBaseDic = dict()

minQBaseDic = dict()

maxQBaseDic = dict()

aveQPositionDic = dict()

count = 0

nCount = -1

nSumCount = -1

## Make list of reads files:

readsFilesList = args.reads.split(",")

print readsFilesList

## Open and Parse spacers_file

with open(args.spacers,'rt') as spacers_file:

for n in SeqIO.parse(spacers_file, "fasta"):

print(n.id),(n.seq),len(n.seq)

spacersDic[n.id] = str(n.seq)

readQualDic[n.id] = list()

readNameDic[n.id] = list()

spacerLen.append(len(n.seq))

while len(empList) <= max(spacerLen):

empList.append(0)

## Open and parse names file:

with open(args.names,'rt') as names_file:

for name in names_file.readlines():

nameSpl = name[:-1].split("\t")

namesDict[nameSpl[0]] = nameSpl[1].split(",")

## Open and parse seq file:

with open(args.seq,'rt') as seq_file:

for seq in seq_file.readlines()[1:]:

seqSpl = seq[:-1].split("\t")

seqDict[seqSpl[0]] = seqSpl[2]

## Merge of namesDict and seqDict:

for sp in namesDict:

spacerReadsDic[sp] = list()

for spR in namesDict[sp]:

spacerReadsDic[sp].append(seqDict[spR]) # Dict of spacer:R1,R2,R3,..

namesDict = dict()

seqDict = dict()

for readsin in readsFilesList:

## Open and parse fastq files:

with open(readsin,'rt') as reads_file1:

readsDict = SeqIO.to_dict(SeqIO.parse(reads_file1,"fastq"))

print(len(readsDict))

for s in spacersDic:

for rd in spacerReadsDic[s]:

seqRead = str(readsDict.get(rd,"None"))

if seqRead != "None":

seqSpacer = spacersDic[s]

seqRead = str(readsDict[rd].seq)

alignment = pairwise2.align.localms(seqRead,seqSpacer,2,-.1,-3,-2, one_alignment_only=True)

if alignment[0][2] >= (len(seqSpacer)*2)-(args.r*2.1):

print(alignment[0]), len(seqSpacer)*2

print(format_alignment(*alignment[0]))

readNameDic[s].append(readsDict[rd].id)

readQualDic[s].append(readsDict[rd].letter_annotations["phred_quality"][alignment[0][3]:alignment[0][4]])

## Reverse aligment read 1:

alignmentR = pairwise2.align.localms(reverse_complement(seqRead),seqSpacer,2,-.1,-3,-2, one_alignment_only=True)

if alignmentR[0][2] >= (len(seqSpacer)*2)-(args.r*2.1):

print(alignmentR[0]), len(seqSpacer)*2

print(format_alignment(*alignmentR[0]))

readNameDic[s].append(readsDict[rd].id)

rR = readsDict[rd].letter_annotations["phred_quality"]

rR.reverse()

readQualDic[s].append(rR[alignmentR[0][3]:alignmentR[0][4]])

readsDict = dict() ## clean memory

# Generate out file:

with open(args.out+".resultQ.file.test.csv",'wt') as spacers_out_file:

spacers_out_file.write("id\tNR\tNR100%\tAvQ100%\tstdQ100%\tAveQ100%List\n")

with open(args.out+".resultQ.file.test.full_report.txt",'wt') as spacers_out_file2:

spacers_out_file2.write(args.out+"_full_report\n")

with open(args.out+".resultQ.file.test.short_report.txt",'wt') as spacers_out_file3:

spacers_out_file3.write(args.out+"_short_report\n")

for qval in readQualDic:

if readQualDic[qval] != []:

aveQBaseDic[qval] = list()

stdQBaseDic[qval] = list()

minQBaseDic[qval] = list()

maxQBaseDic[qval] = list()

aveQPositionDic[qval] = list()[0:len(spacersDic[qval])]=empList[0:len(spacersDic[qval])]

for qqval in readQualDic[qval]:

if len(qqval) == len(spacersDic[qval]):

count += 1

print "\n",qqval, "count:",count

print "max Q base:",max(qqval), "| min Q base:", min(qqval), "| Average Q base:", round(float(sum(qqval))/len(qqval),2)

with open(args.out+".resultQ.file.test.full_report.txt",'at') as spacers_out_file2:

spacers_out_file2.write(str(qval)+"| ")

spacers_out_file2.write("max Q base:" + str(max(qqval)) + "| min Q base:" + str(min(qqval)) + "| Average Q base: " + str(round(float(sum(qqval))/len(qqval),2)) + "\n")

spacers_out_file2.write(str(qqval)+" count: "+str(count)+"\n")

aveQBaseDic[qval].append(round(float(sum(qqval))/len(qqval),2))

stdQBaseDic[qval].append(round(np.std(qqval),2))

minQBaseDic[qval].append(min(qqval))

maxQBaseDic[qval].append(max(qqval))

for n in qqval:

nCount +=1

aveQPositionDic[qval][nCount] += n

nCount = -1

if count != 0:

for nSum in aveQPositionDic[qval]:

nSumCount += 1

aveQPositionDic[qval][nSumCount] = round(float(nSum)/count,1)

nSumCount = -1

print(qval), "| Nro de reads100%len:", count,\

"| Nro de reads:",len(readQualDic[qval]),\

"| Q media spacer100%len:",round(float(sum(aveQBaseDic[qval]))/len(aveQBaseDic[qval]),2),\

"| Q std spacer100%len:", round(np.std(aveQPositionDic[qval]),2)

print qval,"Q average per base:",aveQPositionDic[qval]

with open(args.out+".resultQ.file.test.short_report.txt",'at') as spacers_out_file3:

spacers_out_file3.write(str(qval)+" | Q average per base:"+str(aveQPositionDic[qval])+"\n")

## Output files to visualize on gephi

with open(args.out+".resultQ.file.test.csv",'at') as spacers_out_file:

spacers_out_file.write(qval+"\t"+str(len(readQualDic[qval]))+"\t"+str(count)+"\t"+str(round(float(sum(aveQBaseDic[qval]))/len(aveQBaseDic[qval]),2))+"\t"+str(round(np.std(aveQPositionDic[qval]),2))+"\t"+str(aveQPositionDic[qval])+"\n")

count = 0