#!/usr/bin/python

#-*-coding : utf-8-*-

#Copyright(c) 2014 - SunLiang <sunliang@bioinfo.ac.cn>

from optparse import OptionParser

import sys

import math

import time

import os

from Gtf import Gtf

from Table import Table

from sets import Set

def make_dir(filepath):

if not os.path.exists(filepath):

os.makedirs(filepath)

def index_transform(row_data):

return result

def check_lncRNA(data):

return False

def sub_array(A,B):

x=Set(A)

y=Set(B)

return list(x - y)

def intersect_array(A,B):

x=Set(A)

y=Set(B)

return list(x & y)

def union_array(A,B):

x=Set(A)

y=Set(B)

return list(x | y)

def de_redundency(A):

return list(Set(A))

def classify(C,lnc,AM,discard):

return result

usage="""

"""

parser=OptionParser(usage=usage)

parser.add_option('-i','--index',dest='index',action='store',help='(Required.) '

+'The path of coding/noncoding index file. '

+'This file is the output file of CNCI.py, named as CNCI.index under CNCI.py outdir')

parser.add_option('-g','--gtf',dest='gtf',action='store',help='(Required.) '

+'The path of potentially novel lincRNAs gtf file. '

+'This file is generated by compare.py, named as potentially_novel.gtf under compare.py outdir')

parser.add_option('-s','--score',dest='score',action='store',default=0,help='(Optional.) '

+'Threoshold of CNCI score. RNAs with score less than SCORE will be '

+'classified as noncoding. The Default is 0 .')

parser.add_option('-l','--length',dest='length',action='store',default=200,help='(Optional.) '

+'Minimal length of lincRNA. lincRNA with length >= LENGTH will be kept. '

+'The Default is 200. ')

parser.add_option('-e','--exon_num',dest='exon_num',action='store',default=2,help='(Optional.) '

+'Minimal exon number of lincRNA. lincRNA with exon number >= EXON_NUM '

+'will be kept. The Default is 2.')

parser.add_option('-o','--out_dir',dest='out_dir',action='store',help='(Requried.) Output directory of the results.')

(options,args) = parser.parse_args()

if options.index is not None:

index_file=options.index

else:

print parser.print_help()

exit("Error: Coding/noncoding index file is required!")

if options.gtf is not None:

gtf_input=options.gtf

else:

print parser.print_help()

exit("Error: Unannotated gtf file is required!")

if options.out_dir is not None:

out_dir=options.out_dir

out_dir=out_dir.rstrip('/')

else:

print parser.print_help()

exit("Error: Output directory is required!")

TH_score = float(options.score)

TH_length = int(options.length)

exon_num=int(options.exon_num)

start_time = time.time()

print "Run start:"

#######################################

make_dir(out_dir)

#gtf_outdir=out_dir+'/'+'gtf'

#tid_gid_file=gtf_outdir+"/"+"tid_gid"

#make_dir(gtf_outdir)

snc_gtf=out_dir+"/"+'filter_out_noncoding.gtf'

lnc_gtf=out_dir+"/"+'novel_lincRNA.gtf'

coding_gtf=out_dir+"/"+'novel_coding.gtf'

cnc_gtf=out_dir+"/"+'ambiguous_genes.gtf'

Gene_Info = out_dir + '/compare_2_infor.txt'

###################################

gtf=Gtf(gtf_input)

tid_gid=gtf.get_tid_gid()

#index_content=Table(index_file,0,False)

#index_content.write_to_file('index_content')

#transform_index_content=index_content.cal(index_transform,'row')

#transform_index=Table.build_table(1,['tid','cnc_type','score','start','end','length'],transform_index_content)

transform_index=Table(index_file,1,True)

transform_index.left_join(1,tid_gid,1)

gid_cnctype=transform_index.get_col('gid','index')

gid_cnctype=gid_cnctype.de_redundency()

gid_cnctype=gid_cnctype.key_by('gid','index')

coding=gid_cnctype.eget('union','_index=coding')

noncoding=gid_cnctype.eget('union','_index=noncoding')

ambiguity=gid_cnctype.eget('union','_index=coding,noncoding','_index=noncoding,coding')

exon_number=gtf.getExon()

transform_index.left_join(1,exon_number,1)

transform_index=transform_index.get_row_by_func(check_lncRNA)

all_gids=de_redundency(tid_gid.getCol('gid'))

C=coding.getCol('gid')

print "Novel coding genes:",len(C)

N=noncoding.getCol('gid')

AM=ambiguity.getCol('gid')

print "Ambiguous genes:",len(AM)

lnc=de_redundency(transform_index.getCol('gid'))

lnc=intersect_array(N,lnc)

print "Novel lincRNA genes:",len(lnc)

discard=sub_array(all_gids,C)

discard=sub_array(discard,AM)

discard=sub_array(discard,lnc)

print "Filter out noncoding genes:",len(discard)

gene_class=classify(C,lnc,AM,discard)

gene_class.set_colnames('gene_id','class')

gene_class.write_to_file(Gene_Info)

gtf.sub_gtf(gtf.get_tid(discard)).write_to_file(snc_gtf)

gtf.sub_gtf(gtf.get_tid(lnc)).write_to_file(lnc_gtf)

gtf.sub_gtf(gtf.get_tid(C)).write_to_file(coding_gtf)

gtf.sub_gtf(gtf.get_tid(AM)).write_to_file(cnc_gtf)

run_time=int(time.time() - start_time)

exit("Run complete: "+"%d seconds elapsed " % run_time)