# 
# Author: Praveen Kumar
# Updated: Nov 8th, 2017
# 
# 
# 

import re


def main():
    import sys
    if len(sys.argv) == 4:
        inputFile = sys.argv
        infh = open(inputFile[1], "r")
        # infh = open("Mus_musculus.GRCm38.90.chr.gtf", "r")
        
        gtf = {}
        gtf_transcript = {}
        gtf_gene = {}
        for each in infh.readlines():
            a = each.split("\t")
            if re.search("^[^#]", each):
                if re.search("gene_biotype \"protein_coding\"", a[8]) and int(a[4].strip()) != int(a[3].strip()):
                    type = a[2].strip()
                    if type == "gene" or type == "exon" or type == "CDS" or type == "five_prime_utr" or type == "three_prime_utr":
                        chr = "chr" + a[0].strip()
                        strand = a[6].strip()
                        if strand == "+":
                            start = a[3].strip()
                            end = a[4].strip()
                        elif strand == "-":
                            if int(a[4].strip()) > int(a[3].strip()):
                                start = a[3].strip()
                                end = a[4].strip()
                            elif int(a[4].strip()) < int(a[3].strip()):
                                start = a[4].strip()
                                end = a[3].strip()
                            else:
                                print "Something fishy in start end coordinates"
                        else:
                            print "Something fishy in reading"
                        if not gtf.has_key(strand):
                            gtf[strand] = {}
                        if not gtf[strand].has_key(type):
                            gtf[strand][type] = []
                        b = re.search("gene_id \"(.+?)\";", a[8].strip())
                        gene = b.group(1)
                        if type == "gene":
                            transcript = ""
                        else:
                            b = re.search("transcript_id \"(.+?)\";", a[8].strip())
                            transcript = b.group(1)
                        data = (chr, start, end, gene, transcript, strand, type)
                        gtf[strand][type].append(data)
                
                        if type == "exon":
                            if gtf_transcript.has_key(chr+"#"+strand):
                                if gtf_transcript[chr+"#"+strand].has_key(transcript+"#"+gene):
                                    gtf_transcript[chr+"#"+strand][transcript+"#"+gene][0].append(int(start))
                                    gtf_transcript[chr+"#"+strand][transcript+"#"+gene][1].append(int(end))
                                else:
                                    gtf_transcript[chr+"#"+strand][transcript+"#"+gene] = [[],[]]
                                    gtf_transcript[chr+"#"+strand][transcript+"#"+gene][0].append(int(start))
                                    gtf_transcript[chr+"#"+strand][transcript+"#"+gene][1].append(int(end))
                            else:
                                gtf_transcript[chr+"#"+strand] = {}
                                gtf_transcript[chr+"#"+strand][transcript+"#"+gene] = [[],[]]
                                gtf_transcript[chr+"#"+strand][transcript+"#"+gene][0].append(int(start))
                                gtf_transcript[chr+"#"+strand][transcript+"#"+gene][1].append(int(end))
                
                        if type == "gene":
                            if gtf_gene.has_key(chr+"#"+strand):
                                gtf_gene[chr+"#"+strand][0].append(int(start))
                                gtf_gene[chr+"#"+strand][1].append(int(end))
                                gtf_gene[chr+"#"+strand][2].append(gene)
                            else:
                                gtf_gene[chr+"#"+strand] = [[0],[0],["no_gene"]]
                                gtf_gene[chr+"#"+strand][0].append(int(start))
                                gtf_gene[chr+"#"+strand][1].append(int(end))
                                gtf_gene[chr+"#"+strand][2].append(gene)
                        
                        
    
        # "Starting Reading Intron . . ."
    
        gtf["+"]["intron"] = []
        gtf["-"]["intron"] = []
        for chr_strand  in gtf_transcript.keys():
            chr = chr_strand.split("#")[0]
            strand = chr_strand.split("#")[1]
        
            for transcript_gene in gtf_transcript[chr_strand].keys():
                start_list = gtf_transcript[chr_strand][transcript_gene][0]
                end_list = gtf_transcript[chr_strand][transcript_gene][1]
                sorted_start_index = [i[0] for i in sorted(enumerate(start_list), key=lambda x:x[1])]
                sorted_end_index = [i[0] for i in sorted(enumerate(end_list), key=lambda x:x[1])]
                if sorted_start_index == sorted_end_index:
                    sorted_start = sorted(start_list)
                    sorted_end = [end_list[i] for i in sorted_start_index]
                    for x in range(len(sorted_start))[1:]:
                        intron_start = sorted_end[x-1]+1
                        intron_end = sorted_start[x]-1
                        transcript = transcript_gene.split("#")[0]
                        gene = transcript_gene.split("#")[1]
                        data = (chr, str(intron_start), str(intron_end), gene, transcript, strand, "intron")
                        gtf[strand]["intron"].append(data)
                    
    
        # "Starting Reading Intergenic . . ."
    
        gtf["+"]["intergenic"] = []
        gtf["-"]["intergenic"] = []
        for chr_strand  in gtf_gene.keys():
            chr = chr_strand.split("#")[0]
            strand = chr_strand.split("#")[1]
            start_list = gtf_gene[chr_strand][0]
            end_list = gtf_gene[chr_strand][1]
            gene_list = gtf_gene[chr_strand][2]
            sorted_start_index = [i[0] for i in sorted(enumerate(start_list), key=lambda x:x[1])]
            sorted_end_index = [i[0] for i in sorted(enumerate(end_list), key=lambda x:x[1])]
        
            sorted_start = sorted(start_list)
            sorted_end = [end_list[i] for i in sorted_start_index]
            sorted_gene = [gene_list[i] for i in sorted_start_index]
            for x in range(len(sorted_start))[1:]:
                intergene_start = sorted_end[x-1]+1
                intergene_end = sorted_start[x]-1
                if intergene_start < intergene_end:
                    intergene_1 = sorted_gene[x-1]
                    intergene_2 = sorted_gene[x]
                    gene = intergene_1 + "-#-" + intergene_2
                    data = (chr, str(intergene_start), str(intergene_end), gene, "", strand, "intergenic")
                    gtf[strand]["intergenic"].append(data)
    
        import sqlite3
        # conn = sqlite3.connect('gtf_database.db')
        conn = sqlite3.connect(":memory:")
        c = conn.cursor()
        # c.execute("DROP TABLE IF EXISTS gtf_data;")
        # c.execute("CREATE TABLE IF NOT EXISTS gtf_data(chr text, start int, end int, gene text, transcript text, strand text, type text)")
        c.execute("CREATE TABLE gtf_data(chr text, start int, end int, gene text, transcript text, strand text, type text)")
    
        for strand in gtf.keys():
            if strand == "+":
                st = "positive"
            elif strand == "-":
                st = "negative"
            else:
                print "Something fishy in writing . . ."
        
            for type in gtf[strand].keys():
                data = gtf[strand][type]
                c.executemany('INSERT INTO gtf_data VALUES (?,?,?,?,?,?,?)', data)
            
        conn.commit()
    
        infh = open(inputFile[2], "r")
        # infh = open("Mouse_Data_All_peptides_withNewDBs.txt", "r")
        data = infh.readlines()
        # output file
        outfh = open(inputFile[3], 'w')
        # outfh = open("classified_1_Mouse_Data_All_peptides_withNewDBs.txt", "w")
    
        for each in data:
            a = each.split("\t")
            chr = a[0].strip()
            pep_start = a[1].strip()
            pep_end = a[2].strip()
            strand = a[5].strip()
            c.execute("select * from gtf_data where type = 'CDS' and chr = '"+chr+"' and start <= "+pep_start+" and end >= "+pep_end+" and strand = '"+strand+"' ")
            rows = c.fetchall()
            if len(rows) > 0:
                outfh.write(each.strip() + "\tCDS\n")
            else:
                c.execute("select * from gtf_data where type = 'five_prime_utr' and chr = '"+chr+"' and start <= "+pep_start+" and end >= "+pep_end+" and strand = '"+strand+"' ")
                rows = c.fetchall()
                if len(rows) > 0:
                    outfh.write(each.strip() + "\tfive_prime_utr\n")
                else:
                    c.execute("select * from gtf_data where type = 'three_prime_utr' and chr = '"+chr+"' and start <= "+pep_start+" and end >= "+pep_end+" and strand = '"+strand+"' ")
                    rows = c.fetchall()
                    if len(rows) > 0:
                        outfh.write(each.strip() + "\tthree_prime_utr\n")
                    else:
                        c.execute("select * from gtf_data where type = 'exon' and chr = '"+chr+"' and start <= "+pep_start+" and end >= "+pep_end+" and strand = '"+strand+"' ")
                        rows = c.fetchall()
                        if len(rows) > 0:
                            outfh.write(each.strip() + "\texon\n")
                        else:
                            c.execute("select * from gtf_data where type = 'intron' and chr = '"+chr+"' and start <= "+pep_start+" and end >= "+pep_end+" and strand = '"+strand+"' ")
                            rows = c.fetchall()
                            if len(rows) > 0:
                                outfh.write(each.strip() + "\tintron\n")
                            else:
                                c.execute("select * from gtf_data where type = 'gene' and chr = '"+chr+"' and start <= "+pep_start+" and end >= "+pep_end+" and strand = '"+strand+"' ")
                                rows = c.fetchall()
                                if len(rows) > 0:
                                    outfh.write(each.strip() + "\tgene\n")
                                else:
                                    c.execute("select * from gtf_data where type = 'intergenic' and chr = '"+chr+"' and start <= "+pep_start+" and end >= "+pep_end+" and strand = '"+strand+"' ")
                                    rows = c.fetchall()
                                    if len(rows) > 0:
                                        outfh.write(each.strip() + "\tintergene\n")
                                    else:
                                        outfh.write(each.strip() + "\tOVERLAPPING_ON_TWO_REGIONS: PLEASE_LOOK_MANUALLY (Will be updated in next version)\n")
    
        conn.close()
        outfh.close()
    else:
        print "USAGE: python pep_pointer.py <input GTF file> <input tblastn file> <name of output file>"
    return None

if __name__ == "__main__":
    main()