#!/usr/bin/env python3 # -*- coding: utf-8 -*- """ Created on Fri Apr 23 16:02:49 2021 @author: GIRAUD Delphine """ import sys # Automatic classification RIX # Input files: # argv[1] : *_chr_allTEs_nr_noSSR_join_path.annotStatsPerTE.tab Standard Output REPET TEannot Step 7 # argv[2] : *_Profiles.gff3 Standard Output REPET CreateGFF3sForClassifFeatures.py (script included in REPET package) # argv[3] : results_autoClassRLX.txt corresponded to standard output Additional file 2 # Output file: # result_autoClassRLX_autoClassRIX.txt corresponded to argv[1] file with additionnal informations for manual curation listOtherKnown = ["TREP", "OSLINE", "L1", "ATLINE", "profiles", "LINE", "Dl_Skipper", "ATGP", "TONT", "GYPOT", "TY3-", "TIRANT", "POPCOP", "M2_VM", "LTR-", "GYVIT", "CASTOR", "SZ", "SHACOP", "RLG", "RETROFIT", "POPGY", "GYRAV", "GYPSHAN", "COP", "ATHILA", "ATCOPIA", "TermRepeats", "GmCOPIA", "non-termLTR", "Gypsy", "Copia", "other", "termLTR", "termTIR", "SSR", "TElength", "TE_BLRx", "<", ">", "0", "1", "2", "3", "4", "5", "6", "7", "8", "9" ] listeLINEKnown = ["PF13456.6_RVT", "PF13966.6_zf-RVT", "PF00078.27_RVT", "PF14111.6_DUF4283", "PF14392.6_zf-CCHC", "PF03372.23_Exo", "PF14529.6_Exo"] dicoFull = {} file1 = open(sys.argv[1], "r") for line in file1: if not line.startswith("Seq_name"): name = line.split("\t")[0] length = int(line.split("\t")[1]) Wcode = line.split("\t")[11] listOther = [] dicoAN = {} if Wcode == "RIX": coding = line.split("\t")[13] if "profiles" in coding: coding = coding.split("profiles: ")[1] coding = coding.split(" ") for ele in coding : if ele.startswith("PF"): ele = ele.split("_")[0] + "_" + ele.split("_")[1] if ele not in listOtherKnown: dec = False for truc in listOtherKnown: if ele.startswith(truc): dec = True break if dec == False: listOther.append(ele) else: dec = False for truc in listOtherKnown: if ele.startswith(truc): dec = True break if dec == False: listOther.append(ele) other = line.split("\t")[15] if other != "other=(NA)": other = other.split(" ") for ele in other: dec = False for ele2 in listOtherKnown: if ele.startswith(ele2): dec = True break if dec == False: listOther.append(ele) struc = line.split("\t")[14] listStruc = [False, False, False] if "polyAtail" in struc: listStruc[0] = True if "SSR" in struc: if "start" in struc: listStruc[1] = True elif "end" in struc: listStruc[2] = True dicoFull[name] = [length, listOther, listStruc] file1.close() dico_clean = {} for FLF in dicoFull.keys(): listPos = [] for i in range(0, dicoFull[FLF][0]+1): listPos.append(0) listDomPF = dicoFull[FLF][1] RVT = False fileProfile = open(sys.argv[2], "r") for line in fileProfile: if not line.startswith("#"): cons = line.split("\t")[0] if cons == FLF: line = line.replace("\n", "") types = line.split("\t")[2] if types == "exon": evalue = line.split("\t")[5] evalue = evalue.replace(',', '.') evalue = float(evalue) if evalue < 0.01: start = int(line.split("\t")[3]) end = int(line.split("\t")[4]) des = line.split("\t")[8] des = des.split(";") for ele in des: if ele.startswith("Target="): des = ele break des = des.split(" ")[0] des = des.replace("Target=", "") if des.startswith("PF"): PF = des.split("_")[0] + "_" + des.split("_")[1] if PF in listeLINEKnown: for i in range(start, end+1): listPos[i] = PF else: if PF not in listDomPF: listDomPF.append(PF) else: if des.startswith("GyDB"): des = des.split("_")[1] +"_"+des.split("_")[2] if des not in listDomPF: listDomPF.append(des) else: print("Add dest\t" + des) fileProfile.close() pattern = "" start = 0 end = 0 for i in range(0, dicoFull[FLF][0]+1): curDom = listPos[i] if curDom != 0: curDom = curDom.upper() if not pattern.endswith(curDom): if pattern == "": start = i pattern = curDom else: pattern = pattern + "-" + curDom else: end = i pattern = str(start) + "-" + pattern + "-" + str(end) if pattern == "0--0": pattern = "" list_struc = dicoFull[FLF][2] outString = "\tRIX\tAUTO\t" if "RVT" in pattern : if True in list_struc: outString = outString + "keep\tOK: " + pattern else: outString = outString + "\t" + pattern else: outString = outString + "\t" + pattern listKnown = [] listNew = [] for dom in listDomPF: if dom in listeLINEKnown : listKnown.append(dom) else: listNew.append(dom) if listDomPF != []: if listKnown != [] and listNew != []: outString = outString + ";" + "NewDomaine:"+ str(listNew) + ";" + "KnownDomaine:"+ str(listKnown) elif listKnown == [] and listNew != []: outString = outString + ";" + "NewDomaine:"+ str(listNew) elif listKnown != [] and listNew == []: outString = outString + ";" + "KnownDomaine:"+ str(listKnown) dico_clean[FLF] = outString file3 = open(sys.argv[3], "r") file1_clean = open("result_autoClassRLX_autoClassRIX.txt", "w") file1_clean.write("Seq_name\tlength\tcovg\tclusMCL\tfrags\tfullLgthFrags\tcopies\tfullLgthCopies\tmeanId\tconfused\tWcode_old\tWcode\tnew-classif\tqui\tdecision\tcommentaire\tcovSSR%\tcoding\tstruc\tother\tresB_LTRStruc(nbMatch;bestHit;%id,len;eval)\tresB_SINE_Finder(nbMatch;bestHit;%id,len;eval)\n") for line in file1: if not line.startswith("Seq_name"): cons = line.split("\t")[0] if cons in dico_clean.keys(): list_start = line.split("\t")[0:12] list_end = line.split("\t")[16:] outString2 = "" for ele in list_start: if outString2 == "": outString2 = ele else: outString2 = outString2 + "\t" + ele outString2 = outString2 + dico_clean[cons] for ele in list_end: outString2 = outString2 + "\t" + ele file1_clean.write(outString2) else: file1_clean.write(line) file3.close() file1_clean.close()