#!/usr/bin/env python3 # -*- coding: utf-8 -*- """ Created on Fri Apr 23 16:02:49 2021 @author: GIRAUD Delphine """ import sys # Automatic classification RLC RLG RLX # 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) # Output file: # results_autoClassRLX.txt corresponded to argv[1] file with additionnal informations for manual curation listOtherKnown = ["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" ] 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] if Wcode == "RLG" or Wcode == "RLC" or Wcode == "RLG|RLC" or Wcode == "RLC|RLG" or Wcode == "RLX": dicoAN = {} listOther = [] coding = line.split("\t")[13] if "profiles" in coding: coding = coding.split("profiles: ")[1] coding = coding.split(" ") for ele in coding : if ele.startswith("GyDB"): ele = ele.split("_")[2] if ele not in dicoAN.keys(): dicoAN[ele] = [] elif ele.startswith("PF"): ele = ele.split("_")[0] + "_" + ele.split("_")[1] listOther.append(ele) other = line.split("\t")[15] if other != "other=(NA)": other = other.split(" ") for ele in other: if ele.startswith("PF"): listOther.append(ele.split("_")[0] + "_" + ele.split("_")[1]) else: dec = False for ele2 in listOtherKnown: if ele.startswith(ele2): dec = True break if dec == False: listOther.append(ele) dicoFull[name] = [length, Wcode, dicoAN, listOther] file1.close() listECV = ["PF00077.20_RVP", "PF01623.17_Carla", "PF08284.11_RVP", "PF06618.11_DUF1148", "PF01693.16_Cauli", "PF06143.11_Baculo"] MP = "PF01107.18_MP" Vsp51 = "PF08700.11_Vps51" RVP1 = "PF00077.20_RVP" RVP2 = "PF08284.11_RVP" listDomKnownET = listECV + ["PF15288.6_zf-CCHC", "PF13975.6_gag-asp","PF13966.6_zf-RVT","PF13917.6_zf-CCHC","PF13696.6_zf-CCHC","PF13683.6_rve","PF09668.10_Asp","PF00078.27_RVT", "PF00385.24_Chromo", "PF14244.6_Retrotran", "PF13961.6_DUF4219", "PF03078.15_ATHILA", "PF02892.15_zf-BED" ,"PF14223.6_Retrotran", "PF00665.26_rve", "PF07727.14_RVT", "PF17919.1_RT", "PF17921.1_Integrase", "PF16588.5_zf-C2H2", "PF00098.23_zf-CCHC", "PF16588.5_zf-C2H2", "PF13976.6_gag_pre-integrs", "PF13650.6_Asp", "PF13976.6_gag", "PPF03732.17_Retrotrans", "PF13456.6_RVT", "PF00075.24_RNase", "PF13456.6_RVT", "PF03732.17_Retrotrans", "PF04195.12_Transposase", "PF17917.1_RT"] dico_clean = {} for FLF in dicoFull.keys(): dicoAN_cur = dicoFull[FLF][2] for annot in dicoAN_cur.keys(): for i in range(0, dicoFull[FLF][0]+1): dicoAN_cur[annot].append(0) listDomPF = dicoFull[FLF][3] confu = [] file2 = open(sys.argv[2], "r") for line in file2: 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("GyDB"): ET = des.split("_")[2] Dom = des.split("_")[1] if ET in dicoAN_cur.keys(): for i in range(start, end+1): dicoAN_cur[ET][i] = Dom elif des.startswith("PF"): PF = des.split("_")[0] + "_" + des.split("_")[1] if PF == MP or PF == Vsp51 or PF == RVP1 or PF == RVP2: for i in range(start, end+1): if "caulimovirus" in dicoAN_cur.keys(): dicoAN_cur["caulimovirus"][i] = PF.split("_")[1] elif "badnavirus" in dicoAN_cur.keys(): dicoAN_cur["badnavirus"][i] = PF.split("_")[1] elif "cavemovirus" in dicoAN_cur.keys(): dicoAN_cur["cavemovirus"][i] = PF.split("_")[1] else: if PF not in listDomPF: listDomPF.append(PF) else: print("ERROR Target name\t") print(line) file2.close() dico_pattern = {} for fam in dicoAN_cur.keys(): pattern = "" start = 0 end = 0 for i in range(0, dicoFull[FLF][0]+1): curDom = dicoAN_cur[fam][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 dico_pattern[fam] = str(start) + "-" + pattern + "-" + str(end) new = False listKnown = [] listNew = [] for domPF in listDomPF: if domPF != MP and domPF != Vsp51 and domPF != RVP1 and domPF != RVP2: if domPF not in listDomKnownET : new = True listNew.append(domPF) else: listKnown.append(domPF) outString = "\t" if len(dico_pattern) == 1: for cle in dico_pattern.keys(): fam = cle cptTiret = dico_pattern[fam].count("-") Wcode = dicoFull[FLF][1] if fam == "caulimovirus" or fam == "badnavirus" or fam == "cavemovirus": outString = outString + "ERV_"+fam + "\t" if "MP-RVP-RT-RNASEH" in dico_pattern[fam] and cptTiret == 5: outString = outString + "AUTO\tkeep\t" if new == False: outString = outString + "OK: " + dico_pattern[fam] else: outString = outString + dico_pattern[fam] elif "MP-Vsp51-RVP-RT-RNASEH" in dico_pattern[fam] and cptTiret == 6: outString = outString + "AUTO\tkeep\t" if new == False: outString = outString + "OK: " + dico_pattern[fam] else: outString = outString + dico_pattern[fam] else: outString = outString + "AUTO\t\t" + dico_pattern[fam] else: outString = outString + Wcode + "_"+fam + "\t" if Wcode == "RLC": if "GAG-AP-INT-RT-RNASEH" in dico_pattern[fam] and cptTiret == 6: outString = outString + "AUTO\tkeep\t" if new == False: outString = outString + "OK: " + dico_pattern[fam] else: outString = outString + dico_pattern[fam] else: outString = outString + "AUTO\t\t" + dico_pattern[fam] elif Wcode == "RLG": if "GAG-AP-RT-RNASEH-INT" in dico_pattern[fam] and cptTiret == 6: outString = outString + "AUTO\tkeep\t" if new == False: outString = outString + "OK: " + dico_pattern[fam] else: outString = outString + dico_pattern[fam] else: outString = outString + "AUTO\t\t" + dico_pattern[fam] elif Wcode == "RLG|RLC" or Wcode == "RLC|RLG" or Wcode == "RLX": if "GAG-AP-INT-RT-RNASEH" in dico_pattern[fam] and cptTiret == 6: #donc RLC outString = outString + "AUTO\tkeep\t" if new == False: outString = outString + "OK: " + dico_pattern[fam] else: outString = outString + dico_pattern[fam] elif "GAG-AP-RT-RNASEH-INT" in dico_pattern[fam] and cptTiret == 6: #donc RLG outString = outString + "AUTO\tkeep\t" if new == False: outString = outString + "OK: " + dico_pattern[fam] else: outString = outString + dico_pattern[fam] else: outString = outString + "AUTO\t\t" + dico_pattern[fam] elif len(dico_pattern) == 0: outString = outString + "\tAUTO\t\t" else: outString = outString + "\tAUTO\t\tchimera "+str(len(dico_pattern))+" element" for fam in dico_pattern.keys(): outString = outString + " " + fam + ":" + dico_pattern[fam] 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 file1 = open(sys.argv[1], "r") file1_clean = open("results_autoClassRLX.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] list_start = line.split("\t")[0:12] list_end = line.split("\t")[12:] outString2 = "" for ele in list_start: if outString2 == "": outString2 = ele else: outString2 = outString2 + "\t" + ele if cons in dico_clean.keys(): outString2 = outString2 + dico_clean[cons] else: outString2 = outString2 + "\t\t\t\t" for ele in list_end: outString2 = outString2 + "\t" + ele file1_clean.write(outString2) file1.close() file1_clean.close()