#Author, date: Xyrus (last modified by him Sept 17 2020), most recently updated by Auden on October 17 2024 #Motivation: Generate lots of codon usage statistics to aid in identifying useful characteristics for de novo ORF calling #Intent: Summarize nucleotide composition statistics for a fasta file or folder of fasta files #Dependencies: Python3, numpy, BioPython #Inputs: Fasta file or folder of fasta files #Outputs: A fasta file filtered for properly formatted sequences and several spreadsheets summarizing GC, ENc, RSCU, etc. #Example: python3 CUB.py -i seqs.fasta #Note: Use "python3 CUB.py -i seqs.fasta --require_start --require_stop" when using the on R2G files. #Dependencies import os import re import sys import numpy as np from Bio import SeqIO from Bio.Seq import Seq from Bio.SeqUtils import GC import argparse class CalcCUB: """ Returns the Effective Number of Codons used (observed and expected) following the equations originally from Wright 1990. """ def expWrightENc(gc3): # Calculates the expected ENc from a sequence's GC3 under Wright 1990 if gc3 > 1: # If GC3 looks as though it is > 1 (e.g. 100%), converts to a float ≤ 1. # Calculations expect a value between 0 and 1 gc3 = gc3/100 exp_enc = 2+gc3+(29/((gc3**2)+(1-gc3)**2)) return round(exp_enc, 4) def nullENcGC3(): # Calculates the expected ENc from the null distribution of GC3 # values (0, 100% GC) null = [CalcCUB.expWrightENc(n) for n in np.arange(0,.51,0.01)] null += null[:-1][::-1] return [str(i)+'\t'+str(j) for i, j in zip([n for n in range(0, 101)],null)] def calcWrightENc(cdnTable): # Follows Wright's (1990) calculations for determining ENc scores. def faCalcWright(aa_counts): # Returns the codon homozygosity (fa) for a given "type" of AA (e.g. # 2-fold degeneracy). counts = [i[2] for i in aa_counts] # n_aa --> number of this particular AA n_aa = sum(counts) # fa --> codon homozygosity try: fa = (((n_aa*sum([(i/float(n_aa))**2 for i in counts]))-1)/(n_aa-1)) except: fa = 0 return fa def ENcWright_by_Degen(fa_data): # Same as used in Wright 1990, averages the homozygosity across all codons # of a given class (e.g. 2-fold degeneracy) # Codons without any degeneracy (e.g. ATG == M) have 100% homozygosity # and provide a "base" for the ENc score enc = 2 for k, v in fa_data.items(): non_zero_vals, non_zero_sum = len([i for i in v if i != 0]), sum([i for i in v if i != 0]) try: f_aa = non_zero_sum/non_zero_vals except: f_aa = 1 enc += k/f_aa return enc # Determines the number of degenerate groups to use (i.e. whether 6-Fold # degeneracy is present). degen_cdns = {} for k, v in cdnTable.items(): if v[1] not in degen_cdns.keys(): degen_cdns[v[1]] = [v[0]] else: if v[0] not in degen_cdns[v[1]]: degen_cdns[v[1]] += [v[0]] # Calculates codon homozygosity (fa) for each amino acid. Groups the # resulting values based on the amino acids degeneracy (e.g. 'two-fold'). fa_cdns = {len(v):[] for k, v in degen_cdns.items() if 'one' not in k} for k, v in degen_cdns.items(): # Skip codons lacking degeneracy if 'one' in k: continue for aa in v: aa_counts = [cdnTable[k] for k in cdnTable.keys() if cdnTable[k][0] == aa] fa_cdns[len(v)] += [faCalcWright(aa_counts)] enc_val = min(61, round(ENcWright_by_Degen(fa_cdns),4)) return enc_val def SunEq5(cdnTable): def calcFcf(aa_counts): counts = [i[2] for i in aa_counts] pseudocounts = [i+1 for i in counts] na = sum(pseudocounts) fcf = sum([(i/float(na))**2 for i in pseudocounts]), sum(pseudocounts) return fcf ENcWeightedPsuedo = 0 degen_cdns = {} for k, v in cdnTable.items(): if v[1] == 'none': continue if v[1] not in degen_cdns.keys(): degen_cdns[v[1]] = [v[0]] else: if v[0] not in degen_cdns[v[1]]: degen_cdns[v[1]] += [v[0]] for k, v in degen_cdns.items(): fcf_nc = [] for aa in v: aa_counts = [cdnTable[k] for k in cdnTable.keys() if cdnTable[k][0] == aa] fcf_nc.append(calcFcf(aa_counts)) weightedENc = (len(fcf_nc) / (sum([i[0]*i[1] for i in fcf_nc]) / sum([i[1] for i in fcf_nc]))) ENcWeightedPsuedo += weightedENc return round(ENcWeightedPsuedo,4) def calcRCSU(cdnTbl): rscu = {k:[v[0]] for k, v in cdnTbl.items() if v[0].isalpha()} for k, v in rscu.items(): try: aa_info = [(key, val[-1]) for key, val in cdnTbl.items() if val[0] == v[0]] aa_cnts = [x[1] for x in aa_info] cdn_rscu = (cdnTbl[k][-1]*len(aa_cnts))/sum(aa_cnts) rscu[k] += [str(round(cdn_rscu,4))] except: rscu[k] += ['NA'] return rscu class GenUtil(object): """ "Overflow" of functions for now. Just a precaution to make the code a little cleaner/easier to manage. This class inclues means to normalize/check the user-provided genetic code, which if not valid will default to the "universal" genetic code. Similarly, This class will return the appropriate codon count table and provides a function to update its values. """ def convertGenCode(gCode): # Will interpret the user provided genetic code (gcode) and checks that # it is currently available for use with the NCBI/biopython # supported translation tables. Default is universal. # Dictionary of the possible/functional genetic codes that are supported. # --- Chilodonella and condylostoma are to come! transTable = {'universal':1, 'blepharisma':4, 'ciliate':6, 'euplotes':10, 'mesodinium':29, 'myrionecta':29, 'peritrich':30, '1':1, '4':4, '6':6, '10':10, '29':29, '30':30, 'chilo':'chilo'} if str(gCode).lower() not in transTable: print("\nWarning: Provided genetic code is not supported (yet).\n") print("Currently running using the UNIVERSAL genetic code.\n\n") print("Alternative genetic codes are as follows (Note: numbers "\ "correspond to NCBI genetic code tables):\n") print('\n'.join(list(transTable.keys()))+'\n') return 'Universal',1 else: return gCode,transTable[str(gCode).lower()] def getCDNtable(gCode): # Returns the appropriate codon table to be used for the ENc calculations. # Universal codon table, with 6-fold degenerate codons split # into four-fold and two-fold groups. universal_no6fold = { 'GCT': ['A', 'four', 0], 'GCC': ['A', 'four', 0], 'GCA': ['A', 'four', 0], 'GCG': ['A', 'four', 0], 'CGT': ['R', 'four', 0], 'CGC': ['R', 'four', 0], 'CGG': ['R', 'four', 0], 'CGA': ['R', 'four', 0], 'AGA': ['R_', 'two', 0], 'AGG': ['R_', 'two', 0], 'AAT': ['N', 'two', 0], 'AAC': ['N', 'two', 0], 'GAT': ['D', 'two', 0], 'GAC': ['D', 'two', 0], 'TGT': ['C', 'two', 0], 'TGC': ['C', 'two', 0], 'CAA': ['Q', 'two', 0], 'CAG': ['Q', 'two', 0], 'GAA': ['E', 'two', 0], 'GAG': ['E', 'two', 0], 'GGT': ['G', 'four', 0], 'GGC': ['G', 'four', 0], 'GGA': ['G', 'four', 0], 'GGG': ['G', 'four', 0], 'CAT': ['H', 'two', 0], 'CAC': ['H', 'two', 0], 'ATT': ['I', 'three', 0], 'ATC': ['I', 'three', 0], 'ATA': ['I', 'three', 0], 'ATG': ['M', 'one', 0], 'TTA': ['L_', 'two', 0], 'TTG': ['L_', 'two', 0], 'CTT': ['L', 'four', 0], 'CTC': ['L', 'four', 0], 'CTA': ['L', 'four', 0], 'CTG': ['L', 'four', 0], 'AAA': ['K', 'two', 0], 'AAG': ['K', 'two', 0], 'TTT': ['F', 'two', 0], 'TTC': ['F', 'two', 0], 'CCT': ['P', 'four', 0], 'CCC': ['P', 'four', 0], 'CCA': ['P', 'four', 0], 'CCG': ['P', 'four', 0], 'TCT': ['S', 'four', 0], 'TCC': ['S', 'four', 0], 'TCA': ['S', 'four', 0], 'TCG': ['S', 'four', 0], 'AGT': ['S_', 'two', 0], 'AGC': ['S_', 'two', 0], 'ACT': ['T', 'four', 0], 'ACC': ['T', 'four', 0], 'ACA': ['T', 'four', 0], 'ACG': ['T', 'four', 0], 'TGG': ['W', 'one', 0], 'TAT': ['Y', 'two', 0], 'TAC': ['Y', 'two', 0], 'GTT': ['V', 'four', 0], 'GTC': ['V', 'four', 0], 'GTA': ['V', 'four', 0], 'GTG': ['V', 'four', 0], 'TAA': ['*', 'none', 0], 'TGA': ['*', 'none', 0], 'TAG': ['*', 'none', 0], 'XXX': ['_missing', 'none', 0]} # Universal codon table, with 6-fold degenerate codons kept # whole, no splitting! Traditional Universal codon table. universal_6fold = { 'GCT': ['A', 'four', 0], 'GCC': ['A', 'four', 0], 'GCA': ['A', 'four', 0], 'GCG': ['A', 'four', 0], 'CGT': ['R', 'six', 0], 'CGC': ['R', 'six', 0], 'CGG': ['R', 'six', 0], 'CGA': ['R', 'six', 0], 'AGA': ['R', 'six', 0], 'AGG': ['R', 'six', 0], 'AAT': ['N', 'two', 0], 'AAC': ['N', 'two', 0], 'GAT': ['D', 'two', 0], 'GAC': ['D', 'two', 0], 'TGT': ['C', 'two', 0], 'TGC': ['C', 'two', 0], 'CAA': ['Q', 'two', 0], 'CAG': ['Q', 'two', 0], 'GAA': ['E', 'two', 0], 'GAG': ['E', 'two', 0], 'GGT': ['G', 'four', 0], 'GGC': ['G', 'four', 0], 'GGA': ['G', 'four', 0], 'GGG': ['G', 'four', 0], 'CAT': ['H', 'two', 0], 'CAC': ['H', 'two', 0], 'ATT': ['I', 'three', 0], 'ATC': ['I', 'three', 0], 'ATA': ['I', 'three', 0], 'ATG': ['M', 'one', 0], 'TTA': ['L', 'six', 0], 'TTG': ['L', 'six', 0], 'CTT': ['L', 'six', 0], 'CTC': ['L', 'six', 0], 'CTA': ['L', 'six', 0], 'CTG': ['L', 'six', 0], 'AAA': ['K', 'two', 0], 'AAG': ['K', 'two', 0], 'TTT': ['F', 'two', 0], 'TTC': ['F', 'two', 0], 'CCT': ['P', 'four', 0], 'CCC': ['P', 'four', 0], 'CCA': ['P', 'four', 0], 'CCG': ['P', 'four', 0], 'TCT': ['S', 'six', 0], 'TCC': ['S', 'six', 0], 'TCA': ['S', 'six', 0], 'TCG': ['S', 'six', 0], 'AGT': ['S', 'six', 0], 'AGC': ['S', 'six', 0], 'ACT': ['T', 'four', 0], 'ACC': ['T', 'four', 0], 'ACA': ['T', 'four', 0], 'ACG': ['T', 'four', 0], 'TGG': ['W', 'one', 0], 'TAT': ['Y', 'two', 0], 'TAC': ['Y', 'two', 0], 'GTT': ['V', 'four', 0], 'GTC': ['V', 'four', 0], 'GTA': ['V', 'four', 0], 'GTG': ['V', 'four', 0], 'TAA': ['*', 'none', 0], 'TGA': ['*', 'none', 0], 'TAG': ['*', 'none', 0], 'XXX': ['_missing', 'none', 0]} # Blepharisma (table 4) genetic code codon table, with 6-fold degenerate # codons kept whole, no splitting! blepharisma_6fold = {**universal_6fold, 'TGA': ['W', 'two', 0], 'TGG': ['W', 'two', 0], 'TAA': ['*', 'two', 0], 'TAG': ['*', 'two', 0]} # Blepharisma (table 4) genetic code codon table, with 6-fold degenerate # codons split into four-fold and two-fold groups. blepharisma_no6fold = {**universal_no6fold, 'TGA': ['W', 'two', 0], 'TGG': ['W', 'two', 0], 'TAA': ['*', 'two', 0], 'TAG': ['*', 'two', 0]} # Chilodonella genetic code codon table, with 6-fold degenerate # codons kept whole, no splitting! chilo_6fold = {**universal_6fold, 'CAA': ['Q', 'four', 0], 'CAG': ['Q', 'four', 0], 'TAA': ['*', 'one', 0], 'TAG': ['Q', 'four', 0], 'TGA': ['Q', 'four', 0]} # Chilodonella genetic code codon table, with 6-fold degenerate # codons split into four-fold and two-fold groups. # Note that this also splits four-fold degenerate codons that OUGHT to # be in "different" functional categories (e.g. CAG =/= TAG) chilo_no6fold = {**universal_no6fold, 'TAA': ['*', 'one', 0], 'TAG': ['Q_', 'one', 0], 'TGA': ['Q_', 'one', 0]} # Ciliate (table 6) genetic code codon table, with 6-fold degenerate # codons kept whole, no splitting! Traditional ciliate codon table. ciliate_6fold = {**universal_6fold, 'CAA': ['Q', 'four', 0], 'CAG': ['Q', 'four', 0], 'TAA': ['Q', 'four', 0], 'TAG': ['Q', 'four', 0], 'TGA': ['*', 'one', 0]} # Ciliate (table 6) genetic code codon table, with 6-fold degenerate # codons split into four-fold and two-fold groups. # Note that this also splits four-fold degenerate codons that OUGHT to # be in "different" functional categories (e.g. CAA =/= TAA) ciliate_no6fold = {**universal_no6fold, 'TAA': ['Q_', 'two', 0], 'TAG': ['Q_', 'two', 0], 'TGA': ['*', 'one', 0]} # Euplotes codon table, with 6-fold degenerate codons kept # whole, no splitting! Traditional Universal codon table. euplotes_6fold = {**universal_6fold, 'TGA': ['C', 'three', 0], 'TGT': ['C', 'three', 0], 'TGC': ['C', 'three', 0], 'TAA': ['*', 'two', 0], 'TAG': ['*', 'two',0]} # Euplotes genetic code codon table, with 6-fold degenerate codons # split into four-fold and two-fold groups. euplotes_no6fold = {**universal_no6fold, 'TGA': ['C', 'three', 0], 'TGT': ['C', 'three', 0], 'TGC': ['C', 'three', 0], 'TAA': ['*', 'two', 0], 'TAG': ['*', 'two',0]} # Mesodinium/Myrionecta (table 29) genetic code codon table, with 6-fold # degenerate codons kept whole, no splitting! Traditional ciliate codon table. mesodinium_6fold = {**universal_6fold, 'TAA': ['Y', 'four', 0], 'TAT': ['Y', 'four', 0], 'TAG': ['Y', 'four', 0], 'TAC': ['Y', 'four', 0], 'TGA': ['*', 'one', 0]} # Mesodinium/Myrionecta (table 29) genetic code codon table, with 6-fold # degenerate codons split into four-fold and two-fold groups. mesodinium_no6fold = {**universal_no6fold, 'TAA': ['Y', 'four', 0], 'TAT': ['Y', 'four', 0], 'TAG': ['Y', 'four', 0], 'TAC': ['Y', 'four', 0], 'TGA': ['*', 'one', 0]} # Peritrich (table 30) genetic code codon table, with 6-fold degenerate # codons kept whole, no splitting! Traditional ciliate codon table. peritrich_6fold = {**universal_6fold, 'GAA': ['E', 'four', 0], 'GAG': ['E', 'four', 0], 'TAA': ['E', 'four', 0], 'TAG': ['E', 'four', 0], 'TGA': ['*', 'one', 0]} # Peritrich (table 30) genetic code codon table, with 6-fold degenerate # codons split into four-fold and two-fold groups. # Note that this also splits four-fold degenerate codons that OUGHT to # be in "different" functional categories (e.g. CAA =/= TAA) peritrich_no6fold = {**universal_no6fold, 'TAA': ['E_', 'two', 0], 'TAG': ['E_', 'two', 0], 'TGA': ['*', 'one', 0]} cdnTableDict = {1:[universal_no6fold,universal_6fold], 4:[blepharisma_no6fold, blepharisma_6fold], 6:[ciliate_no6fold,ciliate_6fold], 10:[euplotes_no6fold,euplotes_6fold], 29:[mesodinium_no6fold,mesodinium_6fold], 30:[peritrich_no6fold,peritrich_6fold], 'chilodonella':[chilo_no6fold,chilo_6fold], 'chilo':[chilo_no6fold,chilo_6fold]} return cdnTableDict[gCode] def mapCdns(seq, cdnTable): # Updates the codon counts for a given sequence to the respective codon # count table (e.g. with or without 6-fold degeneracy). codons = [seq[n:n+3] for n in range(0, len(seq)-len(seq)%3, 3)] amb_cdn = 0 for c in codons: try: cdnTable[c][-1] += 1 except: amb_cdn += 1 if cdnTable['TCC'][1] == 'six': return cdnTable, amb_cdn else: return cdnTable class GCeval(): """ Returns %GC values from DNA sequences of various types. """ def gcTotal(seq): # This function returns global GC content return round(GC(seq), 4) def gc1(seq): # This function return the GC content of the first position of a codon return round(GC(''.join([seq[n] for n in range(0, len(seq), 3)])), 4) def gc2(seq): # This function return the GC content of the second position of a codon return round(GC(''.join([seq[n] for n in range(1, len(seq)-len(seq[1:]) % 3, 3)])), 4) def gc3(seq): # This function return the GC content of the third position of a codon return round(GC(''.join([seq[n] for n in range(2, len(seq)-len(seq[2:]) % 3, 3)])), 4) def gc3s(cdnTbl): # This function return the GC content of the third position of a codon excluding Tryp and Met syn = round(GC(''.join([k[-1]*v[-1] for k, v in cdnTbl.items() if v[0] != 'W' and v[0] != 'M'])), 4) return syn def gc3_4F(cdnTbl): # # This function return the GC content of the third position of four-fold # # degenerate codons FrFold = round(GC(''.join([k[-1]*v[-1] for k, v in cdnTbl.items() if v[1] == 'four'])), 4) return FrFold class SeqInfo(object): """ Provides a means to harbor the data for each individual contig/gene in a given fasta file. This includes GC content (various types), Effective Number of codons (ENc; again various calculations), Relative Synonymous Codon Usage (RSCU). """ def __init__(self,seq,gcode='universal'): self.ntd = str(seq) self.gcode, self.transTable = GenUtil.convertGenCode(gcode) # Dictionary of the GC-related functions/calculations self.gcFuncs = {'gcOverall':GCeval.gcTotal,'gc1':GCeval.gc1,'gc2':GCeval.gc2,'gc3':GCeval.gc3} def countCodons(self): # Stores the different codon tables and updates their codon counts cdnTbls = GenUtil.getCDNtable(self.transTable) self.cdnCounts_6F,self.amb_cdn = GenUtil.mapCdns(self.ntd, cdnTbls[1]) self.cdnCounts_No6F = GenUtil.mapCdns(self.ntd, cdnTbls[0]) def ENcStats(self): # Stores the various Effective Number of Codons calculations in the class self.expENc = CalcCUB.expWrightENc(self.gc3s) self.obsENc_6F = CalcCUB.calcWrightENc(self.cdnCounts_6F) self.obsENc_No6F = CalcCUB.calcWrightENc(self.cdnCounts_No6F) self.SunENc_6F = CalcCUB.SunEq5(self.cdnCounts_6F) self.SunENc_No6F = CalcCUB.SunEq5(self.cdnCounts_No6F) def GCstats(self): # Stores the various GC-stats in the class for k, v in self.gcFuncs.items(): setattr(self,k,v(self.ntd)) self.gc4F = GCeval.gc3_4F(self.cdnCounts_No6F) self.gc3s = GCeval.gc3s(self.cdnCounts_No6F) def RSCUstats(self): self.rscu_No6Fold = CalcCUB.RSCU(self.cdnCounts_No6F) self.rscu_6Fold = CalcCUB.RSCU(self.cdnCounts_6F) def prepFolders(outName): if os.path.isdir(outName) == False: os.mkdir(outName) if os.path.isdir(outName+'/SpreadSheets') == False: os.mkdir(outName+'/SpreadSheets') def CalcRefFasta(fasta, gCode): seqDB = {i.description:SeqInfo(i.seq, gCode) for i in SeqIO.parse(fasta,'fasta')} GenCDNtable = {} for k, v in seqDB.items(): v.countCodons() v.GCstats() v.ENcStats() for k, v in v.cdnCounts_6F.items(): if k.isalpha() and k not in GenCDNtable .keys(): GenCDNtable[k] = [v[0],v[-1]] else: GenCDNtable[k][-1] += v[-1] RSCU = CalcCUB.calcRCSU(GenCDNtable) return seqDB, RSCU def WriteWrightOut(seqData, outName, comp): if comp == False: with open(outName+'/SpreadSheets/'+outName.split('/')[-1]+'.ENc.Raw.tsv','w+') as w: w.write('SequenceID\tAmbiguousCodons\tGC-Overall\tGC1\tGC2\tGC3\tGC3S\t' 'GC3-Degen\tExpWrightENc\tObsWrightENc_6Fold\tObsWrightENc_No6Fold\t' 'ObsWeightedENc_6Fold\tObsWeightedENc_No6Fold\n') for k, v in seqData.items(): name = [k] gcs = [str(v.gcOverall),str(v.gc1),str(v.gc2),str(v.gc3),str(v.gc3s),str(v.gc4F)] ENc = [str(v.expENc),str(v.obsENc_6F),str(v.obsENc_No6F), str(v.SunENc_6F),str(v.SunENc_No6F)] w.write('\t'.join(name+[str(v.amb_cdn)]+gcs+ENc)+'\n') else: with open(outName+'/SpreadSheets/'+outName.split('/')[-1]+'.CompTrans.ENc.Raw.tsv','w+') as w: w.write('SequenceID\tAmbiguousCodons\tGC-Overall\tGC1\tGC2\tGC3\tGC3S\t' 'GC3-Degen\tExpWrightENc\tObsWrightENc_6Fold\tObsWrightENc_No6Fold\t' 'ObsWeightedENc_6Fold\tObsWeightedENc_No6Fold\n') for k, v in seqData.items(): name = [k] gcs = [str(v.gcOverall),str(v.gc1),str(v.gc2),str(v.gc3),str(v.gc3s),str(v.gc4F)] ENc = [str(v.expENc),str(v.obsENc_6F),str(v.obsENc_No6F), str(v.SunENc_6F),str(v.SunENc_No6F)] w.write('\t'.join(name+[str(v.amb_cdn)]+gcs+ENc)+'\n') def getCompFasta(fasta, gCode, require_start, require_stop): stopCDNs = {'1':['TAA','TAG','TGA'], '4':['TAA','TAG'], '6':['TGA'], '10':['TAA','TAG'], '29':['TGA'], '30':['TGA'], 'universal':['TAA','TAG','TGA'], 'blepharisma':['TAA','TAG'], 'ciliate':['TGA'],'euplotes':['TAA','TAG'], 'mesodinium':['TGA'], 'peritrich':['TGA'], 'chilo':['TAA']} if gCode.lower() not in stopCDNs.keys(): print('\nWARNING: Genetic code not recognized. Defaulting to universal.') stops = stopCDNs['1'] else: stops = stopCDNs[gCode] with open(fasta.replace('.fasta','.Comp.fasta'),'w+') as w: for i in SeqIO.parse(fasta,'fasta'): if len(i.seq) % 3 == 0: if ((require_start and str(i.seq).upper().startswith('ATG')) or not require_start) and ((require_stop and str(i.seq).upper()[-3:] in stops) or not require_stop): w.write('>'+i.description+'\n'+str(i.seq)+'\n') return fasta.replace('.fasta','.Comp.fasta') def WriteNullENcOut(outName): with open(outName+'/SpreadSheets/' + outName.split('/')[-1] + '.ENc.Null.tsv','w+') as w: w.write('GC3S\tENc\n') w.write('\n'.join(CalcCUB.nullENcGC3())) def WriteRSCUtbl(RSCUtbl, outName): with open(outName+'/SpreadSheets/' + outName.split('/')[-1] + '.RSCU.tsv','w+') as w: w.write('Codon\tAmino Acid\tRSCU\n') for k,v in RSCUtbl.items(): w.write(k+'\t'+'\t'.join(v)+'\n') def get_args(): parser = argparse.ArgumentParser( prog = 'Codon Usage Bias Statistic Calculator version 2.1', description = "Written by Xyrus Maurer-Alcala, updated July 19, 2023 by Auden Cote-L'Heureux" ) parser.add_argument('--input', '-i', required = True, default = None, type = str, help = 'Path to a fasta file of nucleotide sequences OR to a folder containing multiple fasta files') parser.add_argument('--genetic_code', '-g', default = 'universal', choices = { 'universal', 'blepharisma', 'ciliate', 'euplotes', 'mesodinium', 'peritrich', 'chilo'}, help = 'Genetic code to use (be careful if you are working with ciliates). Available codes are: "universal" (TAA, TAG, TGA), "blepharisma" (TAA, TAG), "ciliate" (TGA), "euplotes" (TAA, TAG), "mesodinium" (TGA), "peritrich" (TGA) and "chilo" (TAA). Ask someone if you are not sure what this means.') parser.add_argument('--require_start', action = 'store_true', help = 'Filter for sequences that begin with a start codon') parser.add_argument('--require_stop', action = 'store_true', help = 'Filter for sequences that end with a stop codon') return parser.parse_args() if __name__ == "__main__": args = get_args() #If only one file was input if os.path.isfile(args.input) and args.input.split('.')[-1] in ('fasta', 'fas', 'fa', 'fna'): outName = args.input.split('.')[0] compFasta = getCompFasta(args.input, args.genetic_code, args.require_start, args.require_stop) prepFolders(outName) fastaDataRaw, RSCUtbl = CalcRefFasta(args.input, args.genetic_code) fastaDataComp, RSCUtbl = CalcRefFasta(compFasta, args.genetic_code) WriteWrightOut(fastaDataRaw, outName, comp=False) WriteWrightOut(fastaDataComp, outName, comp=True) WriteNullENcOut(outName) WriteRSCUtbl(RSCUtbl, outName) os.system('mv ' + compFasta + ' ' + outName + '/') #If a folder of files was input elif os.path.isdir(args.input): if not os.path.isdir('TemporaryCUBOutput'): os.mkdir('TemporaryCUBOutput') for file in os.listdir(args.input): if file != '.DS_Store': if os.path.isfile(args.input + '/' + file) and file.split('.')[-1] in ('fasta', 'fas', 'fa', 'fna'): outName = 'TemporaryCUBOutput/' + file.split('.')[0] compFasta = getCompFasta(args.input + '/' + file, args.genetic_code, args.require_start, args.require_stop) prepFolders(outName) fastaDataRaw, RSCUtbl = CalcRefFasta(args.input + '/' + file, args.genetic_code) fastaDataComp, RSCUtbl = CalcRefFasta(compFasta, args.genetic_code) WriteWrightOut(fastaDataRaw, outName, comp=False) WriteWrightOut(fastaDataComp, outName, comp=True) WriteNullENcOut(outName) WriteRSCUtbl(RSCUtbl, outName) os.system('mv ' + compFasta + ' ' + outName + '/') else: print('\nWARNING: One of the files in your input folder (' + file + ') is not formatted correctly. These should be nucleotide fasta files with a file extension fasta, fas, fa, or fna. Skipping this file.\n') folders = ['TemporaryCUBOutput/' + folder for folder in os.listdir('TemporaryCUBOutput') if os.path.isfile('TemporaryCUBOutput/' + folder + '/SpreadSheets/' + folder + '.CompTrans.ENc.Raw.tsv')] #Combining information from all input files into one set of outputs if len(folders) > 0: if not os.path.isdir('CUBOutput'): os.mkdir('CUBOutput') if not os.path.isdir('CUBOutput/SpreadSheets'): os.mkdir('CUBOutput/SpreadSheets') os.system('cat TemporaryCUBOutput/*/*.Comp.fasta > CUBOutput/AllFiles.Comp.fasta') os.system('cp ' + folders[0] + '/SpreadSheets/*.ENc.Null.tsv CUBOutput/SpreadSheets/ENc.Null.tsv') with open('CUBOutput/SpreadSheets/RSCU.tsv', 'w') as o: o.write('File\tCodon\tAmino Acid\tRSCU\n') for folder in folders: for line in open(folder + '/SpreadSheets/' + folder.split('/')[-1] + '.RSCU.tsv'): if 'Amino Acid' not in line: o.write(folder.split('/')[-1] + '\t' + line) with open('CUBOutput/SpreadSheets/ENc.Raw.tsv', 'w') as o: o.write('File\tSequenceID\tAmbiguousCodons\tGC-Overall\tGC1\tGC2\tGC3\tGC3S\tGC3-Degen\tExpWrightENc\tObsWrightENc_6Fold\tObsWrightENc_No6Fold\tObsWeightedENc_6Fold\tObsWeightedENc_No6Fold\n') for folder in folders: for line in open(folder + '/SpreadSheets/' + folder.split('/')[-1] + '.ENc.Raw.tsv'): if 'SequenceID' not in line: o.write(folder.split('/')[-1] + '\t' + line) with open('CUBOutput/SpreadSheets/CompTrans.ENc.Raw.tsv', 'w') as o: o.write('File\tSequenceID\tAmbiguousCodons\tGC-Overall\tGC1\tGC2\tGC3\tGC3S\tGC3-Degen\tExpWrightENc\tObsWrightENc_6Fold\tObsWrightENc_No6Fold\tObsWeightedENc_6Fold\tObsWeightedENc_No6Fold\n') for folder in folders: for line in open(folder + '/SpreadSheets/' + folder.split('/')[-1] + '.CompTrans.ENc.Raw.tsv'): if 'SequenceID' not in line: o.write(folder.split('/')[-1] + '\t' + line) os.system('rm -r TemporaryCUBOutput') else: print('\nERROR: No composition information was created -- something probably went wrong with the formatting and/or filtering of the input fasta files. Make sure that this contains properly formatted nucleotide sequences (divisible by 3, etc).') else: print('\nERROR: Invalid --input. This should be a nucleotide fasta file with a file extension fasta, fas, fa, or fna or a folder of such files.\n')