def column_parser(lines): """Parser column format""" record = False result = [] struct = [] seq = '' for line in lines: if line.startswith('; ------'): #structure part beginns record = True continue if line.startswith('; ******'): #structure part ends record = False struct = adjust_base(struct,-1) struct = Pairs(struct).directed()#remove duplicates struct.sort() result.append([seq,struct]) struct = [] seq = '' continue if record: sline = line.split() if sline[4] == '.': #skip not paired seq = ''.join([seq,sline[1]]) continue seq = ''.join([seq,sline[1]]) pair = (int(sline[3]),int(sline[4])) #(alignpos,align_bp) struct.append(pair) return result
def column_parser(lines): """Parser column format""" record = False result = [] struct = [] seq = '' for line in lines: if line.startswith('; ------'): #structure part beginns record = True continue if line.startswith('; ******'): #structure part ends record = False struct = adjust_base(struct, -1) struct = Pairs(struct).directed() #remove duplicates struct.sort() result.append([seq, struct]) struct = [] seq = '' continue if record: sline = line.split() if sline[4] == '.': #skip not paired seq = ''.join([seq, sline[1]]) continue seq = ''.join([seq, sline[1]]) pair = (int(sline[3]), int(sline[4])) #(alignpos,align_bp) struct.append(pair) return result
def ilm_parser(lines=None, pseudo=True): """Ilm format parser Takes lines as input and returns a list with Pairs object. Pseudo - if True returns pairs with possible pseudoknot if False removes pseudoknots """ pairs = [] for line in lines: if line.startswith('Final') or len(line) == 1: #skip these lines continue line = line.strip('\n') line = map(int, line.split(None, 2)) if line[1] == 0: continue #Skip this line, not a pair else: pairs.append(line) pairs = adjust_base(pairs, -1) tmp = Pairs(pairs).directed() tmp.sort() if not pseudo: tmp = opt_single_random(tmp) tmp.sort() result = [] result.append(tmp) return result
def ilm_parser(lines=None,pseudo=True): """Ilm format parser Takes lines as input and returns a list with Pairs object. Pseudo - if True returns pairs with possible pseudoknot if False removes pseudoknots """ pairs = [] for line in lines: if line.startswith('Final') or len(line)==1:#skip these lines continue line = line.strip('\n') line = map(int,line.split(None,2)) if line[1] == 0: continue #Skip this line, not a pair else: pairs.append(line) pairs = adjust_base(pairs,-1) tmp = Pairs(pairs).directed() tmp.sort() if not pseudo: tmp = opt_single_random(tmp) tmp.sort() result = [] result.append(tmp) return result
def parse_residues(residue_lines, num_base, unpaired_symbol): """Return RnaSequence and Pairs object from residue lines. residue_lines -- list of lines or anything that behaves like it. Lines should contain: residue_position, residue_identiy, residue_partner. num_base -- int, basis of the residue numbering. In bpseq files from the CRW website, the numbering starts at 1. unpaired_symbol -- string, symbol in the 'partner' column that indicates that a base is unpaired. In bpseq files from the CRW website, the unpaired_symbol is '0'. This parameter should be a string to allow other symbols that can't be casted to an integer to indicate unpaired bases. Checks for double entries both in the sequence and the structure, and checks that the structre is valid in the sense that if (up,down) in there, that (down,up) is the same. """ #create dictionary/list for sequence and structure seq_dict = {} pairs = Pairs() for line in residue_lines: try: pos, res, partner = line.strip().split() if partner == unpaired_symbol: # adjust pos, not partner pos = int(pos) - num_base partner = None else: # adjust pos and partner pos = int(pos) - num_base partner = int(partner) - num_base pairs.append((pos, partner)) #fill seq_dict if pos in seq_dict: raise BpseqParseError(\ "Double entry for residue %s (%s in bpseq file)"\ %(str(pos), str(pos+1))) else: seq_dict[pos] = res except ValueError: raise BpseqParseError("Failed to parse line: %s" % (line)) #check for conflicts, remove unpaired bases if pairs.hasConflicts(): raise BpseqParseError("Conflicts in the list of basepairs") pairs = pairs.directed() pairs.sort() # construct sequence from seq_dict seq = RnaSequence(construct_sequence(seq_dict)) return seq, pairs
def parse_residues(residue_lines, num_base, unpaired_symbol): """Return RnaSequence and Pairs object from residue lines. residue_lines -- list of lines or anything that behaves like it. Lines should contain: residue_position, residue_identiy, residue_partner. num_base -- int, basis of the residue numbering. In bpseq files from the CRW website, the numbering starts at 1. unpaired_symbol -- string, symbol in the 'partner' column that indicates that a base is unpaired. In bpseq files from the CRW website, the unpaired_symbol is '0'. This parameter should be a string to allow other symbols that can't be casted to an integer to indicate unpaired bases. Checks for double entries both in the sequence and the structure, and checks that the structre is valid in the sense that if (up,down) in there, that (down,up) is the same. """ #create dictionary/list for sequence and structure seq_dict = {} pairs = Pairs() for line in residue_lines: try: pos, res, partner = line.strip().split() if partner == unpaired_symbol: # adjust pos, not partner pos = int(pos) - num_base partner = None else: # adjust pos and partner pos = int(pos) - num_base partner = int(partner) - num_base pairs.append((pos,partner)) #fill seq_dict if pos in seq_dict: raise BpseqParseError(\ "Double entry for residue %s (%s in bpseq file)"\ %(str(pos), str(pos+1))) else: seq_dict[pos] = res except ValueError: raise BpseqParseError("Failed to parse line: %s"%(line)) #check for conflicts, remove unpaired bases if pairs.hasConflicts(): raise BpseqParseError("Conflicts in the list of basepairs") pairs = pairs.directed() pairs.sort() # construct sequence from seq_dict seq = RnaSequence(construct_sequence(seq_dict)) return seq, pairs
def ct_parser(lines=None): """Ct format parser Takes lines from a ct file as input Returns a list containing sequence,structure and if available the energy. [[seq1,[struct1],energy1],[seq2,[struct2],energy2],...] """ count = 0 length = '' energy = None seq = '' struct = [] result = [] for line in lines: count+=1 sline = line.split(None,6) #sline = split line if count==1 or new_struct(line):#first line or new struct line. if count > 1: struct = adjust_base(struct,-1) struct = Pairs(struct).directed() struct.sort() if energy is not None: result.append([seq,struct,energy]) energy = None else: result.append([seq,pairs]) struct = [] seq = '' #checks if energy for predicted struct is given if sline.__contains__('dG') or sline.__contains__('ENERGY'): energy = atof(sline[3]) if sline.__contains__('Structure'): energy = atof(sline[2]) else: seq = ''.join([seq,sline[1]]) if not int(sline[4]) == 0:#unpaired base pair = ( int(sline[0]),int(sline[4]) ) struct.append(pair) #structs are one(1) based, adjust to zero based struct = adjust_base(struct,-1) struct = Pairs(struct).directed() struct.sort() if energy is not None: result.append([seq,struct,energy]) else: result.append([seq,struct]) return result