B b:@sdZddlmZddlmZddlmZddlmZddlmZddlm Z d d Z d d Z Gd ddeZ GdddeZ Gddde ZGdddeZddZddZedkrddlmZedddS)zBio.SeqIO support for the "fasta" (aka FastA or Pearson) file format. You are expected to use this module via the Bio.SeqIO functions. )Seq) SeqRecord)_clean)_get_seq_string)SequenceIterator)SequenceWriterccsx.|D]"}|ddkr|dd}PqWdSg}x\|D]T}|ddkr|d|ddddfVg}|dd}q:||q:W|d|ddddfVdS)aIterate over Fasta records as string tuples. Arguments: - handle - input stream opened in text mode For each record a tuple of two strings is returned, the FASTA title line (without the leading '>' character), and the sequence (with any whitespace removed). The title line is not divided up into an identifier (the first word) and comment or description. >>> with open("Fasta/dups.fasta") as handle: ... for values in SimpleFastaParser(handle): ... print(values) ... ('alpha', 'ACGTA') ('beta', 'CGTC') ('gamma', 'CCGCC') ('alpha (again - this is a duplicate entry to test the indexing code)', 'ACGTA') ('delta', 'CGCGC') r>rN  )rstripjoinreplaceappend)handlelinetitlelinesr0lib/python3.7/site-packages/Bio/SeqIO/FastaIO.pySimpleFastaParsers     rccsd}x~t|D]r\}}|ddkrP|ddkr>td|d|dd}q|ddkrrtd |d |d||fVqW|dkrn2|ddkrtd |dn|ddkstd dS) aIterate over no-wrapping Fasta records as string tuples. Arguments: - handle - input stream opened in text mode Functionally the same as SimpleFastaParser but with a strict interpretation of the FASTA format as exactly two lines per record, the greater-than-sign identifier with description, and the sequence with no line wrapping. Any line wrapping will raise an exception, as will excess blank lines (other than the special case of a zero-length sequence as the second line of a record). Examples -------- This file uses two lines per FASTA record: >>> with open("Fasta/aster_no_wrap.pro") as handle: ... for title, seq in FastaTwoLineParser(handle): ... print("%s = %s..." % (title, seq[:3])) ... gi|3298468|dbj|BAA31520.1| SAMIPF = GGH... This equivalent file uses line wrapping: >>> with open("Fasta/aster.pro") as handle: ... for title, seq in FastaTwoLineParser(handle): ... print("%s = %s..." % (title, seq[:3])) ... Traceback (most recent call last): ... ValueError: Expected FASTA record starting with '>' character. Perhaps this file is using FASTA line wrapping? Got: 'MTFGLVYTVYATAIDPKKGSLGTIAPIAIGFIVGANI' rr ziExpected FASTA record starting with '>' character. Perhaps this file is using FASTA line wrapping? Got: ''rNzoTwo '>' FASTA lines in a row. Missing sequence line if this is strict two-line-per-record FASTA format. Have '>z' and 'zjMissing sequence line at end of file if this is strict two-line-per-record FASTA format. Have title line 'z+line[0] == '>' ; this should be impossible!) enumerate ValueErrorr stripAssertionError)ridxrrrrrFastaTwoLineParserGs"$    r cs2eZdZdZd fdd ZddZddZZS) FastaIteratorzParser for Fasta files.Ncs,|dk rtd||_tj|ddddS)aNIterate over Fasta records as SeqRecord objects. Arguments: - source - input stream opened in text mode, or a path to a file - alphabet - optional alphabet, not used. Leave as None. - title2ids - A function that, when given the title of the FASTA file (without the beginning >), will return the id, name and description (in that order) for the record as a tuple of strings. If this is not given, then the entire title line will be used as the description, and the first word as the id and name. By default this will act like calling Bio.SeqIO.parse(handle, "fasta") with no custom handling of the title lines: >>> with open("Fasta/dups.fasta") as handle: ... for record in FastaIterator(handle): ... print(record.id) ... alpha beta gamma alpha delta However, you can supply a title2ids function to alter this: >>> def take_upper(title): ... return title.split(None, 1)[0].upper(), "", title >>> with open("Fasta/dups.fasta") as handle: ... for record in FastaIterator(handle, title2ids=take_upper): ... print(record.id) ... ALPHA BETA GAMMA ALPHA DELTA Nz,The alphabet argument is no longer supportedtZFasta)modefmt)r title2idssuper__init__)selfsourceZalphabetr%) __class__rrr's(zFastaIterator.__init__cCs||}|S)z9Start parsing the file, and return a SeqRecord generator.)iterate)r(rrecordsrrrparses zFastaIterator.parsec cs|j}|rFxt|D],\}}||\}}}tt||||dVqWnjxht|D]\\}}y|ddd}Wn(tk r|rtt|d}YnXtt||||dVqPWdS)z.Parse the file and generate SeqRecord objects.)idname descriptionNrrr )r%rrrsplit IndexErrorrrepr) r(rr%rsequencer.r/Zdescr first_wordrrrr+s zFastaIterator.iterate)NN)__name__ __module__ __qualname____doc__r'r-r+ __classcell__rr)r*rr!s-r!cs0eZdZdZfddZddZddZZS)FastaTwoLineIteratorz9Parser for Fasta files with exactly two lines per record.cstj|ddddS)aIterate over two-line Fasta records (as SeqRecord objects). Arguments: - source - input stream opened in text mode, or a path to a file This uses a strict interpretation of the FASTA as requiring exactly two lines per record (no line wrapping). Only the default title to ID/name/description parsing offered by the relaxed FASTA parser is offered. r"ZFASTA)r#r$N)r&r')r(r))r*rrr's zFastaTwoLineIterator.__init__cCs||}|S)z9Start parsing the file, and return a SeqRecord generator.)r+)r(rr,rrrr-s zFastaTwoLineIterator.parsec csnxht|D]\\}}y|ddd}Wn(tk rN|rFtt|d}YnXtt||||dVq WdS)z.Parse the file and generate SeqRecord objects.Nrrr )r.r/r0)r r1r2rr3rr)r(rrr4r5rrrr+s zFastaTwoLineIterator.iterate)r6r7r8r9r'r-r+r:rr)r*rr;s r;cs*eZdZdZdfdd ZddZZS) FastaWriterzClass to write Fasta format files (OBSOLETE). Please use the ``as_fasta`` function instead, or the top level ``Bio.SeqIO.write()`` function instead using ``format="fasta"``. <Ncs,t||r|dkrt||_||_dS)aCreate a Fasta writer (OBSOLETE). Arguments: - target - Output stream opened in text mode, or a path to a file. - wrap - Optional line length used to wrap sequence lines. Defaults to wrapping the sequence at 60 characters Use zero (or None) for no wrapping, giving a single long line for the sequence. - record2title - Optional function to return the text to be used for the title line of each record. By default a combination of the record.id and record.description is used. If the record.description starts with the record.id, then just the record.description is used. You can either use:: handle = open(filename, "w") writer = FastaWriter(handle) writer.write_file(myRecords) handle.close() Or, follow the sequential file writer system, for example:: handle = open(filename, "w") writer = FastaWriter(handle) writer.write_header() # does nothing for Fasta files ... Multiple writer.write_record() and/or writer.write_records() calls ... writer.write_footer() # does nothing for Fasta files handle.close() rN)r&r'rwrap record2title)r(targetr>r?)r*rrr's " zFastaWriter.__init__cCs|jr|||}nL||j}||j}|rN|ddd|krN|}n|r`d||f}n|}d|ksptd|ks|t|jd|t|}d|kstd|kst|j rxJt dt ||j D]"}|j||||j dqWn|j|ddS)z(Write a single Fasta record to the file.Nrrz%s %s r z>%s ) r?Zcleanr.r0r1rrwriterr>rangelen)r(recordrr.r0datairrr write_record&s&      $zFastaWriter.write_record)r=N)r6r7r8r9r'rHr:rr)r*rr<s)r<cs"eZdZdZdfdd ZZS)FastaTwoLineWritera_Class to write 2-line per record Fasta format files (OBSOLETE). This means we write the sequence information without line wrapping, and will always write a blank line for an empty sequence. Please use the ``as_fasta_2line`` function instead, or the top level ``Bio.SeqIO.write()`` function instead using ``format="fasta"``. Ncstj|d|ddS)aOCreate a 2-line per record Fasta writer (OBSOLETE). Arguments: - handle - Handle to an output file, e.g. as returned by open(filename, "w") - record2title - Optional function to return the text to be used for the title line of each record. By default a combination of the record.id and record.description is used. If the record.description starts with the record.id, then just the record.description is used. You can either use:: handle = open(filename, "w") writer = FastaWriter(handle) writer.write_file(myRecords) handle.close() Or, follow the sequential file writer system, for example:: handle = open(filename, "w") writer = FastaWriter(handle) writer.write_header() # does nothing for Fasta files ... Multiple writer.write_record() and/or writer.write_records() calls ... writer.write_footer() # does nothing for Fasta files handle.close() N)r>r?)r&r')r(rr?)r*rrr'PszFastaTwoLineWriter.__init__)N)r6r7r8r9r'r:rr)r*rrIEs rIcCst|j}t|j}|r2|ddd|kr2|}n|rDd||f}n|}d|ksTtd|ks`td|g}t|}d|ks~td|kstx2tdt|dD]}||||ddqWd |S) zTurn a SeqRecord into a FASTA formatted string. This is used internally by the SeqRecord's .format("fasta") method and by the SeqIO.write(..., ..., "fasta") function. Nrrz%s %srAr z>%s r=r ) rr.r0r1rrrCrDrr)rEr.r0rrrFrGrrras_fastars        rJcCst|j}t|j}|r2|ddd|kr2|}n|rDd||f}n|}d|ksTtd|ks`tt|}d|ksttd|kstd||fS)zTurn a SeqRecord into a two-line FASTA formatted string. This is used internally by the SeqRecord's .format("fasta-2line") method and by the SeqIO.write(..., ..., "fasta-2line") function. Nrrz%s %srAr z>%s %s )rr.r0r1rr)rEr.r0rrFrrras_fasta_2lines      rK__main__) run_doctest)verboseN)r9ZBio.SeqrZ Bio.SeqRecordrZ Interfacesrrrrrr r!r;r<rIrJrKr6Z Bio._utilsrMrrrrs"      /BI$O-