:>"^%@sdZddlmZddlmZddlmZddlmZddl m Z ddl Z ddl Z y e jZWnek rdd ZYnXe jdkrddlZejZn6e jdkrddlZd d Zn d d Zy ejWnek r$ddZYn Xdd Zy ejZWn'ek rgddlZejZYnXdddZGdddeZedkrddlmZedddS)aBio.SeqIO support for the UCSC nib file format. Nib stands for nibble (4 bit) representation of nucleotide sequences. The two nibbles in a byte each store one nucleotide, represented numerically as follows: - ``0`` - T - ``1`` - C - ``2`` - A - ``3`` - G - ``4`` - N (unknown) As the first bit in a nibble is set if the nucleotide is soft-masked, we additionally have: - ``8`` - t - ``9`` - c - ``a`` - a - ``b`` - g - ``c`` - n (unknown) A nib file contains only one sequence record. You are expected to use this module via the Bio.SeqIO functions under the format name "nib": >>> from Bio import SeqIO >>> record = SeqIO.read("Nib/test_even_bigendian.nib", "nib") >>> print("%i %s..." % (len(record), record.seq[:20])) 50 nAGAAGagccgcNGgCActt... For detailed information on the file format, please see the UCSC description at https://genome.ucsc.edu/FAQ/FAQformat.html. )print_function) as_handle)SequenceWriter)Seq) SeqRecordNcCs |jdS)Nhex)decode)sr 4/tmp/pip-build-ww9dw3qa/biopython/Bio/SeqIO/NibIO.py7sr cCstj|jdS)Nascii)binasciihexlifyr)br r r r BscCs |jS)N)r)rr r r r EscCsD|dkr tjd|dS|dkr@tjd|dSdS)zConvert byte array to integer.littleziN)structunpack)r byteorderr r r byte2intKs  rcCstj||S)N)int from_bytes)rrr r r r Usc cs|dk rtdt|do}|jd}|sKtdt|}|dkrld}n!|dkrd }n td |jd}t||}|j}t|}|d d krt||kr6td nB|d dkr6t||dkr&td |d|}t|jdsWtdtdd} |j | } t | } t | } | VWdQRXdS)aIterate over a nib file and yield a SeqRecord. - handle - input file in the nib file format as defined by UCSC. This must be opened in binary mode! - alphabet - always ignored. Note that a nib file always contains only one sequence record. The sequence of the resulting SeqRecord object should match the sequence generated by Jim Kent's nibFrag utility run with the -masked option. This function is used internally via the Bio.SeqIO functions: >>> from Bio import SeqIO >>> record = SeqIO.read("Nib/test_even_bigendian.nib", "nib") >>> print("%s %i" % (record.seq, len(record))) nAGAAGagccgcNGgCActtGAnTAtCGTCgcCacCaGncGncTtGNtGG 50 You can also call it directly: >>> with open("Nib/test_even_bigendian.nib", "rb") as handle: ... for record in NibIterator(handle): ... print("%s %i" % (record.seq, len(record))) ... nAGAAGagccgcNGgCActtGAnTAtCGTCgcCacCaGncGncTtGNtGG 50 NzAlphabets are ignored.rbz Empty file.3a3de96br6be93d3arz"unexpected signature in Nib headerrzUnexpected file size 0123489abcz&Unexpected sequence data found in file TCAGNtcagn) ValueErrorrread bytes2hexrlensetissubset maketrans translaterr) handleZalphabetword signaturernumberlengthdataindicestable nucleotidessequencerecordr r r NibIteratoras<               r6c@s.eZdZdZddZddZdS) NibWriterzNib file writer.cCsc||_tj}|dkr'd}n%|dkr<d}ntd||jt|dS)zvInitialize an Nib writer object. Arguments: - handle - Output handle, in binary write mode. rrrrzunexpected system byte order %sN)r+sysr RuntimeErrorwrite hex2bytes)selfr+rr-r r r __init__s      zNibWriter.__init__c Csd}x|D]}|d7}q W|dkr9td|dkrQtd|j}|j}t|}t|}|jtjd|tdd}|d} | d } || 7}t |j d std |j |} |jt | |S) z=Use this to write an entire file containing the given record.rr zMust have one sequencezMore than one sequence foundir"r!rTZ ACGTNacgtnz0Sequence should contain A,C,G,T,N,a,c,g,t,n only) r#r+seqstrr&r:rpackr)r'r(r*r;) r<recordscountr5r+r4r3r/r2paddingsuffixr1r r r write_files*             zNibWriter.write_fileN)__name__ __module__ __qualname____doc__r=rGr r r r r7s  r7__main__) run_doctestverbose)r )r r) rK __future__rZBio.FilerZBio.SeqIO.InterfacesrZBio.SeqrZ Bio.SeqRecordrrr8bytesfromhexr;AttributeError version_inforrr%rrrrAr)stringr6r7rHZ Bio._utilsrMr r r r (sB              ?,