o Rŀge@sdZddlZddlmZzddlmZWn eydZYnwddlmZddl m Z ddl m Z dd l mZd ZGd d d eZdd dZGdddZdS)aSBio.AlignIO support for the "maf" multiple alignment format. The Multiple Alignment Format, described by UCSC, stores a series of multiple alignments in a single file. It is suitable for whole-genome to whole-genome alignments, metadata such as source chromosome, start position, size, and strand can be stored. See http://genome.ucsc.edu/FAQ/FAQformat.html#format5 You are expected to use this module via the Bio.AlignIO functions(or the Bio.SeqIO functions if you want to work directly with the gapped sequences). Coordinates in the MAF format are defined in terms of zero-based start positions (like Python) and aligning region sizes. A minimal aligned region of length one and starting at first position in the source sequence would have ``start == 0`` and ``size == 1``. As we can see on this example, ``start + size`` will give one more than the zero-based end position. We can therefore manipulate ``start`` and ``start + size`` as python list slice boundaries. For an inclusive end coordinate, we need to use ``end = start + size - 1``. A 1-column wide alignment would have ``start == end``. N)islice)dbapi2)MultipleSeqAlignment)Seq) SeqRecord)SequentialAlignmentWriterc@s(eZdZdZddZddZddZdS) MafWriterz9Accepts a MultipleSeqAlignment object, writes a MAF file.cCs|jd|jddS)zWrite the MAF header.z##maf version=1 scoring=none z# generated by Biopython N)handlewriteselfrE/var/www/html/myenv/lib/python3.10/site-packages/Bio/AlignIO/MafIO.py write_header8s zMafWriter.write_headerc Cs|jddkr d}n |jddkrd}nd}dd|jdd d |jd d d |jdtt|jdd|d |jdd t|jg}|jd |ddS)zHWrite a single SeqRecord object to an 's' line in a MAF block (PRIVATE).strandr+-sz%-40s _z%15sstartrz%5ssizesrcSize N) annotationsgetidreplacelenstrseqr r join)rrecordrfieldsrrr _write_record=s  zMafWriter._write_recordcCst|ts tdtdd|Ddkrtdzddd|jD}Wn ty2d }Ynw|j d |d d }|D] }| ||d7}qA|j d |S) zWrite a complete alignment to a MAF block. Writes every SeqRecord in a MultipleSeqAlignment object to its own MAF block (beginning with an 'a' line, containing 's' lines). zExpected an alignment objectcSsh|]}t|qSrr").0xrrr ^sz,MafWriter.write_alignment..rz%Sequences must all be the same lengthrcSs&g|]\}}|dvr|d|qS))scorepass=r)r*r+yrrr hs  z-MafWriter.write_alignment..z score=0.00za rr) isinstancer TypeErrorr" ValueErrorr% _annotationsitemsAttributeErrorr r r()r alignmentannorecs_outr&rrrwrite_alignmentUs(     zMafWriter.write_alignmentN)__name__ __module__ __qualname____doc__rr(r;rrrrr 5s  r c cs d}g}g} zt|}Wn tyd}Ynw|r|dr|}t|dkr1td|ddkr:d }n |dd krCd }nd }t|d t|d |t|dd}|d} d| vr|sftd|dj} g} t | | D]\} } | | dkr~| n| qrd | } | t t | |d |d d|dnu|drno|drni|drnc|drn]|s|durt||ksJt|}||_|Vd}g}g}nszMafIterator..)next StopIteration startswithstriprMr"r4intr$zipappendr%rrrr5countdict)r seq_count in_a_bundlerrecordsline line_splitrr9sequencerefnewletter ref_letterr8 annot_stringsrrr MafIterators                  rdc@szeZdZdZddZddZddZdd Zd d Ze d d Z e ddZ ddZ ddZ dddZddZddZdS)MafIndexzIndex for a MAF file. The index is a sqlite3 database that is built upon creation of the object if necessary, and queried when methods *search* or *get_spliced* are used. c Cstdurddlm}|d||_||_tjtj||_ ||_ t |j |_ tj |rUt||_z||_Wn@tyT}z |j |j|dd}~wwt||_z||_Wnty{}z |j |j|dd}~wwt|j |_dS)z)Indexes or loads the index of a MAF file.Nr)MissingPythonDependencyErrorz.Python was compiled without the sqlite3 module)rBiorf_target_seqname_index_filenameospathabspathdirname_relative_path _maf_fileopen_maf_fpisfileconnect_con_MafIndex__check_existing_db _record_countr4close_MafIndex__make_new_indexrd_mafiter)r sqlite_filemaf_filetarget_seqnamerferrrrr__init__s<         zMafIndex.__init__cCs|jd|_dS)a?Close the file handle being used to read the data. Once called, further use of the index won't work. The sole purpose of this method is to allow explicit handle closure - for example if you wish to delete the file, on Windows you must first close all open handles to that file. rN)rtrwrvr rrrrw.s  zMafIndex.closec Cslzt|jdd}|tkr!dd|d|jg}t||jdd}tj |r4|}ntj |j | dtj j }|tj |jkrWtd|d |jd |jd d}||jkrotd ||jft|jd d}|dkrtdt|jdd}||krtd||f|WStjtjfy}ztd|dd}~ww)zEPerform basic sanity checks upon loading an existing index (PRIVATE).z1SELECT value FROM meta_data WHERE key = 'version'rrz=Index version (%s) incompatible with this version of MafIndexz;You might erase the existing index %s for it to be rebuilt.z2SELECT value FROM meta_data WHERE key = 'filename'/zIndex uses a different file (z != )z8SELECT value FROM meta_data WHERE key = 'target_seqname'z-Provided database indexed for %s, expected %sz6SELECT value FROM meta_data WHERE key = 'record_count'rz$Unfinished/partial database providedz SELECT COUNT(*) FROM offset_dataz.Expected %s records, found %s. Corrupt index?zProblem with SQLite database: N)rTrtexecutefetchoneMAFINDEX_VERSIONr%rir4rjrkisabsrnr!seprlrorhrOperationalError DatabaseError) r idx_versionmsgfilename tmp_mafpath db_target record_count records_foundr}rrr__check_existing_db9s  zMafIndex.__check_existing_dbcCs|jd|jddtf|jd|jdd|jftj|js=tj|js=tj |j|j  tjj d}n.tj tj|jtjj |j tjj rdtj |j|j  tjj d}ntj|j}|jdd|f|jdd }|} tt|d }|sn|jd ||j|t|7}q|jd |jd|jd|jd|d|j|S)z2Read MAF file and generate SQLite index (PRIVATE).z.CREATE TABLE meta_data (key TEXT, value TEXT);z1INSERT INTO meta_data (key, value) VALUES (?, ?);versionz?INSERT INTO meta_data (key, value) VALUES ('record_count', -1);r|rrzSCREATE TABLE offset_data (bin INTEGER, start INTEGER, end INTEGER, offset INTEGER);rTdzCINSERT INTO offset_data (bin, start, end, offset) VALUES (?,?,?,?);z9CREATE INDEX IF NOT EXISTS bin_index ON offset_data(bin);z=CREATE INDEX IF NOT EXISTS start_index ON offset_data(start);z9CREATE INDEX IF NOT EXISTS end_index ON offset_data(end);zUPDATE meta_data SET value = 'z' WHERE key = 'record_count')rtrrrhrjrkrrorirelpathrnr!rrmrlrR_MafIndex__maf_indexerlistr executemanycommitr")rmafpath insert_count mafindex_funcbatchrrr__make_new_indexsn       zMafIndex.__make_new_indexccs|j}|r|dr||jt|} |j}|r%|dr-td|jf|dr{|}|d|jkr{t |d}t |d}|t|d d d krgtd |t|d d d f||d}| ||d|||fVnq|j}|sd Sd S) zReturn index information for each bundle (PRIVATE). Yields index information for each bundle in the form of (bin, start, end, offset) tuples where start and end are 0-based inclusive coordinates. rLTz1Target for indexing (%s) not found in this bundlerrr rBrDrrz=Invalid length for target coordinates (expected %s, found %s)N) rqreadlinerRtellr"rSr4rhrMrTr!_ucscbin)rr\offsetr]rrendrrr __maf_indexers<          zMafIndex.__maf_indexercCsddg}|td|d?d|dd?|td|d?d|dd?|td|d ?d |dd ?|td |d ?d |dd ?t|S)zFind bins that a region may belong to (PRIVATE). Converts a region to a list of bins that it may belong to, including largest and smallest bins. rrr  IJIiJ)extendrangeset)rrbinsrrr _region2bins $$$$zMafIndex._region2bincCs^gd}d}d}|}|d}||L}||L}|D]}||kr$||S||L}||L}qdS)zReturn the smallest bin a given region will fit into (PRIVATE). Adapted from http://genomewiki.ucsc.edu/index.php/Bin_indexing_system )rrrrrrrBrrr)rr bin_offsets_bin_first_shift_bin_next_shift start_binend_bin bin_offsetrrrrs  zMafIndex._ucscbincCs|j|t|jS)zFRetrieve a single MAF record located at the offset provided (PRIVATE).)rqseekrPry)rrrrr _get_record+s  zMafIndex._get_recordc cspt|t|kr tdt||D]\}}||}|dkr'td|||fq|j}t}t||D]\}}zdtt|||}Wnt yUt d||fdw| d|||d|df} | } | D]I\} } } | | f|vrwqk| | | f| t| }|D])}|j|jkr|jd}||jd d}|| kr|| kstd | | | ||fq|Vqkq3dS) aSearch index database for MAF records overlapping ranges provided. Returns *MultipleSeqAlignment* results in order by start, then end, then internal offset field. *starts* should be a list of 0-based start coordinates of segments in the reference. *ends* should be the list of the corresponding segment ends (in the half-open UCSC convention: http://genome.ucsc.edu/blog/the-ucsc-genome-browser-coordinate-counting-systems/). z2Every position in starts must have a match in endsrz7Exon coordinates (%d, %d) invalid: exon length (%d) < 1z, z4Exon coordinates must be integers (start=%d, end=%d)NzSELECT DISTINCT start, end, offset FROM offset_data WHERE bin IN (%s) AND (end BETWEEN %s AND %s OR %s BETWEEN start AND end) ORDER BY start, end, offset ASC;rrz'Expected %s-%s @ offset %s, found %s-%s)r"r4rUrtrr%mapr#rr3rfetchalladdrrTr rhr)rstartsends exonstartexonendexonlenconyielded_rec_coords possible_binsresultrows rec_startrec_endrfetchedr&rrrrrsearch0sj      zMafIndex.searchrc Cs|dvr td|t|||}tddt||D}t|dkr3tttd||j dgSdd |D}d d |D}d}d } |D]} | D]n} | j |j krz&| d urf| j d } | dvretdn| | j d krxtd| j d | fWnt ytd|j d wt| } | j d} | j d}| |d}||7}| D]} t | |dD] }d|| j |<qqn qKtd|j d| }t | D],}| D]} | j |j kr| j|}|| j || j|7<q|dkr||kr|d7}qqGt||j |krtd|j t||j |fdd ||j D}i}|D]Q}||}g}||j kr0dnd}|j}t||D].\}}t ||D]#}||vrR|||qC||vra||||qC||qCq:d|||<q t||j dd|krtdt||j dd|j |ft||j }|D]\}}t||krtdt|||fqg}|D]\}}t|}|| kr|n|}|t|||ddqt|S)aReturn a multiple alignment of the exact sequence range provided. Accepts two lists of start and end positions on target_seqname, representing exons to be spliced in silico. Returns a *MultipleSeqAlignment* of the desired sequences spliced together. *starts* should be a list of 0-based start coordinates of segments in the reference. *ends* should be the list of the corresponding segment ends (in the half-open UCSC convention: http://genome.ucsc.edu/blog/the-ucsc-genome-browser-coordinate-counting-systems/). To ask for the alignment portion corresponding to the first 100 nucleotides of the reference sequence, you would use ``search([0], [100])`` )rrzStrand must be 1 or -1, got css|] \}}||VqdSNr)r*rrrrrrOsz'MafIndex.get_spliced..rNr cSsh|] }|D]}|jqqSrr)r*multiseqr^rrrr,sz'MafIndex.get_spliced..cSsi|]}|iqSrr)r*seq_namerrr sz(MafIndex.get_spliced..NrzStrand must be 1 or -1z8Encountered strand='%s' on target seqname, expected '%s'z-No strand information for target seqname (%s)rrrrz Did not find z in alignment bundlerz/Target seqname (%s) has %s records, expected %scSs&i|]\}}t|dkr|t|qSrr))r*posgapped_fragmentrrrrs  z9Returning %s letters for target seqname (%s), expected %sz'Returning length %s for %s, expected %s)r rFrG)r4rrsumrUr"rrrrhr rKeyErrorrr$r6rVr%r!reverse_complement) rrrrrexpected_letters all_seqnamessplit_by_positiontotal_rec_lengthref_first_strandrseqrec rec_lengthrungapped_lengthrrreal_pos gapped_pos track_valrealpos_to_lensubseqseqid seq_split seq_splice filler_charrVrrref_subseq_lenr$result_multiseqrrr get_spliceds         $            zMafIndex.get_splicedcCsd|jj|jfS)z,Return a string representation of the index.z%MafIO.MafIndex(%r, target_seqname=%r))rqrFrhr rrr__repr__VszMafIndex.__repr__cCs|jS)z*Return the number of records in the index.)rvr rrr__len__]szMafIndex.__len__Nr)r<r=r>r?r~rwrurxr staticmethodrrrrrrrrrrrres") FR1   cD rer)r?rj itertoolsrsqlite3r ImportError Bio.AlignrBio.Seqr Bio.SeqRecordr Interfacesrrr rdrerrrrs         L|