o Rŀgu@sdZddlZddlmZddlmZddlmZddlmZddlm Z ddlm Z dd l m Z dd l mZdd lmZed e      d$ddZddZd%ddZ d&ddZddZddZ d&ddZd d!Zed"kr~dd#lmZedSdS)'z'Code for dealing with Codon Alignments.N)Iterable)Mapping)BiopythonExperimentalWarning)BiopythonWarning)CodonAlignment)mktest)CodonSeq CodonTable) SeqRecordztBio.codonalign is an experimental module which may undergo significant changes prior to its future official release.-XF c  Csddlm} t|| stdt|} |durRzt|} Wnty.t|}t|} Ynw| | kr>td| d| dt|trFd} nMt|trNd} nEtd t|t s[td t|| krt|trgni} |D]}|j }|| vr|td |d || |<qk| }d } n t dt|d| d| dkrt ||}ni| dkrt |}dd|D}||r||}tdd|dg}|D] }||||j fqn3| d krg}|D])}z||j }Wntytd|j dtdYnw||||fq|durtjd}g}d}|D]B}t|d|d||||d}|s:td|dj d|dj dt|d|d|||||d}|||dd krWd}q|rbtt|St|S)a\Build a codon alignment from protein alignment and corresponding nucleotides. Arguments: - pro_align - a protein MultipleSeqAlignment object - nucl_seqs - an object returned by SeqIO.parse or SeqIO.index or a collection of SeqRecord. - corr_dict - a dict that maps protein id to nucleotide id - complete_protein - whether the sequence begins with a start codon Return a CodonAlignment object. The example below answers this Biostars question: https://www.biostars.org/p/89741/ >>> from Bio.Seq import Seq >>> from Bio.SeqRecord import SeqRecord >>> from Bio.Align import MultipleSeqAlignment >>> from Bio.codonalign import build >>> seq1 = SeqRecord(Seq('ATGTCTCGT'), id='pro1') >>> seq2 = SeqRecord(Seq('ATGCGT'), id='pro2') >>> pro1 = SeqRecord(Seq('MSR'), id='pro1') >>> pro2 = SeqRecord(Seq('M-R'), id='pro2') >>> aln = MultipleSeqAlignment([pro1, pro2]) >>> codon_aln = build(aln, [seq1, seq2]) >>> print(codon_aln) CodonAlignment with 2 rows and 9 columns (3 codons) ATGTCTCGT pro1 ATG---CGT pro2 r)MultipleSeqAlignmentz:the first argument should be a MultipleSeqAlignment objectNz2Higher Number of SeqRecords in Protein Alignment (z,) than the Number of Nucleotide SeqRecords (z ) are found!zBNucl Sequences Error, Unknown type to assign correspondence methodzHcorr_dict should be a dict that corresponds protein id to nucleotide id!zDuplicate key ''zNumber of items in corr_dict (z*) is less than number of protein records ()cSsh|]}|jqSid.0irrK/var/www/html/myenv/lib/python3.10/site-packages/Bio/codonalign/__init__.py szbuild..Protein Record z, z= cannot find a nucleotide sequence match, please check the idzProtein record (z) is not in corr_dict!F)gap_char codon_tablecomplete_protein anchor_len and Nucleotide Record  do not match!)rrr max_scoreT) Bio.Alignr isinstance TypeErrorlentuple ValueErrorrrdictr RuntimeErrorzipsetkeysjoinappendKeyErrorprintexitr generic_by_id _check_corr_get_codon_recr_align_shift_recs) pro_align nucl_seqs corr_dictrunknownrrr r#rpro_numnucl_num corr_methoddrecordkey pro_nucl_pairnucl_idpro_iddiffpro_rec codon_alnshiftpair corr_span codon_recrrrbuilds ,                    rLcCsTd}t|D]!}tt|dkr|dt|7}q|ddt|d7}q|S)zFGenerate regular expression based on a given list of codons (PRIVATE).r[])r,r'r-r/)codonsregrrrr _codons2res  rR*cCsddlm}t||stdi}|jD] \}}||g|q|D] \}}t|||<q(t|j ||<d||<|S)aSet up the regular expression of a given CodonTable (PRIVATE). >>> from Bio.Data.CodonTable import generic_by_id >>> p = generic_by_id[1] >>> t = _get_aa_regex(p) >>> print(t['A'][0]) G >>> print(t['A'][1]) C >>> print(sorted(list(t['A'][2:]))) ['A', 'C', 'G', 'T', 'U', '[', ']'] >>> print(sorted(list(t['L'][:5]))) ['C', 'T', 'U', '[', ']'] >>> print(sorted(list(t['L'][5:9]))) ['T', 'U', '[', ']'] >>> print(sorted(list(t['L'][9:]))) ['A', 'C', 'G', 'T', 'U', '[', ']'] rr z;Input table is not a instance of Bio.Data.CodonTable objectz...) Bio.Data.CodonTabler r%r& forward_tableitems setdefaultr0rR stop_codons)rstopr;r aa2codoncodonaarPrrr _get_aa_regexs  r]cs&ddl}t|trt|tstdt|}d}|jD] } | |kr'||| 7}qt|j|d} | || } | rA| dfSt|j|dfddt dt D} t | dkrl| d| d| d<g}d} g}t | D]\}}t |}d}d}|kr|D]"} |r|dkr|t|j7}||d 7}q||| 7}||d 7}q| || } n&|krd}d}t t |D]}||||7}||d 7}q| || } | r|| | |f|kr||qv||qv|kr||qv||qvd|}| || } | r| d fSd }d}|r|dd dkrd d d d d d dg}| d}|D]p}|dkrKd}nf|dvrWd |}n|d d d d fvrpd d |}|dd|kr| |dd||dd}n | d|dd}t|||||\}}|dur|dkr||d<nq@|dkrtd|jtt t |d D]x}||d d||dd }d| ||d ||d d }| ||d||d d}d}|dkrt|||||\}}|dur0|dkr0|g|||d ||d d <d}q|dkr?td|jtq|dd d t | d kr| d}t |}d d d |d d |d dg}|D]x}|dkrtd}nn|dvr|d |}n|d |d d |d fvr|d d ||}t | |dd|kr| |dd|dd|}n | |ddd}t||||||\}}|dur|dkr|||d<nqi|dkrtd|jtd|}| || } | r| d | fStd|jd|jd)znCheck if the nucleotide can be translated into the protein (PRIVATE). Expects two SeqRecord objects. rNzB_check_corr accepts two SeqRecord object. Please check your input.rMcsg|] }||qSrrrr pro_seqrr sz_check_corr..r rTr)rrz&first frameshift detection failed for z'middle frameshift detection failed for z%last frameshift detection failed for zProtein SeqRecord (z) and Nucleotide SeqRecord (z) do not match!)rer%r r&r]seqstrupperreplacesearchspanranger' enumeraterR start_codonsr0startendr/_get_shift_anchor_rewarningswarnrrpopr+)pronuclrrrr rdaa2repro_rer\nucl_seqmatchanchorsanchor_distance anchor_posranchorthis_anchor_lenqcodonfncodon last_qcodon last_fcodonj full_pro_re first_anchor shift_id_pos shift_val_lstsh_anc shift_val sh_nuc_lensh_nucrr^rr5s2                      $&   "               r5c Csddl}ddtddD}d|krd|dkrsnnUtt|D]F}d } t|D]\} } | |krC| || d ||d 7} q,| || 7} q,| d 7} || |} | rj| d d d d } |d7}| |fSq$| sqd|fSdS|d|dd|dfvrd||}tdt|D]S}d } t|D]*\} } | |dkrq| |kr| t||d||||||7} q| || 7} q| d 7} || |} | r| d d d d } |d7}| |fSq| sd|fSdSdS)aFind a regular expression matching a potentially shifted anchor (PRIVATE). Arguments: - sh_anc - shifted anchor sequence - sh_nuc - potentially corresponding nucleotide sequence of sh_anc - shift_val - 1 or 2 indicates forward frame shift, whereas 3*anchor_len-1 or 3*anchor_len-2 indicates backward shift - aa2re - aa to codon re dict - anchor_len - length of the anchor - shift_id_pos - specify current shift name we are at rNcSsg|]}t|qSr)chrrrrrr`z(_get_shift_anchor_re..akrcr^(?P..*)$rMrra)rdrkr'rlrirh _merge_aa2rerg) rrrrvr rrdshift_idrrkr\ryrrrrps\         rpc Csndd}tt|||||f}|dkr_dt|ddt|dd@}d|d}|d |ddd d |ddd d |d d |ddd d |ddd 7}nR|dkrdt|ddt|dd@} dt|ddt|dd@} d|d}|d |ddd d | d d | d d |ddd 7}|d 7}|S) zDMerge two amino acids based on detected frame shift value (PRIVATE).cSshg}d}|D]+}|dkrd}|dq|dkrd}q|dkr(|d||d<q|dkr1||q|S)NrrNrarMrO)r0)re_aaaasmrrrrget_aa_from_codonres  z)_merge_aa2re..get_aa_from_codonrerrMrrr>rNrOr)listmapr/r-) aa1aa2rrvreidrscodon intersectscodonre intersect1 intersect2rrrrs&        &&     rcCsddl}ddlm}|j|d} |d} |d} t|} | dvrt|j|dd| d| dkrAtd|jd |jd d} t }|jD]}|d krT|d 7}qI|r| dkr| | d| dd}| t |j t |s|d8}td |jd|d| d|jd|d t|dkrtd|jd||7}| d7} qI| | dd| | dd| d}|j|d|kr|d8}td|j|| |j|ft|dkrtd|jd||7}| d7} qIt||jdS| dkrddlm}|g}g}|d}t|}|D]}||||||qg}|} |||d7}||vrm|||rm||}d||d||d}||||d|||d}n||vr|||r|||d}|| krnq-t }d} |jD]}|d kr|d 7}q|r| dkr| |d|dd}| t |j t |s|d8}td |jd|d| d|jd|d t||7}| d7} q| t|jd ddkr%|| d|| ddkr%|d8}|| }|d|}|}| ||d |}ny|| || dddkri|d8}|| dd}|| }d|| || dd}| ||d || || || d}n5|| }|d}| ||}|j|d|kr|d8}td|jd|d| d|jd|d t|dkrtd|jd||7}| d7} q||_!t||jdSdS)a3Generate codon alignment based on regular re match (PRIVATE). span_mode is a tuple returned by _check_corr. The first element is the span of a re search, and the second element is the mode for the match. mode - 0: direct match - 1: mismatch (no indels) - 2: frameshift rN)SeqrMr)rrrcrr!r"r ---zstart codon of z ( z) does not correspond to rzmax_score reached for z=! Please raise up the tolerance to get an alignment in anyway)tablez'%s(%s %d) does not correspond to %s(%s)rr)dequeT(z Codon of )"rdBio.Seqrrerhr]r'r)rrrirRrmrfrgrqrrrr+ translater collectionsrr groupdictr.r0rjrnindexisupperislowerrorf_table)rtru span_moderrrr#rdrrxrjmodervaa_num codon_seqr\ this_codonr shift_pos shift_startry m_groupdictrr shift_indexrrnrongaprrrr6;s &      (                 r6cs2dd}dd|D}dgt|}t|D]1\}}|jD]'ttr/||d7<q|jttddkrF||d7<qqtt|dkrTtd dt|} g}z fd d|D}Wn tyrY|Swt|D]\} }| | t|ft|t r|| jt|t|ddkr|||| \} } | |ddkrd| |dnddkr d t} t || jj d } | || jdt|| || jt|d7} || }| |}|d|fdd||ddD}||| <| || _|| |7| d7qwt||kr|D]l\} }t || jj d } d t} | || jdt|| || jt|d7} || }| |}g}tdt| dD] }| ||dqg|d||fdd||dD}||| <| || _q'd7q[)zBuild alignment according to the frameshift detected by _check_corr (PRIVATE). Argument: - recs - a list of SeqRecords containing a CodonSeq dictated by a rf_table (with frameshift in some of them). cSs<||}d} t|||tr||||fS|d7}q)NrTr)rr%int)rlstidxprrr find_next_ints z(_align_shift_recs..find_next_intcSsg|]}|jqSr)reget_full_rf_table)rrecrrrr`sz%_align_shift_recs..rrrcrz3Number of alignable codons unequal in given recordsTcsg|]}|qSrr)rr)rrrr`rr )rNcg|]}|tqSrrrvgap_numrrr` s g@crrrrrrrr`s)r'rlrerr%rr-r+ IndexErrorr0floatrrrrsrk)recsrfull_rf_table_lstrf_numrrrec_numadd_lst col_rf_lstrrrgapsre full_rf_tablebpinter_rftr)rrrr7s   )0  4      4  r7__main__) run_doctest)Nr r NFrr)rSr )Fr)__doc__rqcollections.abcrrBiorrBio.codonalign.codonalignmentrrBio.codonalign.codonseqrBio.Datar Bio.SeqRecordr rrrLrRr]r5rprr6r7__name__ Bio._utilsrrrrrsN           . # DD? P