xTc@ sddlmZddlmZddlZddlZddlmZddlmZddl Z ddl Z de fdYZ deddZed Zedd Zd Zejjd Zdde jd ZdZdS(i(tdivision(teasy_runN(tSorry(tadopt_init_argstalign_pdb_residuesc B seZdZdddddddedd ZddZdZdZdZ dZ ddd Z d Z d Z d Zd ZdZRS(s Provides mapping between original residue numbers or IDs and a reference numbering, via multiple sequence alignment. Depending on whether or not the insertion code was used in the original PDB files, it can use either the resseq or the resid as the lookup key. t reference_seqic C s0t|tt|} |dk s:|dk s:t| dkrX| t|ks^t|dk r|dks|tt|t|kst|dks|dkstn<t|t|kst|dks|dk stg|_d|_|jd| |jd|_ dS(Nitout( RtlocalstlentNonetAssertionErrort fasta_namest _name_lookupt run_alignmenttbuild_lookup_tableR( tselft pdb_sequencest pdb_namest pdb_offsetst pdb_residstreference_sequencetreference_sequence_nametreference_sequence_offsettreference_sequence_residstreference_indextsubstitute_namesRtn_models((s7/public2/ftp/pub/ccp4/6.5/ccp4-6.5/lib/py2/mmtbx/msa.pyt__init__s  $ !    c C st}|jrdi|_xUt|jD]5\}}t||j|<|jjt|q(Wn |j|_|jdk r|j dkst |j dk sT|j dk sTt n|j }t}|dkrd}|_ n|j||_|jdk r|j||_ n|j||_ |jrD|j||_n|j||_djgt|j|jD]\}}d||f^qp}|sd|j|jf} | d|}nt|d||_|jdk st dS(Nis s>%s %sR(tFalseRR t enumerateRtstrR tappendRR RR RRtTrueRRRRtjointziptget_muscle_alignmentt muscle_aln( RRtuse_pdb_sequencetitnameti_reftntstfastat ref_seq_fasta((s7/public2/ftp/pub/ccp4/6.5/ccp4-6.5/lib/py2/mmtbx/msa.pyR 5s<      /  cC si|_i|_i|_|jjj|j}|jj||_|dk sXt xMt |jjD]9\}}||krUd|j|<|j dkri}dg|j }d}}xp|jD]e}|dkr'|j|} | dk r| j} ||| <| ||toffsetti_resR]((s7/public2/ftp/pub/ccp4/6.5/ccp4-6.5/lib/py2/mmtbx/msa.pyRQs  N(t__name__t __module__t__doc__R RRR RR5RNRPRTRVRXRYRRRQ(((s7/public2/ftp/pub/ccp4/6.5/ccp4-6.5/lib/py2/mmtbx/msa.pyRs*  ' I       cC s"|dkrtj}ng}d}d}g}d} xt||D]\} } | jjdkspt| jj} | j dt } |j | | j dt }|j g|D]}| |^q| |kr|}|d| IJqI|dkrI|dkrI|}|d| IJqIqIW|dk rq|}t d|d|d|d |d |d |}n|jjdkst|jj} | j dt }d}}| j dt }g|D]}| |^q}t d|d|d|d |d |d |d |}|S(sT Convenience function: takes a collection of *single-model, single-chain* PDB hierarchies, plus optional reference hierarchy, extracts sequences and resseq offsets or resids, and generates alignment object. By default, if any of the hierarchies contain atoms with insertion codes, the resid mapping will be used automatically. icS s|dkrdS|jS(N(R R6(tstring_or_none((s7/public2/ftp/pub/ccp4/6.5/ccp4-6.5/lib/py2/mmtbx/msa.pyR6s itskip_insertionss! Using %s for sequence numberingRRRRRRRRN(R tsyststdoutR"toverall_countstn_chainsR t only_modelt only_chaintas_padded_sequenceRRtget_residue_idsR(t hierarchiesthierarchy_namestreference_hierarchyRtlogRR&RRR6t hierarchyR'tchaint chain_seqRWR=Rt msa_managerRRR((s7/public2/ftp/pub/ccp4/6.5/ccp4-6.5/lib/py2/mmtbx/msa.pytalign_pdb_hierarchiessX    &     cC s|s ttjjdds0tdntjjd}tjdkr^|d7}ntjj |stdnt |t std|}|r|d 7}nt j |d |}|jS( NR'tmuscles'MUSCLE not available or not configured.smuscle/exe/muscletnts.exes#muscle executable is not available.s%s -quiet -clws -groupt stdin_lines(R tlibtbxtenvt has_moduleR\t under_buildtosR'tpathtisfileRZRRtfully_bufferedt stdout_lines(tfasta_sequencestgroup_sequencestexe_pathtcmdt muscle_out((s7/public2/ftp/pub/ccp4/6.5/ccp4-6.5/lib/py2/mmtbx/msa.pyt run_muscle9s      cC s`t||}ddlm}|dj|\}}|dk r\|dj|IJn|S(Ni(tclustal_alignment_parses (Rtiotbx.bioinformaticsRR!R (RRRRRt alignmenttnull((s7/public2/ftp/pub/ccp4/6.5/ccp4-6.5/lib/py2/mmtbx/msa.pyR#Ls  c  sddlmi}x@t|ddD],\}}|j|d|}|||`sc3 s|]}|kVqdS(N((RR)(tlookup(s7/public2/ftp/pub/ccp4/6.5/ccp4-6.5/lib/py2/mmtbx/msa.pys fsR1R3tprogram(tiotbxRRtgetR#R!titemstdictR"R1R3talltvaluesR tclustal_alignmentR'R(t sequencestname_forR&RR'R((RRs7/public2/ftp/pub/ccp4/6.5/ccp4-6.5/lib/py2/mmtbx/msa.pytget_muscle_alignment_orderedTs ( !s muscle .caption = PHENIX includes the open-source multiple sequence alignment program MUSCLE, written by Bob Edgar. It can produce output identical in format to CLUSTALW, which is suitable for input to the Sculptor model preparation program. You may provide all sequences in a single file, or in as many different files as desired. Alternately, you may provide one or more PDB files from which the chain sequence(s) will be extracted (note however that this is only useful for homogenous models). .style = caption_img:icons/custom/msa64.png caption_width:480 { seq_file = None .type = path .multiple = True .short_caption = Sequence or PDB file .style = file_type:seq,pdb use_list input_file output_file = None .type = path .style = bold file_type:aln new_file load_in_text_editor = True .type = bool .short_caption = Open alignment in text editor when complete }cC s|dk st|jj}|jj}|dksB|dkr`tjjtjd}nddl m }ddl m }m }g}x|D]{} | jds| jds| jds| jd r|j| d d j} | j} | jd } t} x| jD]r}|jr|j}tjj| }d tjj|d |jf}|j|||t} qqW| stdd| qqy)|| dt\}}|j|Wqtk r }tdd| t|fqXqWt |dkr3tdndjg|D]}|j!d^q@}t"|}t#|dj$dj||dj|fS(Nts muscle.alni(t file_reader(tany_sequence_formatRs.pdbs.ents.pdb.gzs.ent.gzt force_typetpdbis%s_%ss2The PDB file %s does not contain any recognizable sprotein chains.tassign_name_if_not_defineds1Error parsing '%s' - not a recognizable sequence sformat. (Original message: %s)is0Need at least two valid sequences to run MUSCLE.s iPRH(%R R Rxtseq_filet output_fileRRR!tgetcwdRRRRRtendswithtany_filet file_objectRstmodelsRtchainst is_proteinRmtbasenametsplitexttidRR Rtextendt ExceptionRRtformatRRIRJ(targstparamsRt seq_filesRRRRtseqsRLtpdb_inRst first_modelt found_proteinRtRut base_nameRUt seq_objectst non_compliantR]RtcombinedR((s7/public2/ftp/pub/ccp4/6.5/ccp4-6.5/lib/py2/mmtbx/msa.pytrunsL      #  "+ cC s+t|jjdkr'tdndS(NisNo sequence files provided!(RRxRR(R((s7/public2/ftp/pub/ccp4/6.5/ccp4-6.5/lib/py2/mmtbx/msa.pytvalidate_paramss((t __future__RR{Rt libtbx.philtlibtbx.load_envt libtbx.utilsRRRRgtobjectRR R RwRR#Rtphiltparset master_philRhRR(((s7/public2/ftp/pub/ccp4/6.5/ccp4-6.5/lib/py2/mmtbx/msa.pyts$     :    )