ó ¼™xTc@ sžddlmZddlZddlZddlZddlZddlZddl Zddl Zddl Z ddl Z ddl Z ddlZ ddlZddlZddlZddlmZdZidd6dd6d d 6Zddd gZd efd „ƒYZd efd„ƒYZdefd„ƒYZded„Zde jj fd„ƒYZ!d„Z"d„Z#e$dkršedej%dƒndS(i’’’’(tdivisionN(tSorrysā input { pdb_file_name_fixed = None .type = path .short_caption = Fixed model .help = Name of PDB file with model to fit to .style = bold file_type:pdb input_file pdb_file_name_moving = None .type = path .short_caption = Moving model .help = Name of PDB file with model that will be fit to pdb_file_name_fixed .style = bold file_type:pdb input_file } output { file_name = None .help = Name of PDB file with model that best fits to pdb_file_name_fixed .short_caption = Output model .type = path .style = bold new_file file_type:pdb include scope libtbx.phil.interface.tracking_params } selection_fixed = None .type = str .short_caption = Selection for fixed model .input_size = 400 .help = Selection of the target atoms to fit to (optional) selection_moving = None .type = str .short_caption = Selection for moving model .input_size = 400 .help = Selection of the atoms that will be fit to selection_fixed (optional) selection_fixed_preset = * ca backbone all .type = choice .help = Selection preset for fixed model. selection_moving_preset = * ca backbone all .type = choice .help = Selection preset for moving model. selection_fixed_chain = None .type = str .help = Selection chain for fixed model. selection_moving_chain = None .type = str .help = Selection chain for moving model. alignment .help = Set of parameters for sequence alignment. Defaults are good for most of cases .short_caption = Sequence alignment .style = box auto_align { alignment_style = local *global .type = choice gap_opening_penalty = 1 .type = float gap_extension_penalty = 1 .type = float similarity_matrix = blosum50 dayhoff *identity .type = choice }sIpepnames and (name ca or name n or name c) and altloc " " chain %(chain)stbackbones5pepnames and (name ca) and altloc " " chain %(chain)stcasall chain %(chain)stallt SuperposePDBc B seZdZiZdddddeddd„Zeddd„ƒZd„Z d„Z dddd„Z dddd„Z d„Z dd „Zdd „Zed „Zd „Zd „Zdd„Zeed„Zdd„Zd„Zd„Zd„Zd„Zed„ƒZRS(sØSuperimpose PDB files. Instantiate with a filename or pdb instance. You may also provide a default selection string, selection preset, or chain to use for atom selection. Otherwise, the select method will attempt to find a reasonable default. Use the superpose(target) method to perform superposition. This will return the RMSD, and the transformation used. To write output, use output(lsq, filename). The selectomatic(target) method performs a pairwise comparison of chain sequence alignments, and updates the selections to use the chains with the highest similarity. You can also update the selection using the select_update() method, or return a given selection using select(). Multiple models are supported by creating multiple instances. The class method open_models() helper can be used to open a file with multiple models. To create a subclass with a modified fitting routine, override fit(). c C sĖ||_||_|p'tjj|ƒ|_|dkrT|dkrTtdƒ‚n|pitj j |ƒ|_ |j j ƒ|_ t |j jƒƒdkr«tdƒ‚n|jd|d|d|ƒdS(sÕA filename or pdb instance is required. You may also provide arguments for the atom selection. The quiet option will suppress informational output. The desc option provides a label for the log output. s"Filename or PDB instance required.isQOnly one model supported; use SuperposePDB.open_models() to open multiple models.t selectiontchaintpresetN(t_quiett_logtostpathtbasenametdesctNoneRtiotbxtpdbtinputt pdb_inputtconstruct_hierarchytlentmodelst select_update( tselftfilenameRRRRtquiettlogR((s=/public2/ftp/pub/ccp4/6.5/ccp4-6.5/lib/py2/mmtbx/superpose.pyt__init__‡s  c K s tjj|ƒ}tjj|ƒjƒjƒ}|jddƒ}g}t|ƒdkr||j|d|d||ƒnŒt|ƒdkrüxwt|ƒD]Z\}}tjj j ƒ} | j |j ƒƒ|j|d| dd||f|ƒq›Wn t dƒ‚|S(s:Class method to return instances for each model in a file.RtmovingiRs%s-%ssNo models found!(RRRRRtpopRtappendt enumeratet hierarchytroott append_modelt detached_copyR( tclsRRtkwargsRRtrettcounttmodeltpdb_new((s=/public2/ftp/pub/ccp4/6.5/ccp4-6.5/lib/py2/mmtbx/superpose.pyt open_modelsŸs"0 cC s t‚dS(sReturn a transformed model.N(tNotImplemented(R((s=/public2/ftp/pub/ccp4/6.5/ccp4-6.5/lib/py2/mmtbx/superpose.pytget_transformedµsc C s}g}xå|jjƒD]Ō}|jdd|jƒ\}}}x©|jjƒD]˜}|jdd|jƒ\}}}|j||ƒ\}}|j||ƒ\} }|j|| ƒ} |j| ƒ} |j| |j|j| fƒqNWqW|stdƒ‚nt |ƒd} |j d| d| dfƒ|j | dƒ|j d | dƒ|j d | dƒ| d| dfS( s,Perform pairwise sequence alignments and find the best aligned chain pair. This method performs pairwise sequence alignments between all chains in itself and the target. The selections are updated to the pair with the highest sequence similarity, with chain IDs in the return value. Rs all chain %ssNo alignments?i’’’’s4Select-o-matic: Aligning chain %s to target chain %siiiR( Rtchainstselecttidt_extract_sequence_chaint _seqalignt_seqalign_scoreRRtsortedRt_print_seqalignR( Rttargett alignmentst moving_chaint_tmoving_selectiont target_chainttarget_selectiont moving_seqt target_seqt alignmenttscoretbest((s=/public2/ftp/pub/ccp4/6.5/ccp4-6.5/lib/py2/mmtbx/superpose.pyt selectomatic¹s$""'cC sL|d krd}nt|jjƒdd„ƒ}|pC|dj}|rj|jjƒjd|ƒ}n§tj|ƒr·tj|ƒ}|i|d6}|jjƒjd|ƒ}nZxWt D]O}t|}|i|d6}|jjƒjd|ƒ}|j t ƒr¾Pq¾q¾W|jj ƒ}|j ƒ|jƒj|ƒ}|||fS( saUpdate the selection. Specify either an explicit selection, or one of the handy preset selections (ca, backbone, all). Presets will use the longest chain by default, but this can also be specified using chain. If no selection or preset is specified, the various presets will be tested before falling back to all atom selection. RtnonetkeycS s |jƒS(N(t atoms_size(tx((s=/public2/ftp/pub/ccp4/6.5/ccp4-6.5/lib/py2/mmtbx/superpose.pytčsi’’’’tstringRN(NsNoneRC(RR4RR.R0tatom_selection_cacheRtPRESETStgett PRESET_ORDERR(tTruetatomst reset_i_seqt extract_xyzR/(RRRRR.tselected_atomsRNt selected_xyz((s=/public2/ftp/pub/ccp4/6.5/ccp4-6.5/lib/py2/mmtbx/superpose.pyR/Śs(     cC sZ|j|||ƒ\|_|_|_|jd|jt|jƒt|jƒfƒdS(sselect() and update state.s2Using selection: %s, atoms total: %s, selected: %sN(R/RRQRRRR(RRRR((s=/public2/ftp/pub/ccp4/6.5/ccp4-6.5/lib/py2/mmtbx/superpose.pyRs'cG s=|jr dS|jd|jIdjtt|ƒƒIJdS(sLog.Ns%s: t (R R Rtjointmaptstr(Rtmsg((s=/public2/ftp/pub/ccp4/6.5/ccp4-6.5/lib/py2/mmtbx/superpose.pyRs tcC sˆ|j|ƒ}||jƒ}y-tjj|ƒ}|jd||fƒWntk rl|jdƒnX|jd||fƒdS(sBPrint statistics on the RMSD between sites_moving and sites_fixed.s#Percentile-based spread (%s): %-.3fs:Could not calculate percentile-based spread... Please fix!s/RMSD between fixed and moving atoms (%s): %-.3fN(trms_differencetnormstlibtbxt math_utilstpercentile_based_spreadRt Exception(Rt sites_movingt sites_fixedRtrmsdtdeltastpbs((s=/public2/ftp/pub/ccp4/6.5/ccp4-6.5/lib/py2/mmtbx/superpose.pyt _print_rmsds c C sl|jdƒ|jd|jjdddtddƒƒ|jdƒ|j|jjdd ddƒƒd S( s,Print the transformation described by a lsq.s Rotation:s tlabeltrtone_row_per_linetformats%8.5fs Translation:ttN(RRftmathematica_formRMRi(Rtlsq((s=/public2/ftp/pub/ccp4/6.5/ccp4-6.5/lib/py2/mmtbx/superpose.pyt _print_lsqs , c C s‹|jƒ}|jdƒ|jd|jdƒ|jdƒƒ|jdƒ|pY|js‡|jd|ddd d d d d dƒndS(s#Print a sequence alignment details.sAlignment details:s matches after alignment: %st|t*s sequence alignment:tmatchest block_sizei2tn_blockittop_nameRt bottom_nametfixedN(RoRR(R t pretty_print(RR?RRo((s=/public2/ftp/pub/ccp4/6.5/ccp4-6.5/lib/py2/mmtbx/superpose.pyR5 s  '  cC sČt|ƒt|ƒkr=tdt|ƒt|ƒfƒ‚nt|ƒdkr^tdƒ‚nt|ƒdkrtdƒ‚ntjjjd|d|ƒ}|jƒ}|j|ƒ}||||fS(s‚Perform a least squares fit between sites_moving and sites_fixed. Return the rmsd, lsq, and updated sites_moving, sites_fixed.s1Cannot align different number of atoms: %s and %sis#Cannot align an empty set of atoms.isCannot align a single atom.treference_sitest other_sites(RRtscitbxtmatht superposetleast_squares_fittother_sites_best_fitRY(RR_R`Rkt sites_moving2Ra((s=/public2/ftp/pub/ccp4/6.5/ccp4-6.5/lib/py2/mmtbx/superpose.pytfit0s% c C s9|jƒ\}}|jƒ\}}|jdƒ|j||ƒ}|j|ƒ|j||||||j|jƒ}t|dd„dtƒ}|s§tdƒ‚n|j ƒ\}} } } |jdƒ|j | | ddƒ|jd ƒ|jd | j ƒƒ|j | j ƒ| dd ƒ|j | ƒ|| fS( s”Superpose to target. Return rmsd and lsq. This method will perform a sequence alignment first if the target is not sequence identical. s Let's go!RDcS s|dS(Ni((RF((s=/public2/ftp/pub/ccp4/6.5/ccp4-6.5/lib/py2/mmtbx/superpose.pyRG\streversesAFailed to get a good sequence alignment, or failed to fit models.sInitial stats:Rtstarts Best fit:s"Number of atoms for LS fitting: %stfinal(t_extract_sequence_rgsRR2R5t_superpose_alignRQR4RMRRRdtsizeR|Rl( RR6t seq_movingt str_movingt seq_fixedt str_fixedR?trmsdsRaRkt xyz_movingt xyz_fixed((s=/public2/ftp/pub/ccp4/6.5/ccp4-6.5/lib/py2/mmtbx/superpose.pyRzDs0      c c s6g} d}x#t|ƒD]\} } tjjjƒ} tjjjƒ} |j|d| ƒ}x¢|D]š\}}}xˆ||jjƒD]s}xj||jjƒD]U}|j|jkr¢||j r¢||j r¢| j |j ƒ| j |j ƒq¢q¢WqˆWqeWy|j | | ƒVWqt k r-}qXqWdS(sŠPerform a sequence alignment, try to find a sequence alignment window that produces the lowest lsq fit. Note that this is a generator.iiitwindowN(iii(R tcctbxt array_familytflext vec3_doublet_seqalign_pickmatchestrgRNtnameti_seqRtxyzR~R^(RR?R…R†R‡Rˆt sel_movingt sel_fixedtwindowsR‰R(RŒRŠR‹talignedtitjtktatom1tatom2te((s=/public2/ftp/pub/ccp4/6.5/ccp4-6.5/lib/py2/mmtbx/superpose.pyRƒis ,c C s||f}|jj|ƒr/|r/|j|Sidd6dd6dd6dd6}||d <||d ·s( Rotzipt itertoolsR(ti_seqs_ati_seqs_btatbRRR(RR?RŒRoR™RštiatibtimR“RµtlefttrighttwR((s=/public2/ftp/pub/ccp4/6.5/ccp4-6.5/lib/py2/mmtbx/superpose.pyR‘s(  "    cC sh|jƒ}t|jƒ|jjdƒ}|jdƒ}|jdƒ}d||td|ƒ}|S(Nt-RmRngY@i(RoRR“R(tmax(RR?RottotaltequaltsimilarR@((s=/public2/ftp/pub/ccp4/6.5/ccp4-6.5/lib/py2/mmtbx/superpose.pyR3½s  cC sng}g}xL|jjƒD];}|j||jƒ\}}|j|ƒ|j|ƒqWdj|ƒ|fS(s>Extract the sequence and residue groups of the selected atoms.RX(RR.R1RQRtextendRT(RtsequencetrgsRR“Rµ((s=/public2/ftp/pub/ccp4/6.5/ccp4-6.5/lib/py2/mmtbx/superpose.pyR‚Ås c C sgg}g}d}x?|jƒD]1}t}x.|jƒjƒD]}||r>t}Pq>q>W|rt|jƒƒdkrP|jƒd} tjj j j | dƒ} | dkrź|j | ƒ|j t jd|d|ƒƒ|d7}qMtjjd| jƒƒdkrM|j | jƒƒ|j t jd|d|ƒƒ|d7}qMqPqqWdj|ƒ|fS( s3Extract the sequence and residue groups of a chain.iitXR”R’R“tcommon_rna_dnaRX(tresidue_groupstFalseRNt extract_i_seqRMRtunique_resnamesRRtamino_acid_codestone_letter_given_three_letterRKRR[t group_argstcommon_residue_names_get_clasststripRT( RRRRĀRĆtcounterR’tgoodtaitresnametolc((s=/public2/ftp/pub/ccp4/6.5/ccp4-6.5/lib/py2/mmtbx/superpose.pyR1Ļs,    !c C s“|jjƒ}|j|jj|jƒ|jjƒ|jtj j j |j ƒddddddgƒƒ|jj d|d|jjƒƒdS(s6Output PDB model to filename given transformation lsq.i’’’’t file_nametcrystal_symmetryg(RRNtset_xyzRftelemsRPRitset_uijRRŽRtsym_mat3_doubleR„twrite_pdb_fileRRÕ(RRkRtoutput((s=/public2/ftp/pub/ccp4/6.5/ccp4-6.5/lib/py2/mmtbx/superpose.pyRŪės'7 cC sPtjjjƒ}x'|D]}|j|jƒjƒƒqW|jd|ƒdS(s5Merge multiple instances back into a single PDB file.RŌN(RRR!R"R#R-R$RŚ(R%t instancesRRRš((s=/public2/ftp/pub/ccp4/6.5/ccp4-6.5/lib/py2/mmtbx/superpose.pyt output_mergeös N(t__name__t __module__t__doc__RØRRĒRt classmethodR+R-RBR/RRRdRlR5R~RzRƒRMR2R‘R3R‚R1RŪRŻ(((s=/public2/ftp/pub/ccp4/6.5/ccp4-6.5/lib/py2/mmtbx/superpose.pyRls0   !)     %    tSuperposePDBSievecB seZd„ZRS(cC sItjjjd|d|ƒ}|jƒ}|j|ƒ}||||fS(NR`R_(RxRyRzt sieve_fitR|RY(RR_R`RkR}Ra((s=/public2/ftp/pub/ccp4/6.5/ccp4-6.5/lib/py2/mmtbx/superpose.pyR~’s (RŽRßR~(((s=/public2/ftp/pub/ccp4/6.5/ccp4-6.5/lib/py2/mmtbx/superpose.pyRāžstSuperposePDBSCEDScB seZd„ZRS(cC s t‚dS(N(R,(RR_R`((s=/public2/ftp/pub/ccp4/6.5/ccp4-6.5/lib/py2/mmtbx/superpose.pyR~s(RŽRßR~(((s=/public2/ftp/pub/ccp4/6.5/ccp4-6.5/lib/py2/mmtbx/superpose.pyRässphenix.superpose_pdbsc C sˆddl}t}tjjtdtƒ}|jjd|d|dtƒ}|j }d\}}t |j ƒdkr|j d|j d}}g}t jjjd |d dƒ} |j| jd d |ƒƒ|j| jd d |ƒƒ|jd|dtƒ\}} n|jƒ} | jj| jj| jj}}} |dkrh|dkrhtdƒ‚n|jd|ƒdGHt|d| jd| jd|d|d|ƒ}td| jd| j d|d|d|ƒ} x§t!tj"|| ƒD]\}}d|j#|j#fGH| js-|j$|ƒn|j%|ƒ\}}| ród| |f}d|j#|fGH|j|d|ƒqóqóWdS(Ni’’’’tprocess_includestargst master_paramst suppress_symmetry_related_errorsiiit master_philt home_scopetargspdb_file_name_fixed=%sspdb_file_name_moving=%stsourcesttrack_unused_definitionssNeed two input PDB files.touts ===== Init =====RRRRRs ===== Aligning %s to %s =====s %s-%s.pdbs$ ===== Writing %s output to %s =====R(NN(&t mmtbx.utilsRĒRtphiltparset PHIL_PARAMSRMtutilstprocess_command_line_argstparamsRRtpdb_file_namesR[t command_linetargument_interpreterRtprocesstfetchtextractRtpdb_file_name_fixedtpdb_file_name_movingRŪRŌRtshowRtselection_fixedtselection_fixed_presettdicttselection_movingtselection_moving_presetR R+RRBRz(Ręt command_nameRR©RRõRtRRģtparameter_interpreterR9tpeRŪt moving_argsR(RaRkR((s=/public2/ftp/pub/ccp4/6.5/ccp4-6.5/lib/py2/mmtbx/superpose.pytrun sR !  ! &      % tlaunchercB seZd„ZRS(cC stdt|jƒdtjƒS(NRęR(RtlistRętsyststdout(R((s=/public2/ftp/pub/ccp4/6.5/ccp4-6.5/lib/py2/mmtbx/superpose.pyRFs(RŽRßR(((s=/public2/ftp/pub/ccp4/6.5/ccp4-6.5/lib/py2/mmtbx/superpose.pyR EscC sS|dk rI|\}}|dfg}dtd|ƒfg}||fSggfS(NsSuperposed modeltRMSDs%.3f(Rt format_value(tresultt output_fileRat output_filest statistics((s=/public2/ftp/pub/ccp4/6.5/ccp4-6.5/lib/py2/mmtbx/superpose.pyt finish_jobIs    cC sX|jjdks$|jjdkr3tdƒ‚n|jjdkrTtdƒ‚ndS(NsOne or both PDB files missing.s*Please specify a name for the output file.(RRüRRżRRŪRŌ(Rõ((s=/public2/ftp/pub/ccp4/6.5/ccp4-6.5/lib/py2/mmtbx/superpose.pytvalidate_paramsQs t__main__Ręi(&t __future__RR R R±tcctbx.array_family.flexRt iotbx.pdbRtiotbx.pdb.amino_acid_codest iotbx.philR[tlibtbx.math_utilst libtbx.philtlibtbx.runtime_utilstmmtbx.alignmentR©t scitbx.mathRxtscitbx.math.superposet libtbx.utilsRRņRJRLtobjectRRāRäRRt runtime_utilsttarget_with_save_resultR RRRŽtargv(((s=/public2/ftp/pub/ccp4/6.5/ccp4-6.5/lib/py2/mmtbx/superpose.pyts>              N ’“<