ó Æ;ìWc@€s‘ddlmZddlZddlZddlZddlZdefd„ƒYZd„Zde d„Z d„Z d„Z d efd „ƒYZd efd „ƒYZddlZd efd„ƒYZddlZddlZidd6dd6dd6dd6dd6dd6dd6dd6dd6dd6d d!6d"d#6d$d%6d&d'6d"d(6dd)6d"d*6Zd+efd,„ƒYZd-„Zed.krejd/ƒZgZed0ƒZeeƒZejeƒxejƒD] ZeGHqÕWx¦eD]›Z e j!ƒGe j"ƒGe j#ƒj$ƒGHx5e%j&e ƒD]$Z'd1e'd2e'd3e'd4fGHq&Wx5ej&e ƒD]$Z'd1e'd2e'd3e'd4fGHq^WqëWndS(5iÿÿÿÿ(twith_statementNtSequencecB€s}eZdZdddd„Zd„Zd„Zd„Zd„Zd„Z d„Z d „Z d „Z d d „Z d „ZRS(sBiological sequence class. Sequence data is immutable. data: the sequence data as a tuple or string alpha: the alphabet from which symbols are taken name: the sequence name, if any info: can contain additional sequence information apart from the name tcC€sÇ||_||_t|ƒtkr6t|ƒ|_nNt|ƒtkrT||_n0t|ƒtkrxt|ƒ|_n tdƒ‚|d krÀt j |jƒ}|d krÃtdƒ‚qÃnt |t ƒrV|}t j |ƒ}|d krtd|dƒ‚n|j|jƒsÃtddj|jƒƒddj|jƒdƒ‚qÃnmt |t jƒr·|j|jƒsÃtddj|jƒƒddj|jƒdƒ‚qÃn td ƒ‚d S( sCreate a sequence with sequence data. Specifying the alphabet is optional, so is the name and info. Example: >>> myseq = sequence.Sequence('MVSAKKVPAIAMSFGVSF') will create a sequence with name "", and assign one of the predefined alphabets on basis of what symbols were used. >>> myseq.getAlphabet().getSymbols() will most likely output the standard protein alphabet: ('A', 'C', 'D', 'E', 'F', 'G', 'H', 'I', 'K', 'L', 'M', 'N', 'P', 'Q', 'R', 'S', 'T', 'V', 'W', 'Y') sASequence data is not specified correctly: must be string or tuples)Could not identify alphabet from sequences"No predefined alphabet with name "s"sInvalid alphabet specified: Rs" is not compatible with sequence 't'sUnexpected type for alphaN(tnametinfottypetstrttupletdatatlistt RuntimeErrortNonetalphabettgetBySeqt isinstancet basestringt getByNamet isValidStringtjoint getSymbolstAlphabet(tselftsequencetalphaRtseqinfot alphaname((sD/woldlab/castor/home/georgi/programs/meme_4.11.2/scripts/sequence.pyt__init__s0       ==cC€s|jS(sGet the name of the sequence(R(R((sD/woldlab/castor/home/georgi/programs/meme_4.11.2/scripts/sequence.pytgetName9scC€s|jS(sKGet additional info of the sequence (e.g. from the defline in a FASTA file)(R(R((sD/woldlab/castor/home/georgi/programs/meme_4.11.2/scripts/sequence.pytgetInfo<scC€s|jS(s7Retrieve the alphabet that is assigned to this sequence(R(R((sD/woldlab/castor/home/georgi/programs/meme_4.11.2/scripts/sequence.pyt getAlphabet?scC€s ||_dS(sChange the name of the sequenceN(R(RR((sD/woldlab/castor/home/georgi/programs/meme_4.11.2/scripts/sequence.pytsetNameBscC€sFt|tjƒr6|j|jƒsBtdƒ‚qBn tdƒ‚dS(sTSet the alphabet, throws an exception if it is not compatible with the sequence datasInvalid alphabet specifiedsUnexpected type for alphaN(RR RRR R (RR((sD/woldlab/castor/home/georgi/programs/meme_4.11.2/scripts/sequence.pyt setAlphabetEscC€s|jS(s/Retrieve the sequence data (a tuple of symbols)(R (R((sD/woldlab/castor/home/georgi/programs/meme_4.11.2/scripts/sequence.pyt getSequenceMscC€sdj|jƒS(s<Retrieve the sequence data as a string (copy of actual data)R(RR (R((sD/woldlab/castor/home/georgi/programs/meme_4.11.2/scripts/sequence.pyt getStringPscC€s t|jƒS(s2Get the length of the sequence (number of symbols)(tlenR (R((sD/woldlab/castor/home/georgi/programs/meme_4.11.2/scripts/sequence.pytgetLenSsicC€sh|dkrN||jƒ|krN|dkr9|j|S|j|||!Sntd|jƒƒ‚dS(sÊRetrieve a site in the sequence of desired length. Note that positions go from 0 to length-1, and that if the requested site extends beyond those the method throws an exception. iis/Attempt to access invalid position in sequence N(R$R R R(Rtpositiontlength((sD/woldlab/castor/home/georgi/programs/meme_4.11.2/scripts/sequence.pytgetSiteVs "  cC€s|jƒGdG|jƒGHdS(s% A short description of the sequence t:N(RR$(R((sD/woldlab/castor/home/georgi/programs/meme_4.11.2/scripts/sequence.pytnicecsN(t__name__t __module__t__doc__R RRRRRR R!R"R$R'R)(((sD/woldlab/castor/home/georgi/programs/meme_4.11.2/scripts/sequence.pyR s(         cC€s@g}t|ƒ}x'|jƒD]}|j|jƒƒqW|S(s‡ Read one or more lines of text from a file--for example an alignment. Return as a list of strings. filename: name of file (topent readlinestextendtsplit(tfilenamettxtlisttftline((sD/woldlab/castor/home/georgi/programs/meme_4.11.2/scripts/sequence.pyt readStringsgs  c C€sÂg}d }d }g}t|ƒ}|jƒ}xý|r/|ddkrç|rÁy>|rjdj|ƒ} nt||||ƒ} |j| ƒWqÁtk r½} tjd|dI| IJqÁXn|dd!}|j ƒd}g}n9|j ƒ} x*| D]"} |j t | j dƒƒƒqúW|jƒ}q3W|r¨y>|rQdj|ƒ} nt||||ƒ} |j| ƒWq´tk r¤} tjd|dI| IJq´Xn td ƒ‚|j ƒ|S( sð Read one or more sequences from a file in FASTA format. filename: name of file to load sequences from alpha: alphabet that is used (if left unspecified, an attempt is made to identify the alphabet for each individual sequence) it>Rs Warning: s is invalid (ignored): iiÿÿÿÿt*s/No sequences on FASTA format found in this fileN(R R-treadlineRRtappendR tsyststderrR0R/Rtstriptclose( R1Rt string_onlytseqlisttseqnameRtseqdatatfhtthislinetseqnewtet cleanlineR4((sD/woldlab/castor/home/georgi/programs/meme_4.11.2/scripts/sequence.pyt readFASTArsD   "     "  cC€s|jd|jƒdƒ|jƒ}|jƒddd}xGt|ƒD]9}dj||d|dd!ƒd}|j|ƒqLWdS(s:Write one sequence in FASTA format to an already open fileR6s ii<RN(twriteRR!R$trangeR(Rt filehandleR tlinestit lineofseq((sD/woldlab/castor/home/georgi/programs/meme_4.11.2/scripts/sequence.pyt_writeOneFASTAŸs  &cC€s‘t|dƒ}t|tƒr.t||ƒnKxH|D]@}t|tƒrZt||ƒq5tjd|jƒdIJq5W|jƒ|jƒdS(sEWrite a list (or a single) of sequences to a file in the FASTA formattwsWarning: could not write s (ignored).N( R-RRRNR:R;RtflushR=(RR1RBtseq((sD/woldlab/castor/home/georgi/programs/meme_4.11.2/scripts/sequence.pyt writeFASTAªs  tDistribcB€sheZdZdd„Zd„Zd d„Zd„Zd„Zd d„Z d„Z d „Z d „Z RS( sXClass for storing a multinomial probability distribution over the symbols in an alphabetgcC€sN||_||jjƒ|_gt|jjƒƒD] }|^q5|_dS(N(RR$ttotRItcnt(RRt pseudo_countt_((sD/woldlab/castor/home/georgi/programs/meme_4.11.2/scripts/sequence.pyR¼s cC€sCt|jƒ}tj|j|ƒ|_tj|j|ƒ|_|S(N(RSRtcopytdeepcopyRTRU(Rtmemotdup((sD/woldlab/castor/home/georgi/programs/meme_4.11.2/scripts/sequence.pyt __deepcopy__ÁscC€sg|dkr|jjƒ}nxB|D]:}|jj|ƒ}|j|cd7<|jd7_q%WdS(s Count an observation of a symbolgð?iN(R RRtgetIndexRURT(Rtsymstsymtidx((sD/woldlab/castor/home/georgi/programs/meme_4.11.2/scripts/sequence.pytcountÇs   cC€s|jjƒstdƒ‚n|jjƒ}g}xMtt|ƒƒD]9}||}|dkrk|}n|j|j|ƒqFW||_|S(s;Complement the counts, throw an error if this is impossiblesGAttempt to complement a Distrib based on a non-complementable alphabet.N( RtisComplementableR tgetComplementsRIR#R R9RU(Rtcomst new_countR`tcidx((sD/woldlab/castor/home/georgi/programs/meme_4.11.2/scripts/sequence.pyt complementÐs    cC€s8d|_gt|jjƒƒD] }d^q|_dS(s2Reset the distribution, that is, restart counting.igN(RTRIRR$RU(RRW((sD/woldlab/castor/home/georgi/programs/meme_4.11.2/scripts/sequence.pytresetßs cC€ss|jdkro|dkrHtg|jD]}||j^q(ƒ}|S|jj|ƒ}|j||jSndS(s£Determine the probability distribution from the current counts. The order in which probabilities are given follow the order of the symbols in the alphabet.iN(RTR RRURR](RR_tytfreqR`((sD/woldlab/castor/home/georgi/programs/meme_4.11.2/scripts/sequence.pytgetFreqäs )cC€sldjg|jjƒD]}d|^qƒg}|jdjgtj|ƒD]}d|^qNƒƒ|S(s_Retrieve the probabilites for all symbols and return as a pretty table (a list of text strings)Rs%4s s%3.2f (RRRR9RSRk(RtsttableRi((sD/woldlab/castor/home/georgi/programs/meme_4.11.2/scripts/sequence.pytprettyðs26cC€s |jjƒS(sMGet the symbols in the alphabet in the same order as probabilities are given.(RR(R((sD/woldlab/castor/home/georgi/programs/meme_4.11.2/scripts/sequence.pyRöscC€s|jS(s8Get the alphabet over which the distribution is defined.(R(R((sD/woldlab/castor/home/georgi/programs/meme_4.11.2/scripts/sequence.pyRúsN( R*R+R,RR\R RaRgRhRkRnRR(((sD/woldlab/castor/home/georgi/programs/meme_4.11.2/scripts/sequence.pyRSºs      tMotifcB€s2eZdZd„Zd„Zd„Zd„ZRS(s± Sequence motif class--defining a pattern that can be searched in sequences. This class is not intended for direct use. Instead use and develop sub-classes (see below). cC€sd|_||_dS(Ni(R#R(RR((sD/woldlab/castor/home/georgi/programs/meme_4.11.2/scripts/sequence.pyRs cC€s|jS(sGet the length of the motif(R#(R((sD/woldlab/castor/home/georgi/programs/meme_4.11.2/scripts/sequence.pyR$scC€s|jS(s*Get the alphabet that is used in the motif(R(R((sD/woldlab/castor/home/georgi/programs/meme_4.11.2/scripts/sequence.pyR scC€s:|}t|ƒtkr'|jƒ}n|jƒj|ƒS(s4Check if the sequence can be processed by this motif(RRR"RR(Rtseqstrtmystr((sD/woldlab/castor/home/georgi/programs/meme_4.11.2/scripts/sequence.pyt isAlphabets(R*R+R,RR$RRr(((sD/woldlab/castor/home/georgi/programs/meme_4.11.2/scripts/sequence.pyRos    tRegExpcB€s eZdZd„Zd„ZRS(sGA motif class that defines the pattern in terms of a regular expressioncC€s&tj||ƒtj|ƒ|_dS(N(RoRtretcompiletpattern(RRt re_string((sD/woldlab/castor/home/georgi/programs/meme_4.11.2/scripts/sequence.pyRscc€s¢|}t|ƒtk r-t||jƒ}n|jƒ}tj||ƒsdtd|jƒƒ‚nx7tj |j |ƒD] }|j ƒ|j ƒdfVqzWdS(sBFind matches to the motif in a specified sequence. The method is a generator, hence subsequent hits can be retrieved using next(). The returned result is a tuple (position, match-sequence, score), where score is always 1.0 since a regular expression is either true or false (not returned). s)Motif alphabet is not valid for sequence gð?N( RRRR"RoRrR RRttfinditerRvtstarttgroup(RRQtmyseqRqtm((sD/woldlab/castor/home/georgi/programs/meme_4.11.2/scripts/sequence.pytmatchs (R*R+R,RR}(((sD/woldlab/castor/home/georgi/programs/meme_4.11.2/scripts/sequence.pyRss tNAtLOGOTYPEt12tFONTSIZEt TITLEFONTSIZEt6t SMALLFONTSIZEs0.9t TOPMARGINt BOTTOMMARGINttruetYAXIStbitst YAXISLABELRt XAXISLABELtTITLEs1.0tERRORBARFRACTIONtfalset SHOWINGBOXs2.0tBARBITSt1tTICBITStSHOWENDSt NUMBERINGtOUTLINEtPWMcB€s˜eZdZd„Zdd„Zd„Zd„Zd„Zddd„Z d„Z d „Z d „Z d „Z d „Zd d„Zejƒd„ZRS(sãThis motif subclass defines a pattern in terms of a position weight matrix. An alphabet must be provided. A pseudo-count to be added to each count is optional. A uniform background distribution is used by default. cC€sBtj||ƒt|ƒ|_|jj|jƒƒd|_dS(Ni(RoRRSt backgroundRaRtnsites(RR((sD/woldlab/castor/home/georgi/programs/meme_4.11.2/scripts/sequence.pyRHsgcC€shd|_t|ƒ|_g}x¸|D]°}t|ƒtk rXt|tj|ƒƒ}n+tj||ƒsƒtd|j ƒƒ‚n|jdkr¤|j ƒ|_n$|j|j ƒkrÈtdƒ‚n|j |ƒq%Wgt |jƒD]}t tj|ƒ|ƒ^qé|_xEt |jƒD]4}x+|D]#}|j|j|j|ƒƒq-Wq W|j|_dS(s¦Set the probabilities in the PWM from an alignment. The alignment is a list of equal-length strings (see readStrings), OR a list of Sequence. iÿÿÿÿs)Motif alphabet is not valid for sequence s.Sequences in alignment are not of equal lengthN(tcolsR#R˜RRRoRRrR RR$R9RIRStcountsRaR'(RtalignedRVtseqsRlRWtc((sD/woldlab/castor/home/georgi/programs/meme_4.11.2/scripts/sequence.pytsetFromAlignmentNs$  7 %cC€s´d}t|jƒd}xt||kr|j|}|j||j|<||j|<|j|jƒ|j|jƒ|d7}|d8}qW||kr°|j|jƒn|S(sReverse complement the PWMii(R#RšRg(RRLtjttemp((sD/woldlab/castor/home/georgi/programs/meme_4.11.2/scripts/sequence.pytreverseComplementps    cC€s|jS(s)Get the number of sites that made the PWM(R˜(R((sD/woldlab/castor/home/georgi/programs/meme_4.11.2/scripts/sequence.pyt getNSites€scC€s7|jƒtj|ƒks*tdƒ‚n||_dS(sSet the background distributionsIncompatible alphabetsN(RRoR R—(Rtdistrib((sD/woldlab/castor/home/georgi/programs/meme_4.11.2/scripts/sequence.pyt setBackground„scC€sD|dkr,g|jD]}|jƒ^qS|j|j|ƒSdS(sGGet the probabilities for all positions in the PWM (a list of Distribs)N(R RšRk(RtcolR_Ri((sD/woldlab/castor/home/georgi/programs/meme_4.11.2/scripts/sequence.pyRkŠs  cC€sQg}xDtj|ƒD]3}|jdjg|D]}d|^q/ƒƒqW|S(saRetrieve the probabilites for all positions in the PWM as a pretty table (a list of text strings)Rs%8.6f (R–RkR9R(RRmtrowRi((sD/woldlab/castor/home/georgi/programs/meme_4.11.2/scripts/sequence.pyRn‘s1c C€sg}xrtj|ƒD]a}|jdjgtt|ƒƒD]2}dtj||||ƒtjdƒ^q;ƒƒqW|S(sfRetrieve the (base-2) log-odds for all positions in the PWM as a pretty table (a list of text strings)Rs%8.6f i(R–RkR9RRIR#tmathtlog(RtbkgRmR¦RL((sD/woldlab/castor/home/georgi/programs/meme_4.11.2/scripts/sequence.pyt logoddsPretty™s_cC€s©d}tj|ƒjƒ}x‡t|jƒD]v}|d}|j|j|ƒ}xC|dD]7}|j|j|ƒ}||kr\|}|}q\q\W||7}q+W|S(s¢ Get the consensus sequence corresponding to a PWM. Consensus sequence is the letter in each column with the highest probability. Rii(RoRRRIR™RšRk(Rt consensusR tpost best_lettertbest_ptlettertp((sD/woldlab/castor/home/georgi/programs/meme_4.11.2/scripts/sequence.pytconsensus_sequence£s   cC€s†g}xyt|jƒD]h}g}xLtj|ƒjƒD]5}|j|j|ƒ}|dkr8||7}q8q8W|j|ƒqW|S(s¢ Get the consensus corresponding to a PWM. Consensus at each column of motif is a list of characters with non-zero probabilities. i(RIR™RoRRRšRkR9(RR«R¬tmatchesR¯R°((sD/woldlab/castor/home/georgi/programs/meme_4.11.2/scripts/sequence.pyR«·s cC€sd}|jƒ|||j!}xvtt|ƒƒD]b}|j|j||ƒ}|dkrhd}ntj||jj||ƒƒ}||7}q3W|S(s[Score this particular list of symbols using the PFM (background needs to be set separately)gig-Cëâ6?( R!R™RIR#RšRkR§R¨R—(RRQRytsumRAR¬tqtlogodds((sD/woldlab/castor/home/georgi/programs/meme_4.11.2/scripts/sequence.pytgetScoreÈs  #iöÿÿÿcc€sÊ|}t|ƒtk r-t||jƒ}ntj||ƒsXtd|jƒƒ‚nxkt|jƒ|j ƒD]P}t j |||ƒ}||krr|dj |j ||j ƒƒ|fVqrqrWdS(s6Find matches to the motif in a specified sequence. The method is a generator, hence subsequent hits can be retrieved using next(). The returned result is a tuple (position, match-sequence, score). The optional parameter _LOG0 specifies a lower bound on reported logodds scores. s)Motif alphabet is not valid for sequence RN(RRRRoRrR RRIR$R™R–R¶RR'(RRQt_LOG0R{R¬tscore((sD/woldlab/castor/home/georgi/programs/meme_4.11.2/scripts/sequence.pyR}Ôs  c" C€s d}d}tj||ƒ}tj||ƒ}dd} d} |jƒdd} td| ƒ} t| | ƒ} t| | ƒ} tjƒ}||d <||d > def /mutedColourDict << s$ ({0:s}) [{1:.4f} {2:.4f} {3:.4f}] s>> def sUnknown variable "s" in EPS templateN(ttimetstrftimeR$tmintintt _eps_defaultsRXRR-RtRutsearchRyRHRzRItformatRRRkR§R¨R9tsortt getColourtroundtgetMutedColourR tendR#("Rtprogramt template_fileteps_fht timestampt small_dfmtt full_dfmtt small_datet full_datet points_per_cmtheighttwidtht points_heightt points_widthtdefaultst template_fhtm_varR4tlastR}tprevtkeyR¬tstacktalpha_icthR_Rjtstack_ictsymhtrgbRÃtgtb((sD/woldlab/castor/home/georgi/programs/meme_4.11.2/scripts/sequence.pytwriteEPSäs                       N(R*R+R,RRžR¡R¢R¤R RkRnRªR±R«R¶R}RÆt localtimeRí(((sD/woldlab/castor/home/georgi/programs/meme_4.11.2/scripts/sequence.pyR–Cs  "       cC€sG|jtƒ}x1tt|ƒƒD]}||j|ƒ||s\$ ^ - F  ù       %"