dTc@sddlZddlZddlmZddlmZmZddlmZm Z m Z ddl m Z ddl m Z dfdYZed krddlZddlZddlZejejjd d e_ndS( iN(tdeepcopy(trt FloatVector(tlog10tsqrttlog(tSummits(ttimetWigcBseZdddedZdZdZdZdZdZddd Z d d Z dFdd ddd dd edeedd Z idFd ddd eedZ dd d dded d dd ddFdFd Zdd d ddd dd ddFd Zdd ddFdZdZddZdZdZdZdZdZdd Zd!Zd"Zd#Zd$Zd%ed&Zed'Zed(Zed)Z ed*Z!d+Z"d,Z#d-Z$d.Z%dd/dd0d1gd%ed2Z&d3Z'd4Z(dd5Z)dFdFd6Z*d7Z+d8Z,d9dFed:Z-dFdFd9d;d<Z.d=Z/d>Z0d1d?Z1d@Z2dAZ3dBZ4dCZ5dFdDZ6dEZ7RS(GticCsi|_||_|dkrxt|D]}|j}tjdg|j|d<|dkr|j|djt|d|dddq+|j|djt|ddddq+Wn|dkr|j|d|d|ndS( s Parameter: file: a path to a file in Wiggle format step: each chrosome will present as an vector, each value in the vector represent a short region of the chrosome, the step value is the size of the short region. R giitrefchecktgfiletsuppressN( tdatatsteptopentsplittnumpytarraytresizetinttload(tselftfileR RR tlinetcol((sC/oak/stanford/groups/akundaje/marinovg/programs/danpos-2.2.2/wig.pyt__init__ s     33 cCsI|j}d}x3|D]+}x"||D]}|t|7}q'WqW|S(s Description: the sum of absolute value at each data point. Parameter: None Value: float value g(R tabs(Rtwigtsumtchrtv((sC/oak/stanford/groups/akundaje/marinovg/programs/danpos-2.2.2/wig.pytabsSums  cCsd|j|jkr1t|}|j|jni}x|jD]}d||twidthtdistancetpheighttheighttcalculate_P_valuetmodet title_linetpos_onlytfoldR tfdrt fdrSampleSizetppoistmtdicRttwidthtpksRtlthtstarttposR-tpstitoutftptwidth_above_cutoffRFteRtauctsmtstsmttpvl((sC/oak/stanford/groups/akundaje/marinovg/programs/danpos-2.2.2/wig.pyt callRegionss  <                 $(%         /|jc Cstd} |j} |dkr|dkrt| d}xJdtt| || jdtd|kr|d7}qCWndt| dG|GHt|d } |dkr| jd n | jd |j } |j } d}x|D]}||j }|j x|D]}|||||krFq"n|| |||| d}}| |j |kr| |j|dd dnd}t| |||!}|rt|d}nd}|}x||krZ| |||kr || 7}n|r;| ||dkrM|| ||7}qMn|| ||7}|d7}qW|| }||7}|r6tt| || jdtd}| j|d t|d t|||d t||||dd t|d t|d t|d td|dq"| j|d t|d t|||d t||||dd t|d t|d t|dq"WqWdG|GH| jdS(s& Add description here s?function(q,avg){return(ppois(q,avg,lower.tail=FALSE,log=TRUE))}ig?ii iswhole genome aveage value is s , use cutoffR2sHchr start end center width_above_cutoff total_signal height height_logP s<chr start end center width_above_cutoff total_signal height R s is stotal_width_above_cutoff:N(RRTRR9R:RRRR;RR R"RUR%RR4tclose(RtregionsRR[R\RYR]R`R RdReRnRRfttotal_width_above_cutoffRtstartsRjRFRqRpRRrRmRu((sC/oak/stanford/groups/akundaje/marinovg/programs/danpos-2.2.2/wig.pyt fillRegionssZ  <       !   / i(gh㈵>c% Cst|| }| dkrs| dkrM| r=|jdqp|jdqs| rc|jdqs|jdn|}td}|j}|dkr|dkrt|d}x@dtt|||jd |kr|d 7}qWnd t|d G|GHd GH|jd|dd d|d| }dGH|j d|d||rdGH|j d|d|d|n| dkrdGH|j |d|n|rdGHndGH|d|j :}|j }|d}|d|j }x|D]|}|j||j d }|GH|dkr.qn||d}||d}| dkru||d}||d}n|j}| r|t||d|}nd}x||krr||} d"\}!}"|dkr| dd|j d | dd|j d }!}"n| |j }#|#d }$x&|!dkrP|$|krLd }!n|#|$|dkr|j ||$t|j ||$||$|d !kr|$|j d }!qC|j ||$||kr|$|j d }!qC|j ||$|krC|$|j d }!qCn;|dkrC|$|j ||d krC|$|j d }!n|$d 8}$q+W|#d }$x9|"dkr|$||kr||j d }"n|$|#|dkrJ|j ||$t|j ||$||$|d !kr|$|j d }"q|j ||$||kr|$|j d }"q|j ||$|kr|$|j d }"qn?||d kr|$|j ||d kr|$|j d }"n|$d 7}$q^W|"|!kre| dkr| r?|j|d t|!d t|"d t| d d t||d t||d t||d t||d!qb|j|d t|!d t|"d t| d d t||d t||d!qe| r|j|d t|!d t|"d t| d d t||d t||d!qe|j|d t|!d t|"d t| d d t||d!n|d 7}qWqW|j|S(#s Description: This fuction is designed to call nucleosome positions Parameter: ofile: a path to the file used to save the positions. width: minimal width of positions distance: minimal distance between positions, neighboring positions with distance shorter than this value will be merged. edge: set to 1 if need to search for position edges, else set to 0 pcut: a P value cutoff used to call positions. height: the occupancy cutoff used to call positions. valide only when pheight is set to 1. fill_gap: fill the gap between two neighboring nuclesomes with a new nucleosome if the gap size is reasonable. fill_value: the default value to be set to a filled nucleosome calculate_P_value: calculate P value for each position if set to 1, else set to 0. mode: the mode to write result to ofile, could be either 'w' to create new file or 'a' to append to a existing file title_line: set to 1 if need a title line in the result file ofile poscal: set to 1 if need to calculate nucleosome positioning score and P value,else set to 0 Value: None. R2isRchr start end smt_pos smt_value smt_log10pval fuzziness_score fuzziness_log10pval s0chr start end smt_pos smt_value fuzziness_score s.chr start end smt_pos smt_value smt_log10pval s chr start end smt_pos smt_value sIfunction(q,avg){return(ppois(q,avg,lower.tail=FALSE,log.p=TRUE)/log(10))}g?iiswhole genome aveage value is s, use calling cutoffscalling summits ...RYtpcutR\Rxsmerging summits ...twgRZsfilling gaps ...scalculating positioning scoretrdssearching position edges ...ssaving positions ...iRoRRFtpposiJs s (ii(RR;RRTRR9R:Rt callSummitstmergetfillgapt positioningRR R5R%RtminRw(%RR>RYRZtedgeR|R\tfill_gapt fill_valueR]R^R_tposcalRxR~RnttwigRdReRsRftrhalfdisthalfdisRRitposestvalusRRttlentpvsRmRkRjtendtpppRo((sC/oak/stanford/groups/akundaje/marinovg/programs/danpos-2.2.2/wig.pyt callPositionsPs    2$           6   7' 7+  ffT ic "Cs%|} t}t||_|j| | dkrRdGH|j| d| nt|| }| dkr| dkr|r|jdq|jdq|r|jdq|jdntd }|j}|dkrH|dkrHt |d }x@dt t |||j d |krD|d 7}qWn|rVd GHndGH|d| j :}|j}|d}|d| j }x|D]|}| j|| j d }|GH|dkrqn||d}||d}| dkr||d}||d}n|j}|rD|t||d|}nd}x||kr||}d\}}|dkr|dd|j d |dd|j d }}n|| j } | d }!x&|dkr|!|krd }n| |!|dkr| j||!t| j||!||!|d !krK|!| j d }q| j||!||krz|!| j d }q| j||!|kr|!| j d }qn;|dkr|!| j ||d kr|!| j d }n|!d 8}!qW| d }!x9|dkr6|!||kr.|| j d }n|!| |dkr| j||!t| j||!||!|d !kr|!| j d }q)| j||!||kr|!| j d }q)| j||!|kr)|!| j d }q)n?||d kr)|!| j ||d kr)|!| j d }n|!d 7}!qW||kr| dkrE|r|j|dt |dt |dt |d dt ||dt ||dt ||dt ||dq|j|dt |dt |dt |d dt ||dt ||dq|r|j|dt |dt |dt |d dt ||dt ||dq|j|dt |dt |dt |d dt ||dn|d 7}qMWqW|j|S(s Description: This fuction is designed to call nucleosome positions Parameter: ofile: a path to the file used to save the positions. width: minimal width of positions distance: minimal distance between positions, neighboring positions with distance shorter than this value will be merged. edge: set to 1 if need to search for position edges, else set to 0 pcut: a P value cutoff used to call positions. height: the occupancy cutoff used to call positions. valide only when pheight is set to 1. fill_gap: fill the gap between two neighboring nuclesomes with a new nucleosome if the gap size is reasonable. fill_value: the default value to be set to a filled nucleosome calculate_P_value: calculate P value for each position if set to 1, else set to 0. mode: the mode to write result to ofile, could be either 'w' to create new file or 'a' to append to a existing file title_line: set to 1 if need a title line in the result file ofile poscal: set to 1 if need to calculate nucleosome positioning score and P value,else set to 0 Value: None. iscalculating positioning scoreR~R2sRchr start end smt_pos smt_value smt_log10pval fuzziness_score fuzziness_log10pval s0chr start end smt_pos smt_value fuzziness_score s.chr start end smt_pos smt_value smt_log10pval s chr start end smt_pos smt_value sIfunction(q,avg){return(ppois(q,avg,lower.tail=FALSE,log.p=TRUE)/log(10))}g?iissearching position edges ...ssaving positions ...iRoRRFRiJs s (ii(RRR tfetchValueFromWigRRR;RRTRR9R:RRR5R%RRRw("RRfRRYRZRR|R\R]R^R_RR~RRsRnRdReRRRRiRRRRRRRmRkRjRRRo((sC/oak/stanford/groups/akundaje/marinovg/programs/danpos-2.2.2/wig.pyt fillPositionss       2          6   7' 7+  ffT cCsV|dkrUtd}|j}tt|||jd}dG|GHnt}|d:}|d kri}xA|jD]3}i||<|j|j|j d|||t|||d |D]!}||c|j||7     cCs|jj|S(s Description: remove a chrosome Parameter: k: the name of the chrosome that is to be removed Value: None (R R,(Rtk((sC/oak/stanford/groups/akundaje/marinovg/programs/danpos-2.2.2/wig.pyR,s cCs]i}x|jD]}d|||j|| |j|| kr|| ||j|| :# cCsbi}x|jD]}d||s_ $