ó xÈ Yc@sÔdZddlZddlZddlZddlmZddlZddlmZddl Tddl m Z ddl m Z mZddlmZddlTd „Zd „Zd d „Zd „Zd„ZdS(s³Description: MACS 2 main executable Copyright (c) 2008,2009 Yong Zhang, Tao Liu Copyright (c) 2010,2011 Tao Liu This code is free software; you can redistribute it and/or modify it under the terms of the BSD License (see the file COPYING included with the distribution). @status: release candidate @version: $Id$ @author: Yong Zhang, Tao Liu @contact: taoliu@jimmy.harvard.edu iÿÿÿÿN(tstrftime(t opt_validate(t*(tbinomial_cdf_inv(t PeakModeltNotEnoughPairsException(t PeakDetectcCsšt|jƒƒ}t|jƒƒ}|j|ƒ}t|ƒdkr–|dƒ|ddjt|ƒƒƒ|ddjt|ƒƒƒtjƒndS(scheck common chromosome namesismNo common chromosome names can be found from treatment and control! Check your input files! MACS will quit...s!Chromosome names in treatment: %st,sChromosome names in control: %sN(tsett get_chr_namest intersectiontlentjointsortedtsystexit(ttreattcontrolt error_streamt tchrnamest cchrnamest commonnames((sP/oak/stanford/groups/akundaje/marinovg/programs/MACS-2.1.0/MACS2/callpeak_cmd.pyt check_names(s c Cs t|ƒ}|j}|j}|j}|j}|d|jƒ|jdak|_|jred}nd}|jt _|d|ƒ|jr¢t |ƒ\}}nt |ƒ\}}|d`k rÓt |||ƒn|d||jƒd||jf} |j} | d || f7} |d || ƒ|jd krR|jd krt|d |ƒt|j| ƒ} |d| ƒn|d|ƒt|jƒ} |jr©|d| ƒn |d| ƒ|j| ƒ|j} |d|| ƒ| d|| f7} |jr | d| 7} n| d| 7} |dt| | ƒ| ƒ| dt| | ƒ| 7} n| } |d`k rÇ|j} | d|| f7} |d|| ƒ|jd kr¾|jd krß|d|ƒt|j| ƒ}|d|ƒn|d|ƒt|jƒ}|jr|d| ƒn |d| ƒ|j| ƒ|j}|d||ƒ| d||f7} |jru| d| 7} n| d| 7} |dt| |ƒ| ƒ| d t| |ƒ| 7} qÇ| }n|d!ƒ|d"ƒ|jr€|d#ƒ|jr|j|_n |j|_|jd$kr5|d%|jƒn'|jd$kr\|d&|jd'ƒn|d(|jƒd)|j|_nÑymtd*|d+td,|ƒ}|d-ƒ|d.ƒ|d/|jƒ|d0|jƒ|d1|jƒ|d2|jƒ|d3d4jt t!|j"ƒƒƒ|d5|j#ƒt$||j#|j%ƒ|j|_d)|j|_|jd)|jkrì|d6|jƒ|j&r¼|j|_d)|j|_|d7|jƒqì|d8d4jt t!|j"ƒƒƒ|d9ƒnWnat'k rP|j&st(j)d:ƒn|d#ƒ|j|_d)|j|_|d;|jƒnX|d<ƒ|j*rq|d=ƒn|r|jr|d)}n|rÞ|j+r|d>ƒ| |kr|d?|ƒ|j,d$krÛ|d@ƒn|dA|j,ƒ|j-||j,ƒ|dB|jƒno|| kr|dC|ƒ|j,d$krG|d@ƒn|dA|j,ƒ|j-| |j,ƒ|dD|j|ƒnt.|_/qÞ|j0rº| |kr®t.|_/qÛt1|_/qÞ| |krÒt1|_/qÞt.|_/nt2dE|dF|d,|ƒ}|j3ƒ|j4j5dG|j6ƒ|dH|j7ƒt8|j7dIƒ}|j9dJt:ƒ|j9|jdƒ|j9| ƒ|jd$kr”|j9dK|jƒn*|jd$kr¾|j9dL|jd'ƒn|j9dM|jƒy*|j9dNd4jt t!|j"ƒƒƒWnnX|j*r |j9dOƒn|j4j;|dP|j%ƒ|j<ƒ|j=d`krZ dQ}n|j>d`krr dR}n|j?sH |dS|j@ƒt8|j@dIƒ}|j4jA|dTdUdP|j%dV|dW|jBƒ|j<ƒ|dX|jCƒt8|jCdIƒ}|j4jD|dTdUdP|j%dYdZtEd[ƒd\dV|dW|jBƒ|j<ƒn¼|d]|jFƒt8|jFdIƒ}|j4jG|dTdUdP|j%dY|j%dW|jBƒ|j<ƒ|d^|jHƒt8|jHdIƒ}|j4jI|dTdUdP|j%dY|j%dW|jBƒ|j<ƒ|d_ƒd`S(bs.The Main function/pipeline for MACS. s tBAMPEtBEDPEtfragmentttags#1 read %s files...s#1 %s size = %ds"# %s size is determined as %d bps s# total %ss in treatment: %d s#1 total %ss in treatment: %dtalltautosS#1 calculate max duplicate %ss in single position based on binomial distribution...s'#1 max_dup_tags based on binomial = %ds"#1 user defined the maximum %ss...sN#1 filter out redundant fragments by allowing at most %d identical fragment(s)sc#1 filter out redundant tags at the same location and the same strand by allowing at most %d tag(s)s(#1 %ss after filtering in treatment: %ds'# %ss after filtering in treatment: %d s0# maximum duplicate fragments in treatment = %d s@# maximum duplicate tags at the same position in treatment = %d s%#1 Redundant rate of treatment: %.2fs$# Redundant rate in treatment: %.2f s# total %ss in control: %d s#1 total %ss in control: %dsa#1 for control, calculate max duplicate %ss in single position based on binomial distribution...s&#1 %ss after filtering in control: %ds%# %ss after filtering in control: %d s.# maximum duplicate fragments in control = %d s># maximum duplicate tags at the same position in control = %d s##1 Redundant rate of control: %.2fs"# Redundant rate in control: %.2f s #1 finished!s#2 Build Peak Model...s #2 Skipped...is9#2 Sequencing ends will be shifted towards 3' by %d bp(s)s9#2 Sequencing ends will be shifted towards 5' by %d bp(s)iÿÿÿÿs#2 Use %d as fragment lengthit treatmentt max_pairnumtopts #2 finished!s#2 Summary Model:s#2 min_tags: %ds #2 d: %ds#2 scan_window: %ds&#2 predicted fragment length is %d bpss/#2 alternative fragment length(s) may be %s bpsRs%#2.2 Generate R script for model : %ss…#2 Since the d (%.0f) calculated from paired-peaks are smaller than 2*tag length, it may be influenced by unknown sequencing problem!sˆ#2 MACS will use %d as EXTSIZE/fragment length d. NOTE: if the d calculated is still acceptable, please do not use --fix-bimodal option!sA#2 You may need to consider one of the other alternative d(s): %ss#2 You can restart the process with --nomodel --extsize XXX with your choice or an arbitrary number. Nontheless, MACS will continute computing.isB#2 Since --fix-bimodal is set, MACS will use %d as fragment lengths#3 Call peaks...s# local lambda is disabled!sA#3 User prefers to use random sampling instead of linear scaling.s+#3 MACS is random sampling treatment %ss...sC#3 Your results may not be reproducible due to the random sampling!s#3 Random seed (%d) is used.s"#3 %d Tags from treatment are kepts)#3 MACS is random sampling control %ss...s#3 %d %ss from control are keptRRtfc_lows#4 Write output xls file... %stws,# This file is generated by MACS version %s s9# Sequencing ends will be shifted towards 3' by %d bp(s) s9# Sequencing ends will be shifted towards 5' by %d bp(s) s # d = %d s/# alternative fragment length(s) may be %s bps s# local lambda is disabled! tnametpscoretqscores-#4 Write peak in narrowPeak format file... %st name_prefixs%s_peak_t score_columnt tracklines#4 Write summits bed file... %st descriptions#Summits for %s (Made with MACS v2, s%xt)s2#4 Write broad peak in broadPeak format file... %ss9#4 Write broad peak in bed12/gappedPeak format file... %ssDone!N(RR(JRtinfotwarntdebugterrortargtxttformattPE_MODEttempdirttempfiletload_frag_files_optionstload_tag_files_optionstNoneRttsizettotaltkeepduplicatestcal_max_dup_tagstgsizetintt separate_dupstfloattnomodeltdtextsizetshiftt scanwindowRt MAX_PAIRNUMtmin_tagst scan_windowR tmaptstrt alternative_dtmodelRtmodel2r_scriptR"tonautoRRRtnolambdat downsampletseedt sample_numtFalset tocontrolttolargetTrueRt call_peakstpeakst filter_fctfecutofftpeakxlstopentwritet MACS_VERSIONt write_to_xlstcloset log_pvaluet log_qvaluetbroadtpeakNarrowPeaktwrite_to_narrowPeakR't summitbedtwrite_to_summit_bedRt peakBroadPeaktwrite_to_broadPeaktpeakGappedPeaktwrite_to_gappedPeak(targstoptionsR*R+R,R-RRRttagsinfott0ttreatment_max_dup_tagstt1tc0tcontrol_max_dup_tagstc1t peakmodelt peakdetecttofhd_xlsR&tofhd_bedt ofhd_summits((sP/oak/stanford/groups/akundaje/marinovg/programs/MACS-2.1.0/MACS2/callpeak_cmd.pytrun3st                                        #   #                            *     .  1 1 gñh㈵øä>cCstd||d|ƒS(sæCalculate the maximum duplicated tag number based on genome size, total tag number and a p-value based on binomial distribution. Brute force algorithm to calculate reverse CDF no more than MAX_LAMBDA(100000). igð?(R(t genome_sizet tags_numbertp((sP/oak/stanford/groups/akundaje/marinovg/programs/MACS-2.1.0/MACS2/callpeak_cmd.pyR9VscCs—|jdƒ|j|jdd|jƒ}|jƒ}t|jƒdkrx?|jdD]-}|j|d|jƒ}|j|ƒ}q[Wn|jƒ|j|_ |j rS|jdƒ|j|j dd|jƒ}|jƒ}|j}t|j ƒdkrFx?|j dD]-}|j|d|jƒ}|j|ƒ}qWn|jƒnd}|jd|j ƒ|dk r|jd|ƒn||fS( sVFrom the options, load treatment fragments and control fragments (if available). s#1 read treatment fragments...it buffer_sizeis#1.2 read input fragments...s;#1 mean fragment size is determined as %d bp from treatments@#1 note: mean fragment size in control is %d bp -- value ignoredN( R*tparserttfileR{t build_petrackR tappend_petracktfinalizeR?R6tcfileR5(RjttpRR}tcpRt control_dR((sP/oak/stanford/groups/akundaje/marinovg/programs/MACS-2.1.0/MACS2/callpeak_cmd.pyR3_s0          cCs}|jdƒ|j|jdd|jƒ}|jsM|jƒ}||_n|jƒ}t|jƒdkr°x?|jdD]-}|j|d|jƒ}|j|ƒ}q|Wn|jƒ|j rY|jdƒ|j|j dd|jƒjƒ}t|j ƒdkrLx?|j dD]-}|j|d|jƒ}|j|ƒ}qWn|jƒnd}|jd|jƒ||fS(sLFrom the options, load treatment tags and control tags (if available). s#1 read treatment tags...iR{is#1.2 read input tags...s##1 tag size is determined as %d bpsN( R*R|R}R{R6t build_fwtrackR tappend_fwtrackR€RR5(RjR‚tttsizeRR}RRRƒ((sP/oak/stanford/groups/akundaje/marinovg/programs/MACS-2.1.0/MACS2/callpeak_cmd.pyR4†s,        % (t__doc__tosRtloggingttimeRR2tMACS2.OptValidatorRtMACS2.OutputWritert MACS2.ProbRtMACS2.PeakModelRRtMACS2.PeakDetectRtMACS2.ConstantsRRwR9R3R4(((sP/oak/stanford/groups/akundaje/marinovg/programs/MACS-2.1.0/MACS2/callpeak_cmd.pyts"       ÿ$ '