ó ?îec@stddlZddlZddlmZd„Zd„Zd„Zd„Zd„Zd„Z d „Z d „Z dS( iÿÿÿÿN(t __version__cCsD|j}tƒ}|dkr@ddlm}dGH||ƒn|dkrnddlm}dGH||ƒnÒ|dkrœdd lm}d GH||ƒn¤|d krÊdd lm }d GH||ƒnv|dkrøddl m }dGH||ƒnH|dkr@|j t tdd|jd|jd|jd|jd|jd|jg ƒƒ}|j dd|jdd|jgƒ} dd|jd|jd|jdt|jƒd|jdd|jdd|jd|jd|jdg} |jr| jdgƒn|j | ƒ} |j d d |jd!d"|jd#d|jgƒ} |j dd|jd|jdd$|jd%d|jd|jd|jd|jgƒ} ddlm}ddlm}dd lm}dd lm }ddl m }d&GH||ƒd'GH|| ƒd(GH|| ƒd)GH|| ƒd*GH|| ƒnd+S(,s.The Main function for calling nucleoatac tocciÿÿÿÿ(trun_occsP---------Computing Occupancy and Nucleosomal Insert Distribution----------------tvprocess(t run_vprocesssP---------Processing VPlot-------------------------------------------------------tnuc(trun_nucsP---------Obtaining nucleosome signal and calling positions----------------------tmerge(t run_mergesP---------Merging----------------------------------------------------------------tnfr(trun_nfrsP---------Calling NFR positions--------------------------------------------------truns--beds--bams--fastas--pwms--outs--coress--sizess .nuc_dist.txts --occ_tracks.occ.bedgraph.gzs--vmats.VMats.fragmentsizes.txts --write_alls --occpeakss.occpeaks.bed.gzs--nucposs.nucpos.bed.gzs--callss.nucmap_combined.bed.gzsP---------Step1: Computing Occupancy and Nucleosomal Insert Distribution---------sP---------Step2: Processing Vplot------------------------------------------------sP---------Step3: Obtaining nucleosome signal and calling positions---------------sP---------Step4: Making combined nucleosome position map ------------------------sP---------Step5: Calling NFR positions-------------------------------------------N(tcalltnucleoatac_parsertnucleoatac.run_occRtnucleoatac.run_vprocessRtnucleoatac.run_nucRtnucleoatac.mergeRtnucleoatac.run_nfrR t parse_argstmaptstrtbedtbamtfastatpwmtouttcorest write_alltextend(targsR tparserRRRRR tocc_argst vprocess_argst nuc_args_listtnuc_argst merge_argstnfr_args((s4/tmp/pip-install-bGcd2k/NucleoATAC/nucleoatac/cli.pytnucleoatac_mainsj             !(-% #"    cCs‡tjddddƒ}|jdddddtƒ|jd d ƒ}t|ƒt|ƒt|ƒt|ƒt |ƒt |ƒ|S( sPrepares argparse object t descriptions+%(prog)s -- Nucleosome Calling for ATAC-SeqtepilogsCFor command line options for each command, type %(prog)s COMMAND -hs --versiontactiontversions %(prog)s tdestR ( targparsetArgumentParsert add_argumentRtadd_subparserstadd_run_parsertadd_occ_parsertadd_vprocess_parsertadd_nuc_parsertadd_merge_parsertadd_nfr_parser(t argparsert subparsers((s4/tmp/pip-install-bGcd2k/NucleoATAC/nucleoatac/cli.pyR Es       c Cs|jdddƒ}|jddƒ}|jddddd d tƒ|jd dd dd d tƒ|jdddddd tƒ|jddƒ}|jdddddƒ|jdddddddƒ|jddƒ}|jdddddƒ|jd dd!dd"dd#d$tƒ|jd%d&ƒ}|jd'dd!dd(dd)d$tƒ|jd*dd!dd+dd,d$tƒ|jd-dd.dd/d$tdd0ƒ|jd1dd!dd2d$tdd3ƒ|jd4dd.dd5d$tdd6ƒ|jd7dd!dd8d$tdd9ƒd:S(;s.Add argument parsers for the occ utility Rthelps/nucleoatac function: Call nucleosome occupancytRequiredsNecessary argumentss--bedtmetavartbed_filesPeaks in bed formattrequireds--bamtbam_filesSorted (and indexed) BAM files--outtbasenamesgive output basenamesBias calculation informationsEHighly recommended. If fasta is not provided, will not calculate biass--fastat genome_seqsIndexed fasta files--pwmtTn5_PWMsAPWM descriptor file. Default is Human.PWM.txt included in packagetdefaulttHumansGeneral Optionsts--sizestfragmentsizes_filesUFile with fragment size distribution. Use if don't want calculation of fragment sizes--corestintisNumber of cores to usettypesOccupancy parametersChange with cautions--upperiûs*upper limit in insert size. default is 251s--flanki<sRDistance on each side of dyad to include for local occ calculation. Default is 60.s --min_occtfloatgš™™™™™¹?sHOccupancy cutoff for determining nucleosome distribution. Default is 0.1s --nuc_sepixs:minimum separation between occupany peaks. Default is 120.s--confidence_intervalgÍÌÌÌÌÌì?s`confidence interval level for lower and upper bounds. default is 0.9, should be between 0 and 1s--stepisastep size along genome for comuting occ. Default is 5. Should be odd, or will be subtracted by 1N(t add_parsertadd_argument_groupR.tTrueRERG(R7Rtgroup1tgroup4tgroup2tgroup3((s4/tmp/pip-install-bGcd2k/NucleoATAC/nucleoatac/cli.pyR1\s<     cCsÒ|jdddƒ}|jddƒ}|jdddd tdd ƒ|jd dd d tdd ƒ|jddƒ}|jdddddƒ|jdddddddƒ|jdddddddƒdS(s%Add argument parser for merge utilityRR8s,nucleoatac function: Merge occ and nuc callsR9sNecessary argumentss --occpeaksR:t occpeaks_fileR<sOutput from occ utilitys--nucpost nucpos_filesOutput from nuc utilitytOptionstoptionals--outt out_basenamesoutput file basenames--septmin_separationRAixsminimum separation between calls --min_occtmin_occgš™™™™™¹?s_minimum lower bound occupancy of nucleosomes to be considered for excluding NFR. default is 0.1N(RHRIR.RJ(R7RRKRM((s4/tmp/pip-install-bGcd2k/NucleoATAC/nucleoatac/cli.pyR4sc Cs¿|jdddƒ}|jddƒ}|jddddd d tƒ|jd dd d tdd ƒ|jdddd tddƒ|jddƒ}|jdddddƒ|jdddddƒ|jddƒ}|jdddddƒ|jddddd d!d"ƒ|jd#d$ƒ}|jd%dd&dd'ƒ|jd(dd)d!d*dd+d,tƒ|jd-ƒ}|jd.dd/d!d0dd1d,tƒ|jd2dd/d!d3dd4d,tƒd5S(6s)Add argument parser for nfr utility R R8snucleoatac function: Call NFRsR9sNecessary argumentss--bedR:R;sPeaks in bed formatR<s --occ_tracktocc_filesDbgzip compressed, tabix-indexed bedgraph file with occcupancy track.s--callsRPs%bed file with nucleosome center callssInsertion track optionss'Either input insertion track or bamfiles --ins_tracktins_filesebgzip compressed, tabix-indexed bedgraph file with insertion track. will be generated if not includeds--bamR=sSorted (and indexed) BAM filesBias calculation informationsEHighly recommended. If fasta is not provided, will not calculate biass--fastaR?sIndexed fasta files--pwmR@sAPWM descriptor file. Default is Human.PWM.txt included in packageRARBsGeneral optionsRRs--outRSsoutput file basenames--corest num_coresisNumber of cores to useRFsNFR determination parameterss --max_occRGgš™™™™™¹?s.Maximum mean occupancy for NFR. Default is 0.1s--max_occ_uppergÐ?sEMaximum for minimum of upper bound occupancy in NFR. Default is 0.25N(RHRIR.RJRERG(R7RRKtgroup6RLRMtgroup5((s4/tmp/pip-install-bGcd2k/NucleoATAC/nucleoatac/cli.pyR5Žs2  c Csk|jdddƒ}|jddƒ}|jdddd tƒ|jd d ƒ}|jd dd ddƒ|jddddddtjjƒ|jddƒ}|jddddddddtƒ|jddddddddtƒ|jddddddddtƒ|jd d!ƒ}|jd"dd#dd$dtdd%ƒ|jd&d'd(dt dd)ƒd*S(+s3Add argument parsers for the vprocess utility RR8sHnucleoatac function: Make processed vplot to use for nucleosome callingR9sNecessary argumentss--outR:toutput_basenameR<sVPlot and Insert Size OptionstOptionals--sizestfilesInsert distribution files--vplott vmat_files1Accepts VMat file. Default is Vplot from S. Cer.RAsSize parameerssUse sensible valuess--lowerREiis6lower limit (inclusive) in insert size. default is 105RFs--upperiûs3upper limit (exclusive) in insert size. default 251s--flanki<s(distance on each side of dyad to includeRQRCs--smoothRGgè?s:SD to use for gaussian smoothing. Use 0 for no smoothing.s --plot_extraR)t store_truesMake some additional plotsN( RHRIR.RJt nucleoatactMagict default_vplotRERGtFalse(R7RRKRMRNRL((s4/tmp/pip-install-bGcd2k/NucleoATAC/nucleoatac/cli.pyR2­s*    c Csi|jdddƒ}|jddƒ}|jddddd d tƒ|jd dd dd d tƒ|jdddddd tƒ|jdddddd tƒ|jddƒ}|jdddddƒ|jdddddddƒ|jddƒ}|jd dd!dd"ƒ|jd#dd$dd%ƒ|jd&dd'dd(dd)d*tƒ|jd+d,d-dtdd.ƒ|jd/d0d1d,d2dtdd3ƒ|jd4d5ƒ}|jd6dd7dd8dd9d*tƒ|jd:dd7dd;dd<d*tƒ|jd=dd>dd?d*tdd@ƒ|jdAdd>ddBd*tddCƒ|jdDdd>d*tddEddFƒdGS(Hs.Add argument parsers for the nuc utility RR8sFnucleoatac function: Call nucleosome positions and make signal tracksR9sNecessary argumentss--bedR:R;s&Regions for which to do stuff.R<s--vmatt vdensity_files VMat objects--bamR=sAccepts sorted BAM files--outR>sgive output basenames Bias optionss,If --fasta not provided, bias not calculateds--fastaR?sIndexed fasta files--pwmR@sAPWM descriptor file. Default is Human.PWM.txt included in packageRARBsGeneral optionsRCs--sizesRDsUFile with fragment size distribution. Use if don't want calculation of fragment sizes --occ_trackRVstbgzip compressed, tabix-indexed bedgraph file with occcupancy track. Otherwise occ not determined for nuc positions.s--coresRXisNumber of cores to useRFs --write_allR)R_swrite all trackss --not_atacR+tatact store_falsesdata is not atac-seqsNucleosome calling parameterssChange with cautions--min_zRGis4Z-score threshold for nucleosome calls. Default is 3s--min_lrisALog likelihood ratio threshold for nucleosome calls. Default is 0s --nuc_sepREixsDMinimum separation between non-redundant nucleosomes. Default is 120s--redundant_sepis_Minimum separation between redundant nucleosomes. Not recommended to be below 15. Default is 25s--sdi s´Standard deviation for smoothing. Affect the resolution at which nucleosomes can be positioned. Not recommended to exceed 25 or to be smaller than 10. Default is 10N(RHRIR.RJRERcRG(R7RRKRMRNRZ((s4/tmp/pip-install-bGcd2k/NucleoATAC/nucleoatac/cli.pyR3ÇsJ      c Cs.|jdddƒ}|jddƒ}|jddddd d tƒ|jd dd dd d tƒ|jdddddd tƒ|jdddddd tƒ|jddƒ}|jdddddddƒ|jddƒ}|jdddddddd tƒ|jd!d"d#dtdd$ƒd%S(&s.Add argument parsers for the run utility R R8sVMain nucleoatac utility-- runs through occupancy determination & calling nuc positionsR9sNecessary argumentss--bedR:R;s&Regions for which to do stuff.R<s--bamR=sAccepts sorted BAM files--outR[sgive output basenames--fastaR?sIndexed fasta filesBias calculation parametersRCs--pwmR@sAPWM descriptor file. Default is Human.PWM.txt included in packageRARBsGeneral optionss--coresRXisNumber of cores to useRFs --write_allR)R_swrite all tracksN(RHRIR.RJRERc(R7RRKRLRN((s4/tmp/pip-install-bGcd2k/NucleoATAC/nucleoatac/cli.pyR0ós$     ( R,tnucleoatac.MagicR`RR&R R1R4R5R2R3R0(((s4/tmp/pip-install-bGcd2k/NucleoATAC/nucleoatac/cli.pyts   >  #    ,