o Nrft/@s|ddlmZddlZddlZddlZddlZddlmZddl Z ddl Z ddl Z e eZ d ddZ    d dd ZdS) )print_functionNc Csg}|D]} t| ddkr|t| dg7}q|t| dg7}qtj} |} g} d} d}|r5|dd}td|tt |D]\}}|d dsT|d d rUqB|||\}}}}t |t ||kr~t d ||tjgt ||t |7}||dkrt d |td|dur||krt| ||||t | <d} d}g} |}||}| |kr| | gt |g7} | d7} | |ks | |krd |||}t|t||}| |kr| |g7} | d7} | |kst | | kr7zt| ||||t | <Wn$ty,}ztd |tjdtdtjdWYd}~dSd}~wwg} | }td||qBt| ||||t | <dS)z9 Convert an epilogos bedfile to multivec format. z.gzrtrrNzbase_resolution:Zbrowsertrackz-Lines contain fewer columns than expected: %szDThe start coordinate is not a multiple of the resolution in line: %sz The expected position location does not match the observed location at entry {}:{}-{} This is probably because the bedfile is not sorted. Please sort and try again. zError:filez-Probably need to set the --num-rows parameterzdumping batch:)opsplitextgzipopennpnanreadlineprint enumeratezip startswithlenloggerwarningerrorsysexitarrayformat ValueErrormathceil TypeErrorstderr)Zinput_filenamesZf_outZ!bedline_to_chrom_start_end_vectorZbase_resolution has_header chunk_sizeZnum_rows row_infosfilesZinput_filenameZ FILL_VALUEZ batch_lengthbatchZ curr_indexZbatch_start_indexZ prev_chrom_lineschromstartendZvectorZdata_start_indexmessageZdata_end_indexexr-Y/var/www/html/software/conda/envs/catlas/lib/python3.10/site-packages/clodius/multivec.pybedfile_to_multivecs        "r/r/tmp/my_file.multiresc Cs|}t|r t|t|d}|d||djd<|d|d|} |dt| t |\} } t j | dd} d|jvrJ|jd}|d ur[|dt| j d||dt| d|dt| d |dt| dj d t| f| j| d d |dt| dj dt| f| d d|dj d t| f| j| d d |dj dt| f| d dt | | D]n\} }| |vrtd| tjdq|dt| d j t| || jd dd}d}|dt| d | }tt|t|}|t|kr,|| |||||||<|tt|t|| |7}|t|ksqt| }tt|||td}| }t|D] }|d} |dt| |d urj|dt| j d||dt| d|dt| d |dt| dj d t| f| j| d d |dt| dj dt| f| d dt | | D]\} }| |dt|d vrqd}|dt|d | }d}tt|t|}t|j}t|dd|d<t|}|dt| d j | |d d|t|krd|dt|d | |||}t|ddkr/t ||dgf}|d7}||} ||dt| d | t|dt|d|d<|tt|t||7}|t|ksq| }qH|S)a Create a multires file containing the array data aggregated at multiple resolutions. Parameters ---------- array_data: {'chrom_key': np.array, } The array data to aggregate organized by chromosome chromsizes: [('chrom_key', size),...] agg: lambda The function that will aggregate the data. Should take an array as input and create another array of roughly half the length starting_resolution: int (default 1) The starting resolution of the input data tile_size: int The tile size that we want higlass to use. This should depend on the size of the data but when in doubt, just use 256. winfoz tile-size resolutionschromsS)dtyper#Nvaluesnamer )shaper7data compressionlength)r:r;r<zMissing chrom {} in input filer)r<gj@rr)r existsosremoveh5pyFileZ create_groupattrsstrrr rcreateZcreate_datasetrr7rrrr r:intminsumrrlograngelisttupleZ concatenate)Z array_dataZ chromsizesZaggZstarting_resolutionZ tile_sizeZ output_filer#filenamefZcurr_resolutionr5lengthsZ chrom_arrayr(r=Zstandard_chunk_sizer)Z chrom_datar" total_lengthZmax_zoomZprev_resolutioniZ new_shapeZold_dataZnew_datar-r-r.create_multivec_multiress               #rT)N)rr0r1N) __future__rr loggingrrAZos.pathpathr rrCnumpyr getLogger__name__rr/rTr-r-r-r.s$