U vId=@sddlmZddlmZmZddlmZddlmZm Z m Z m Z ddlm Z m Z mZddlmZddlZddlZddlmZddlZddlZddlZdd lmZd d lmZd d lmZd dl m!Z!d dl"m#Z#zd dl$m%Z%Wn0e&k rZ'ze'Z(ddZ%W5dZ'['XYnXd?eee e)fe e)e e*e)edddZ+d@eee)e,fe)edddZ-dAeedddZ.ee)eddd Z/e e)e e)e e)fe ej0e e)ej1ffd!d"d#Z2ed$d%d&d'd(d)dd*ddeieid+ ee*e*e)e)e e e)e e)fe)e e e)e e)fe)e e)e e)fe e)e e)fed, d-d.Z3dBee)ed/d0d1Z4dCeee e)fe e)e e*e)edd2d3Z5e e)ee)ddfd4d5d6Z6e e)e e)e e*e)ed7d8d9Z7dDd;d<Z8d=d>Z9dS)E)Path)PathLikefspath)MappingProxyType)UnionOptionalMappingTuple)IterableIterator Generator) OrderedDictN)warn)sparse)AnnData)_deprecate_positional_args)is_float) read_h5ad) read_zarrcOstdS)N)e)___rI/home/sam/Atlas/atlas_env/lib/python3.8/site-packages/anndata/_io/read.pyrsr,float32)filename delimiterfirst_column_namesdtypereturncCst||||S)a Read `.csv` file. Same as :func:`~anndata.read_text` but with default delimiter `','`. Parameters ---------- filename Data file. delimiter Delimiter that separates data within text file. If `None`, will split at arbitrary number of white spaces, which is different from enforcing splitting at single white space `' '`. first_column_names Assume the first column stores row names. dtype Numpy data type. ) read_text)rrr r!rrrread_csvsr$)rsheetr!r"cCszddlm}|t||}|jddddf}t|jdddfjtd}ttj |j ddtdd}t |||S)z Read `.xlsx` (Excel) file. Assumes that the first columns stores the row names and the first row the column names. Parameters ---------- filename File name to read from. sheet Name of sheet in Excel file. r) read_excelNr) row_namesr!) col_names) pandasr&rvaluesdictZilocastypestrnparraycolumnsr)rr%r!r&ZdfXrowcolrrrr&9s   r&)rr"cCstj|dddd}tj|d|djj|djjff|d}tjtj|djjddd}tjtj|djjddd}t |||d S) z Read a gzipped condensed count matrix from umi_tools. Parameters ---------- filename File name to read from. category)genecellr(countr7r6)name)index)r2obsvar) pdZ read_tabler csr_matrixcatcodes DataFrameIndex categoriesr)rr!tabler2r;r<rrrread_umi_toolsSs rE)rkeyr"c Cst|d|}dd|D}|dkr>td|d|d||d}iig}td d gD]$\}}||kr^||d|||<q^W5QRXt||d |d }|S) z Read `.h5` (hdf5) file. Note: Also looks for fields `row_names` and `col_names`. Parameters ---------- filename Filename of data file. key Name of dataset in the file. rcSsg|]}|qSrr).0krrr zszread_hdf..z The file z stores the following sheets: z& Call read/read_hdf5 with one of them.rr'r)rr)h5pyFilekeys ValueError enumerater) rrFfrNr2Z rows_colsinamer9adatarrrread_hdfjs   rT)inputidx_name dimm_mappingr"c st}i}|D]&\}}tfdd|Dj||<qD]2\}}|jdkrn|jddkrn|||<qD|||<qD||kr|j|ddd||fS)Ncsg|]}|qSr)pop)rHr9rUrrrJsz(_fmt_loom_axis_attrs..rT)ZdropZinplace) r=rAitemsr/r0TndimshapeZ set_index) rUrVrWZaxis_dfZ axis_mappingrFnamesrIvrrYr_fmt_loom_axis_attrss   r`z0.9)versionTFZsplicedZCellIDZGene) rcleanupX_name obs_names obsm_names var_names varm_namesr! obsm_mapping varm_mapping) rrrbrcrdrerfrgr!rhrir"c  KsP|dk r&tdt| ikr"td|} |dk rLtdt| ikrHtd|} t|}ddlm} | |df| } || jkrd }|r| j|j n| j|d j }|j |d d }t }|d kr|r| jd j n| jd d j |d <| jD]:}|d kr|r$| j|j n| j|d j ||<qt t| j|| \}}t t| j|| \}}i}|ri}t|D]4}tt||dkr||d||<||=q|r||d<i}t|D]4}tt||dkr||d||<||=q|r||d<t|||||r,|nd|r8|nd|d}W5QRX|S)a Read `.loom`-formatted hdf5 file. This reads the whole file into memory. Beware that you have to explicitly state when you want to read the file as sparse data. Parameters ---------- filename The filename. sparse Whether to read the data matrix as sparse. cleanup Whether to collapse all obs/var fields that only store one unique value into `.uns['loom-.']`. X_name Loompy key with which the data matrix :attr:`~anndata.AnnData.X` is initialized. obs_names Loompy key where the observation/cell names are stored. obsm_mapping Loompy keys which will be constructed into observation matrices var_names Loompy key where the variable/gene names are stored. varm_mapping Loompy keys which will be constructed into variable matrices **kwargs: Arguments to loompy.connect Example ------- .. code:: python pbmc = anndata.read_loom( "pbmc.loom", sparse=True, X_name="lognorm", obs_names="cell_names", var_names="gene_names", obsm_mapping={ "X_umap": ["umap_1", "umap_2"] } ) NzbArgument obsm_names has been deprecated in favour of `obsm_mapping`. In 0.9 this will be an error.zfReceived values for both `obsm_names` and `obsm_mapping`. This is ambiguous, only pass `obsm_mapping`.zbArgument varm_names has been deprecated in favour of `varm_mapping`. In 0.9 this will be an error.zfReceived values for both `varm_names` and `varm_mapping`. This is ambiguous, only pass `varm_mapping`.r)connectrGrKrF)copymatrixrzloom-obszloom-var)r;r<layersobsmvarmuns)r FutureWarningrOrZloompyrjrmrNrr[Ztocsrr-r r`r,Z col_attrsZ row_attrslistlensetr)rrrbrcrdrerfrgr!rhrikwargsrjlcr2rmrFr;rnr<rorpZuns_obsZuns_varrSrrr read_looms? (&     rw)rr!r"cCs:ddlm}|t||}ddlm}||}t|S)z Read `.mtx` file. Parameters ---------- filename The filename. dtype Numpy data type. r)mmreadr>)Zscipy.iorxrr-Z scipy.sparser>r)rr!rxr2r>rrrread_mtx.s  rzc Cst|tttfst||||St|}|jdkrftjt|dd}t||||W5QRSQRXnj|jdkrt jt|dd}t||||W5QRSQRXn*|}t||||W5QRSQRXdS)a Read `.txt`, `.tab`, `.data` (text) file. Same as :func:`~anndata.read_csv` but with default delimiter `None`. Parameters ---------- filename Data file, filename or stream. delimiter Delimiter that separates data within text file. If `None`, will split at arbitrary number of white spaces, which is different from enforcing splitting at single white space `' '`. first_column_names Assume the first column stores row names. dtype Numpy data type. z.gzrt)modez.bz2N) isinstancerr.bytes _read_textrsuffixgzipopenbz2)rrr r!rQrrrr#Cs " " r#) file_liker"ccs"|D]}|d}|r|VqdS)zdel_sparse_matrix_keys)drFr>key_csrrrrload_sparse_csrs  &  rcCs4||d=||d=||d=||d=dS)Nrrrrr)mappingrrrrrs   r)rNr)r)N)r)NNr)r2):pathlibrosrrtypesrtypingrrrr r r r collectionsr rrwarningsrrLnumpyr/r*r=ZscipyrrKrcompatrutilsrZh5adrZzarrr ImportError_err.boolr$rr&rErTrAZndarrayr`rwrzr#rrrrrrrrs            "  '  k