U mdH7@sNddlmZmZmZmZddlmZddlZddlZ ddl Z ddl m Z mZmZddl mZmZddlmZddlmZdd lmZmZmZmZmZmZd d lmZd2d dZeeeeeeddddddd dd ddd ee e ee eee!ee e"ee e"ee j#d ddZ$eeeeeddddd d dddee e ee e"ee j#dddZ%eeeeeeeddddddd d ddd ee e ee eee!ee e"e"e"ee"eee j#e j#fd d!d"Z&ee j'efe!d#d$d%Z(d&d'Z)d(d)Z*ee j'efee!e j+d*d+d,Z,ee j'efee!e j+d*d-d.Z-ej.ddd/d0d1Z/dS)3)OptionalTuple CollectionUnion)warnN)issparseisspmatrix_csrisspmatrix_coo)spmatrix csr_matrix)mean_variance_axis)AnnData) doc_expr_repsdoc_obs_qc_argsdoc_qc_metric_namingdoc_obs_qc_returnsdoc_var_qc_returnsdoc_adata_basic) _doc_paramsFcCsJ|dk }|r&|r&td|d|d|r4|j|S|r@|jjS|jSdS)NzECannot use expression from both layer and raw. You provided:'use_raw=z ' and 'layer=') ValueErrorZlayersrawX)adatause_rawlayerZis_layerrQ/home/sam/Atlas/atlas_env/lib/python3.8/site-packages/scanpy/preprocessing/_qc.py_choose_mtx_reps r )rrrrrcountsZgenesr)2diT) expr_typevar_typeqc_vars percent_toprrlog1pinplacerparallel) rr%r&r'r(rrr)r*returnc  Cs | dk rtdt| dkrFt|||} t| r6t| } t| rF| tj|j d} t| rz| j dd| d|d|<nt j | dd| d|d|<|rt | d|d|| d|d|<t | jdd| d |<|rt | d || d |<|r\t|}t| |} t|D]4\} }| dd| fd | d |d |d|<q&|D]}t | dd|j|jfjdd| d |d|<|rt | d |d|| d |d|<| d |d|| d |d | d |d|<q`|r| |j| j<n| SdS)a Describe observations of anndata. Calculates a number of qc metrics for observations in AnnData object. See section `Returns` for a description of those metrics. Note that this method can take a while to compile on the first call. That result is then cached to disk to be used later. Params ------ {doc_adata_basic} {doc_qc_metric_naming} {doc_obs_qc_args} {doc_expr_reps} log1p Add `log1p` transformed metrics. inplace Whether to place calculated metrics in `adata.obs`. X Matrix to calculate values on. Meant for internal usage. Returns ------- QC metrics for observations in adata. If inplace, values are placed into the AnnData's `.obs` dataframe. {doc_obs_qc_returns} N?Argument `parallel` is deprecated, and currently has no effect.indexrZaxisZn_Z_by_Zlog1p_n_Ztotal_Z log1p_total_r#Zpct_Z_in_top__)r FutureWarningr r r reliminate_zerospd DataFrameZ obs_namesgetnnznp count_nonzeror)ravelsumsortedtop_segment_proportions enumeratevarvaluesZobscolumns)rr%r&r'r(rrr)r*rr+ obs_metricsZ proportionsinZqc_varrrr describe_obs&s^2    rD)rrrr)r%r&rrr*r)r)rr%r&rrr,c Cs8|dkr4t|||}t|r$t|}t|r4|tj|jd}t|rp|jdd|d<t |ddd|d<n"t j |dd|d<|j dd|d<|rt |d|d<d|d|jdd |d <t |jdd|d <|rt |d |d <g} |jD]} | | jftq| |_|r0||j|j<n|SdS) a9 Describe variables of anndata. Calculates a number of qc metrics for variables in AnnData object. See section `Returns` for a description of those metrics. Params ------ {doc_adata_basic} {doc_qc_metric_naming} {doc_expr_reps} inplace Whether to place calculated metrics in `adata.var`. X Matrix to calculate values on. Meant for internal usage. Returns ------- QC metrics for variables in adata. If inplace, values are placed into the AnnData's `.var` dataframe. {doc_var_qc_returns} Nr.rr0zn_cells_by_{expr_type}zmean_{expr_type}zlog1p_mean_{expr_type}rr#zpct_dropout_by_{expr_type}ztotal_{expr_type}zlog1p_total_{expr_type})r r r rr3r4r5Z var_namesr6r r7r8Zmeanr)shaper9r:r@appendformatlocalsr>) rr%r&rrr*r)r var_metricsZ new_colnamescolrrr describe_vars>)  rK)rrrrrr) r%r&r'r(rrr*r)r+) rr%r&r'r(rrr*r)r+r,c Csz| dk rtdtt|||} t| r.t| } t| r>| t||||||| |d} t||||| |d} |sv| | fSdS)a Calculate quality control metrics. Calculates a number of qc metrics for an AnnData object, see section `Returns` for specifics. Largely based on `calculateQCMetrics` from scater [McCarthy17]_. Currently is most efficient on a sparse CSR or dense matrix. Note that this method can take a while to compile on the first call. That result is then cached to disk to be used later. Parameters ---------- {doc_adata_basic} {doc_qc_metric_naming} {doc_obs_qc_args} {doc_expr_reps} inplace Whether to place calculated metrics in `adata`'s `.obs` and `.var`. log1p Set to `False` to skip computing `log1p` transformed annotations. Returns ------- Depending on `inplace` returns calculated metrics (as :class:`~pandas.DataFrame`) or updates `adata`'s `obs` and `var`. {doc_obs_qc_returns} {doc_var_qc_returns} Example ------- Calculate qc metrics for visualization. .. plot:: :context: close-figs import scanpy as sc import seaborn as sns pbmc = sc.datasets.pbmc3k() pbmc.var["mito"] = pbmc.var_names.str.startswith("MT-") sc.pp.calculate_qc_metrics(pbmc, qc_vars=["mito"], inplace=True) sns.jointplot( data=pbmc.obs, x="log1p_total_counts", y="log1p_n_genes_by_counts", kind="hex", ) .. plot:: :context: close-figs sns.histplot(pbmc.obs["pct_counts_mito"]) Nr-)r%r&r'r(r*rr))r%r&r*rr)) rr2r r r rr3rDrK) rr%r&r'r(rrr*r)r+rrArIrrrcalculate_qc_metricss<L   rLmtxrCcCs<t|r.t|st|}t|j|jt|St||SdS)a^ Calculates cumulative proportions of top expressed genes Parameters ---------- mtx Matrix, where each row is a sample, each column a feature. n Rank to calculate proportions up to. Value is treated as 1-indexed, `n=50` will calculate cumulative proportions up to the 50th most expressed gene. N) rrr top_proportions_sparse_csrdataindptrr7arraytop_proportions_denserMrrrtop_proportionsIs rTcCs|jdd}ttjd| |d}|ddd|f}tj|tjd}t|jdD]P}||||ddff}|dddt |||}|||ddf<qV|S)Nrr0dtyper) r:r7apply_along_axisZ argpartitionZ zeros_likefloat64rangerEsortcumsum)rNrCsums partitionedr?rBvecrrrrS_s rSc Cstj|jd|ftjd}t|jdD]}||||d}}tj|tjd}|||kr||||d||<|}nt|}t|j|j t |St ||SdS)as Calculates total percentage of counts in top ns genes. Parameters ---------- mtx Matrix, where each row is a sample, each column a feature. ns Positions to calculate cumulative proportion at. Values are considered 1-indexed, e.g. `ns=[50]` will calculate cumulative proportion up to the 50th most expressed gene. rrz$Positions outside range of features.N) maxrEmin IndexErrorrrr "top_segment_proportions_sparse_csrrPrQr7rRtop_segment_proportions_dense)rNrhrrrr<|sr<c Cst|}|jdd}ttjd||jd|dddddfddd|df}t|jdt|f}t|jd}d}t|D]<\}}||dd||fjdd7}||dd|f<|}q||dddfS)Nrr0rWr) r7r[r:rXrdrEr`lenr=) rNrhr]r^r?accprevjrCrrrrms  4 rm)cacher+c Cs|tj}|tj}t|}|d}tj|jd|jd}tj|jdt|ftjd}tj|jd|f|jd}t |jdD]}||||d}} t ||| ||<| ||kr||| ||d| |f<n:| ||krt |||  |d| ||ddf<t ||ddf||||ddf<q|dddddfddd|df}tj|jd|jd} d} t |jD]B} | |dd| || fj dd7} | |dd| f<|| } q|||jddfS)NrWrrUrr0)Zastyper7Zint64r[r`rarVrnrYrbrcr:rdrZZreshape) rPrQrhZmaxidxr]r?r^rBrerfrorprqrrrrls.    ,**$ rl)FN)0typingrrrrwarningsrrbnumpyr7Zpandasr4Z scipy.sparserrr r r Zsklearn.utils.sparsefuncsr Zanndatar Z_docsrrrrrr_utilsrr strintboolr5rDrKrLrRrTrSrOZndarrayr<rmZnjitrlrrrrs        _ H  g