o ؾg@sddlZddlmZddlZddlmZddlm Z m Z m Z ddl m Z mZmZmZddlmZdyd d Zd d Zd dZddZddZdzddZ   d{d d!Z  d|d"d#Z % d}d&d'Z  d~d(d)Zd*d+Z  dd.d/Zd0d1Zd2d3Z 6 7 dd8d9Z!  ddd?Z#dd@dAZ$ $ ddFdGZ% E ddJdKZ& L ddMdNZ' O ddPdQZ( S ddTdUZ)dVdWZ*dddddddXdYdZgdd$d[d\d]d%dddfd^d_Z+d`daZ,dbdcZ-dddeZ.dfdgZ/dhdiZ0djdkZ1dldmZ2dndoZ3 L r ddwdxZ4dS)N) csr_matrix)AnnData)extract rename_net filt_min_n)get_filterbyexpr_inputsget_min_sample_sizeget_cpm_cutoffget_cpm)compute_es_per_rankFcCsT|sz ddlm}W|Stytdwzddl}W|Sty)tdw)NrzKmatplotlib is not installed. Please install it with: pip install matplotlib)matplotlib.pyplotpyplot Exception ImportError matplotlib) return_mplpltmplr\/var/www/html/backend_erp/backend_erp_env/lib/python3.10/site-packages/decoupler/plotting.pycheck_if_matplotlib s   rcC&zddl}W|Stytdw)NrzEseaborn is not installed. Please install it with: pip install seaborn)seabornrr)snsrrrcheck_if_seaborn   rcCr)NrzKadjustText is not installed. Please install it with: pip install adjustText) adjustTextrr)atrrrcheck_if_adjustText"rrcCsNzddl}Wn tytdwddlm}||j|dkr%td|S)NrzCigraph is not installed. Please install it with: pip install igraph)Versionz0.10.0z^igraph version needs to be at least 0.10.0. Please install it with: pip install igraph==0.10.0)igraphrrpackaging.versionr __version__)igr rrrcheck_if_igraph*s   r%cCs>|dur|dur|dur|j|dddStdtddS)Ntight) bbox_inchesz fig is None, cannot save figure.zax is None, cannot save figure.)savefig ValueError)figaxsaverrr save_plot5sr-cCsX|durtj}|durtj}|dt|k}t|dt|k}|j||@}|S)NpvalslogFCs)npinfabsloc)df sign_limit lFCs_limitmsk_signmsk_lFCsrrr filter_limits@sr9sourcetargetweight皙??#D62728#1F77B4grayr:dc# Cs6t}t}|j|j}|j|j}|j||f}|j||f}t|  } d}|dur9|jdd||d\}}|durt ||||d}tj t |dddd}t |ddk}|durdt d ||d |kd }|j|gj|d <t|j|gj |d <|tj t|dd}t|| | d}|r| }nd}|jjd d dd||d|d|d||ni|j|gjj|d idd}t|j|gj |d <|tj t|dd}t|| | d}||d<| |j|d | k|d | k@df<||j|d | k|d | k@df<|jjd d dd|d|d||j| ddd|j| ddd|j| ddd|t |d | k|d | k@jd dd}|jd|}|dg}t|d |d |jD]\} }!}"||| |!|"qjt |dkr|j!|t"ddd|dt#||||r|SdS)a Plot logFC and p-values. If name and net are provided, it does the same for the targets of a selected source. Parameters ---------- logFCs : DataFrame Data-frame of logFCs (contrasts x features). pvals : DataFrame Data-frame of p-values (contrasts x features). contrast : str Name of the contrast (row) to plot. name : str, None Name of the source to plot. If None, plot classic volcano (without subsetting targets). net : DataFrame, None Network dataframe. If None, plot classic volcano (without subsetting targets). top : int Number of top differentially expressed features. source : str Column name in net with source nodes. target : str Column name in net with target nodes. weight : str, None Column name in net with weights. If none, set to None. sign_thr : float Significance threshold for p-values. lFCs_thr : float Significance threshold for logFCs. sign_limit : float Limit of p-values to plot in -log10. lFCs_limit : float Limit of logFCs to plot in absolute value. color_pos: str Color to plot significant positive genes. color_neg: str Color to plot significant negative genes. color_null: str Color to plot rest of the genes. figsize : tuple Figure size. dpi : int DPI resolution of figure. ax : Axes, None A matplotlib axes object. If None returns new figure. return_fig : bool Whether to return a Figure object or not. save : str, None Path to where to save the plot. Infer the filetype if ending on {``.pdf``, ``.png``, ``.svg``}. Returns ------- fig : Figure, None If return_fig, returns Figure object. Nrfigsizedpir;r<r=r=)ddofrz%If net is given, name cannot be None.r;r<r/r.axisr5r6coolwarmF)xyccmapvminvmaxsharexr+z {0} | {1}rPrQrRrVr+z{0}--blackrQ linestylecolorrPr[r\ ascending -log10(pvals)- arrowstyler\ arrowpropsr+)$rrindex intersectioncolumnsr3r0log10subplotsrstdr2anyr) set_indexTpdisnullr9plotscatter set_titleformatrenameaxhlineaxvline sort_valuesiloc set_ylabelzipappendtextlen adjust_textdictr-)#r/r.contrastnamenettopr;r<r=sign_thrlFCs_thrr5r6 color_pos color_neg color_nullrGrHr+ return_figr,rrrfrhr*max_nhas_negr4rTsignstextsrPrQsrrr plot_volcanoPsh: "$,   rcCst}t}t| }|}|||d<t|| |d<t|||d}| |d<|d|k|d|k@}|d| k|d|k@}||j|df<| |j|df<d}| durc|jdd| | d\}} |jj dddd| d | d | j |d d d | j |d d d| j | d d d|||Bj ddd}|jd|}| dg}t|d|d|jD]\}}}|| |||qt|dkr|j|tdd d| dt|| ||r|SdS)a Plot logFC and p-values from a long formated data-frame. Parameters ---------- data : pd.DataFrame Results of DEA in long format. x : str Column name of data storing the logFCs. y : str Columns name of data storing the p-values. top : int Number of top differentially expressed features to show. sign_thr : float Significance threshold for p-values. lFCs_thr : float Significance threshold for logFCs. sign_limit : float Limit of p-values to plot in -log10. lFCs_limit : float Limit of logFCs to plot in absolute value. color_pos: str Color to plot significant positive genes. color_neg: str Color to plot significant negative genes. color_null: str Color to plot rest of the genes. figsize : tuple Figure size. dpi : int DPI resolution of figure. ax : Axes, None A matplotlib axes object. If None returns new figure. return_fig : bool Whether to return a Figure object or not. save : str, None Path to where to save the plot. Infer the filetype if ending on {``.pdf``, ``.png``, ``.svg``}. Returns ------- fig : Figure, None If return_fig, returns Figure object. r/r.rNr=NrrFFrWTrXrYrZr]r^r`rrarbrd)rrr0ricopyr9r3rjrqrr set_axisbelowrvrwrxryrzr{rfr|r}r~rrr-)datarPrQrrrr5r6rrrrGrHr+rr,rrr4up_mskdw_mskr*rrrrrrplot_volcano_dfs@/       r r:r:c Cst} t}||}|}||||k||g}tj||gddd}||dk||dk@||dk||dk@B}d|d<d|j|df<d}| dur\| jdd|| d \}} |jj ||d| d | | d | j dd d d| j dd d d| |t|||||j}|j|ddf}g}t|||||jD]\}}}|| |||qt|dkr|j|tdd d| dt|| | | r|SdS)a Plot the weight and statistic of the target genes of a given source. Parameters ---------- data : pd.DataFrame Results of DEA in long format. stat : str Column name of data storing feature statistics. source_name : str Name of source to plot. net : DataFrame Network dataframe.. source : str Column name in net with source nodes. target : str Column name in net with target nodes. weight : str, None Column name in net with weights. top : int Number of features to show labels. figsize : tuple Figure size. dpi : int DPI resolution of figure. ax : Axes, None A matplotlib axes object. If None returns new figure. return_fig : bool Whether to return a Figure object or not. save : str, None Path to where to save the plot. Infer the filetype if ending on {``.pdf``, ``.png``, ``.svg``}. Returns ------- fig : Figure, None If return_fig, returns Figure object. rinner)rMjoinrrAr\r@NrF)rPrQrRr+TrXrYrZr]rarbrd)rrrrmroconcatr3rjrqrrgridrrvrwrsr0r2rxtailrfr{r|r}r~rrr-)rstat source_namerr;r<r=rrGrHr+rr,rrwposr* top_namesrrPrQrrrr plot_targets8s:) 0  "   rc Cst} t} t||d\}}}t||jd}t|tr$| }n| }d}|dur9| j dd||d\}}|r@t |}| j |||| d|durU|j |ddd|jd d d |dure|||durn||t||| | rx|SdS) a  Plot distribution of features' values per sample. Parameters ---------- mat : list, DataFrame or AnnData List of [features, matrix], dataframe (samples x features) or an AnnData instance. thr : float Threshold to plot horizontal line. log : bool Whether to log1p the data or not. use_raw : bool Use raw attribute of mat if present. figsize : tuple Figure size. dpi : int DPI resolution of figure. ax : Axes, None A matplotlib axes object. If None returns new figure. title : str Text to write as title of the plot. ylabel : str Text to write as ylabel of the plot. color : str Color to plot the violins. return_fig : bool Whether to return a Figure object or not. save : str, None Path to where to save the plot. Infer the filetype if ending on {``.pdf``, ``.png``, ``.svg``}. Returns ------- fig : Figure, None If return_fig, returns Figure object. )use_rawrNrF)rPrQr+r\rXrYrZrPZrMrotation)rrrr0repeatshape isinstancertoarrayflattenrjlog1p violinplotrv tick_paramsrsrzr-)matthrlogrrGrHr+titleylabelr\rr,rrmrrRrPrQr*rrr plot_violinss0'     rcCsV|dur|}|dur|}|dur|}||kr| }||kr&| }|||fSN)minmaxmean)rTvcenterrUvaluesrrr set_limitss rrOc Cst} t}tdd}tt|rtd|j|g}d|j_d|j _|j ddt t |j dfj|j|diddd}|rNd \}}nd \}}d}| durc|jdd|| d \}} | j|||| d |rtd d| jdD}| d| dntdd| jdD}| jddd| d| dt||||d\}}}|jj|||d}||}|||}t| jd|D] \}}||q|j j!||d}|"g| #j$|| dt%|| | | r|SdS)a` Plot barplots showing the top absolute value activities. Parameters ---------- acts : DataFrame Activities obtained from any method. contrast : str Name of the contrast (row) to plot. top : int Number of top features to plot. vertical : bool Whether to plot verticaly or horizontaly. cmap : str Colormap to use. vmin : float, None The value representing the lower limit of the color scale. vcenter : float, None The value representing the center of the color scale. vmax : float, None The value representing the upper limit of the color scale. figsize : tuple Figure size. dpi : int DPI resolution of figure. ax : Axes, None A matplotlib axes object. If None returns new figure. return_fig : bool Whether to return a Figure object or not. save : str, None Path to where to save the plot. Infer the filetype if ending on {``.pdf``, ``.png``, ``.svg``}. Returns ------- fig : Figure, None If return_fig, returns Figure object. Trz&Input acts contains non finite values.NractsrrL)rrf)rfrrF)rrPrQr+cSg|]}|qSr) get_width.0barrrr @z plot_barplot..ActivitycSrr) get_heightrrrrrDrrPrrrTrrU)rSnorm)r+)&rrr0rlisfiniter)r3rfrrhryargsortr2rrnr reset_indexrurxrjbarplotarray containers set_xlabelrzrrcolors TwoSlopeNormget_cmapr{ set_facecolorcmScalarMappable set_array get_figurecolorbarr-)rrrverticalrSrTrrUrGrHr+rr,rrrr4rPrQr*sizesdivnormcmap_f div_colorsrr\smrrr plot_barplotsZ)               rcCsxt|tur |g}|durt|jddg}dg}||fSg}||j}t|}|D] }||k}||q,||fS)NrT)typestrr0fullrruniquer|)r4groupbymskscatssubcatmskrrr build_msks^s    rcCsJ|dur |||dur|}|||dur|}||dSr)rsupperrrz)r+rxlabelrrPrQrrr write_labelsrs  raurocauprcTrYcst}t}t||\}}d}g}t||D]f\}}||jddgddjddjdddd}||j||j|djdurQ|j d d ||d \} dj d d j d |d |r|fddt tD}||q|dur|rjdddd|rtttg}tttg}|||||r|j|tddddt| | | ||t|||r|SdS)a Plot scatter plot of metrics across two different axes. Parameters ---------- df : DataFrame Performance metrics per method, obtained by running run_benchmark. x : str Name of the metric to plot in the x axis. y : str Name of the metric to plot in the y axis. groupby : str Metrics can be gruped by an extra categorical column. show_text : bool Whether to plot text labels. show_legend : bool Whether to plot the legend. mirror_xy : bool Whether to make x and y axis have the same values. figsize : tuple Figure size. dpi : int DPI resolution of figure. ax : Axes, None A matplotlib axes object. If None returns new figure. title : str Text to write as title of the plot. xlabel : str Text to write as xlabel of the plot. ylabel : str Text to write as ylabel of the plot. color : str Color to plot the dots. return_fig : bool Whether to return a Figure object or not. save : str, None Path to where to save the plot. Infer the filetype if ending on {``.pdf``, ``.png``, ``.svg``}. Returns ------- fig : Figure, None If return_fig, returns Figure object. NmethodmetricTobserved numeric_onlyscorerfrhrrrFrzorder)rlabelcs*g|]}j|||ddqS)rKr)r})rir+s_valsx_valsy_valsrrrs*z(plot_metrics_scatter.. center leftrr?Fr3bbox_to_anchorframeonrarBrbrd)rrrr{rrrpivotrrjrrrrranger~extendlegendr0r get_xticks get_yticksrset_xlimset_ylimrrrr-)r4rPrQr show_text show_legend mirror_xyrGrHr+rrrr\rr,rrrrr*rrrrr}rmin_nrrrplot_metrics_scatter~sT0         ""   rr c  Cst} t||j} tt| j|} t||j} t|dj}| D][}| D]V}||||k|||k@jddk}|r|D]<}|dksN|dkrctj |||tj tj gg||dddgd}ntj ||d|tj tj gg||ddddgd}t ||g}qDq+q'| j | |||dddd \}}| }t|jD])}||}|t| kr| |}t||||k|||d d d ||d q|d q|t| d \}}|j||ddd dt||| |r|SdS)a Extension of the function plot_metrics_scatter to group metrics by two categories at the same time. Parameters ---------- df : DataFrame Performance metrics per method, obtained by running run_benchmark. col : str Name of the group column to group by. Each of its categories will become a subplot. x : str Name of the metric to plot in the x axis. y : str Name of the metric to plot in the y axis. groupby : str Name of the group column to additionaly group by. Each of its categories will appear in the legend. n_cols : int Number of columns per row. figsize : tuple Figure size. dpi : int DPI resolution of figure. return_fig : bool Whether to return a Figure object or not. save : str, None Path to where to save the plot. Infer the filetype if ending on {``.pdf``, ``.png``, ``.svg``}. Returns ------- fig : Figure, None If return_fig, returns Figure object. rrrrci)rhrT)rGrH tight_layoutrVshareyF)rPrQrrrrr+roffrrrrN)rr0rrintceilsizerro DataFramenanrrjravelr r~rrMget_legend_handles_labelsr r-)r4colrPrQrn_colsrGrHrr,rrn_rowsgroupbysmetricsrgrpbyis_emptyrtmpr*axesrr+handleslabelsrrrplot_metrics_scatter_colssD#&*    r0c  Kst} t} |dvrtd||d|k}d}|dur'| jdd||d\}}|d|jdd t|tur||d kr||jd |gdd j dd  jd dd  d j }| jdd d ||||t||d| |dur{|jddddn%|dur|d gj dd  d j }| jdd d |||d| ntd|jddd|dur|||durd}|||dur|}||t||| | r|SdS)ao Plot boxplots showing the distribution of scores between methods for a metric. Parameters ---------- df : DataFrame Performance metrics per method, obtained by running run_benchmark. metric : str Name of metric to plot, must be either "mcauroc", "mcauprc", "rank" or "nrank". groupby : str Metrics can be gruped by an extra categorical column. figsize : tuple Figure size. dpi : int DPI resolution of figure. ax : Axes, None A matplotlib axes object. If None returns new figure. title : str Text to write as title of the plot. xlabel : str Text to write as xlabel of the plot. ylabel : str Text to write as ylabel of the plot. return_fig : bool Whether to return a Figure object or not. save : str, None Path to where to save the plot. Infer the filetype if ending on {``.pdf``, ``.png``, ``.svg``}. kwargs : dict Other keyword arguments are passed through to seaborn.boxplot(). Returns ------- fig : Figure, None If return_fig, returns Figure object. )mcaurocmcauprcranknrankrecallzGArgument metric must be either "mcauroc", "mcauprc", "rank" or "nrank".rNrrFTrrrrrr)rPrQhuerr+order hue_orderrrFr)rPrQrr+r7z6Argument groupby must be a string and not be "method".rPrrMethodsr)rrr)rjrrrrrrrrrxrfboxplotr0sortr get_legendr rrsrrrzr-)r4rrrGrHr+rrrrr,kwargsrrr*r7rrrplot_metrics_boxplot7sl'         r>c Kst}t} |j} t| d| d<t| d| d<t|tur+|dur+tdt|tur{| j dt |||dd\} } | } t | |D]1\}} |j dd |d|jddd| || dd ||jd d d | d|d|dqHn>d} |dur| j dd||d\} }|j dd |d|jddd||| dd ||jd d d |d|d|dt| |||r| SdS)ae Quality Control plot to assess the quality of the obtained pseudobulk samples. Parameters ---------- adata : AnnData AnnData obtained after running ``decoupler.get_pseudobulk``. groupby : str, list Name of the ``.obs`` column to group by. Can also be a list of columns. figsize : tuple Figure size. dpi : int DPI resolution of figure. ax : Axes, None A matplotlib axes object. If None returns new figure. return_fig : bool Whether to return a Figure object or not. save : str, None Path to where to save the plot. Infer the filetype if ending on {``.pdf``, ``.png``, ``.svg``}. kwargs : dict Other keyword arguments are passed through to seaborn.scatterplot(). Returns ------- fig : Figure, None If return_fig, returns Figure object. psbulk_n_cells psbulk_countsNz1If a grupby is a list of columns ax must be None.rTrGrHrrr)rPrQr6r+rrrrF)r3rr rzLog10 number of cellsLog10 total sum of countsrFr)rrobsrr0rirlistr)rjr~r#r{rr scatterplotr rrzr-)adatarrGrHr+rr,r=rrr4r*r-grprrrplot_psbulk_sampless>            rHffffff?viridisr:rc Ks8t}t}t||\}}}|durtj|dd}t||||}|d8}|d8}t||}t||}tj||kdd}tj|dd}tj||dk<d}| durZ|j dd| | d\}} |j dt |||dt d d d d | d || j |dddd| jt |ddd| d| dt|| | | r|SdS)a Plot to help determining the thresholds of the ``decoupler.filter_by_expr`` function. Parameters ---------- adata : AnnData AnnData obtained after running ``decoupler.get_pseudobulk``. obs : DataFrame, None If provided, metadata dataframe, only needed if ``adata`` is not an ``AnnData``. group : str, None Name of the ``.obs`` column to group by. If None, it assumes that all samples belong to one group. lib_size : int, float, None Library size. If None, default to the sum of reads per sample. min_count : int Minimum count requiered per gene for at least some samples. min_total_count : int Minimum total count required per gene across all samples. large_n : int Number of samples per group that is considered to be "large". min_prop : float Minimum proportion of samples in the smallest group that express the gene. cmap : str Colormap to use. figsize : tuple Figure size. dpi : int DPI resolution of figure. ax : Axes, None A matplotlib axes object. If None returns new figure. return_fig : bool Whether to return a Figure object or not. save : str, None Path to where to save the plot. Infer the filetype if ending on {``.pdf``, ``.png``, ``.svg``}. kwargs : dict Other keyword arguments are passed through to ``sns.histplot``. Returns ------- fig : Figure, None If return_fig, returns Figure object. NrrLg+=r?rFTg?zNumber of features)shrinkr)FT)rPrQrScbarcbar_kwsdiscreter+r?rBrX)rQrRlsrPrRrRrBzNumber of samplesr)rrrr0sumr r r r"rjhistplotrirrvrwrrzr-)rFrCgrouplib_size min_countmin_total_countlarge_nmin_proprSrGrHr+rr,r=rrrQ var_namesmin_sample_size cpm_cutoffcpm sample_size total_countsr*rrrplot_filter_by_exprs<,      rb皙?rKc Kst} d} t|trl|j} dt| vr$|jd} t| tjr#| j} nt | t j | |kdd}d}|durB| j dd||d\}}|j |fdd d | }|j|d d d |d|dt||||rj|SdSt | )a6 Plot to help determining the thresholds of the ``decoupler.filter_by_prop`` function. Parameters ---------- adata : AnnData AnnData obtained after running ``decoupler.get_pseudobulk``. It requieres ``.layer['psbulk_props']``. min_prop : float Minimum proportion of cells that express a gene in a sample. min_smpls : int Minimum number of samples with bigger or equal proportion of cells with expression than ``min_prop``. cmap : str Colormap to use. figsize : tuple Figure size. dpi : int DPI resolution of figure. ax : Axes, None A matplotlib axes object. If None returns new figure. return_fig : bool Whether to return a Figure object or not. save : str, None Path to where to save the plot. Infer the filetype if ending on {``.pdf``, ``.png``, ``.svg``}. kwargs : dict Other keyword arguments are passed through to ``matplotlib.pyplot.hist``. Returns ------- fig : Figure, None If return_fig, returns Figure object. zadata must be an AnnData object that contains the layer 'psbulk_props'. Please check the function decoupler.get_pseudobulk. psbulk_propsrrLNrrFTrB)rr\rYrXrSz#Number of samples where >= min_propzNumber of genes)rrrlayerskeysrDror!rr)r0rTrjhistrwrrzr-)rFr[ min_smplsrSrGrHr+rr,r=rmsg layer_keyspropsnsmplsr*_rrrplot_filter_by_propCs."        rnRdBu_rc &Cs,t} tdd} Gddd| jj} |jjd}||j}t|}||}||}t|||dd}t ||dd }||d |k}t | }||}||}t |j }t ||d }|j }|}d ||}t|tj|tj|t|\}}}t|}||}|dkr||||k}n||||k}| jd ddgdi|d|d\}}|}|d} | d| j||ddd| j||ddd| jdddd| d| ||d} | d| g| g| j||ddddd|d} | d| g| gtt |d}!tt!|d}"| |!d|"d}#| j"|tj#ddfd|#|d| $d|j d|d} | d| j%||ddd ||dk}$|$j dkrh|d!}%n|$d!d}%| j|%ddd| &d"| d#|j'ddd$t(|| | | r||fSdS)%a) Plot the running score of GSEA. Parameters ---------- df : DataFrame Long format DataFrame with features to be ranked. Assumes features are indexes. stat : str Name of the column containing the ranking metric. net : DataFrame Network in long format. set_name : str Name of the feature set to plot. source : str Column name in net with source nodes. target : str Column name in net with target nodes. cmap : str Colormap to use. figsize : tuple Figure size. dpi : int DPI resolution of figure. return_fig : bool Whether to return a Figure object or not. save : str, None Path to where to save the plot. Infer the filetype if ending on {``.pdf``, ``.png``, ``.svg``}. Returns ------- fig : Figure Returns Figure object. le_c: ndarray List of leading edge features. If ES is positive, these are the top ranked features. If ES is negative, these are the bottom ranked features. Trcs(eZdZdfdd ZdddZZS) z-plot_running_score..MidpointNormalizeNFcs||_t|||dSr)rsuper__init__)selfrTrUrclip __class__rrrqsz6plot_running_score..MidpointNormalize.__init__cSs0|j|j|jggd}}tjt|||S)N)rr?r)rTrrUr0ma masked_arrayinterp)rrvaluersrPrQrrr__call__sz6plot_running_score..MidpointNormalize.__call__)NNNFr)__name__ __module__ __qualname__rqrz __classcell__rrrtrMidpointNormalizesrUNrIr)rr;r<rMrr height_ratios)rr?r?rK) gridspec_kwrGrVrHz#88c544rK)r\ linewidthrXr[r\zEnrichment Scorer?)rr\br) rasterizedrrSz#C9D3DB)y1y2r\Rankz Ranked metric)wspacehspace))rr Normalizerfrastyper0rrrraranger isinrTr float32int64boolsignrjr#marginsrqrwrvrzrsset_yticklabels set_yticksvlines percentilerr pcolormeshnewaxisr fill_betweenrsubplots_adjustr-)&r4rrset_namer;r<rSrGrHrr,rrrrRrrsnetidxrnksset_msk n_features nf_in_setdecmx_valuejesrset_rnksle_cr*r-r+rTrUmidnorm non_zero_rnkszero_rnkrrrplot_running_scores(      * "                 rrrc Kst} ||j}||j}t|}||}||}d}| dur-| jdd||d\}} | jd |||d| | || || j dd|durU| j |d|d|dur^| |t || | | rh|SdS) aN Plot results of enrichment analysis as bars. Parameters ---------- df : DataFrame Results of enrichment analysis. x : str Column name of ``df`` to use as continous value. y : str Column name of ``df`` to use as labels. color : str Color to plot the bars. thr : float, None If specified, x value where to draw a vertical dashed line. thr_color : str Color of the vertical dashed line. title : str, None Text to write as title of the plot. figsize : tuple Figure size. dpi : int DPI resolution of figure. ax : Axes, None A matplotlib axes object. If None returns new figure. return_fig : bool Whether to return a Figure object or not. save : str, None Path to where to save the plot. Infer the filetype if ending on {``.pdf``, ``.png``, ``.svg``}. kwargs : dict Other keyword arguments are passed through to ``matplotlib.pyplot.barh``. Returns ------- fig : Figure, None If return_fig, returns Figure object. NrrF)rQwidthr\g?)rQrXrr) rrrr0rrjbarhrrzrrwrsr-)r4rPrQr\r thr_colorrrGrHr+rr,r=rx_valy_valrr*rrrplot_barplot_dfs*)       r viridis_rrr:c  s`t||j} |dur||j}n|jj}||j}||j}t| }| |} ||}||}||}jdd|| d\}} |jdd}| jdd| j| ||||d}| d | ||j d d d fd dd\}}| j |||ddddd|j |dddddd}|jj|dd| jddd|dur| |t|| | | r|SdS)a  Plot results of enrichment analysis as dots. Parameters ---------- df : DataFrame Results of enrichment analysis. x : str Column name of ``df`` to use as continous value. y : str Column name of ``df`` to use as labels. c : str Column name of ``df`` to use for coloring. s : str Column name of ``df`` to use for dot size. scale : int Parameter to control the size of the dots. cmap : str Colormap to use. title : str, None Text to write as title of the plot. figsize : tuple Figure size. dpi : int DPI resolution of figure. ax : Axes, None A matplotlib axes object. If None returns new figure. return_fig : bool Whether to return a Figure object or not. save : str, None Path to where to save the plot. Infer the filetype if ending on {``.pdf``, ``.png``, ``.svg``}. Returns ------- fig : Figure, None If return_fig, returns Figure object. NrrFlines.markersizerKrPrL)rPrQrRrrSTrrz{x:.2f}cst|jdS)Nr)r0sqrtrcParams)rrscalerrszplot_dotplot..)propnumfmtfuncF)rMg?z upper leftrM)rr rr3 labelspacingg?rr)rMrcright)rNaspect orientationanchorlocationleft)r3皙?)rPrQ)rrrfr0rrjrrrrrrlegend_elementsr rr+rsrr-)r4rPrQrRrrrSrrGrHr+rr,rrc_valsridxsr*nsrrr.r/clbrrr plot_dotplotfsn)             rc sFd}t|drt|dst|dst|dsJ||dus"Jd||jvs-Jd|j|} || jvs;Jd|durG| | j|} |durS| | j|} |dusc|d usc|duscJd |duri|sw|durq|dusw|r{|dur{td |durt|dsJd ||jvsJd t |j|t j r|j|} | | fS| dd dd dd ddfddt |j|jdD} t j |j||jj| d} | | fS|durt|dsJd||jvsJdt j |j||jj|jjd} | | fS|rt|dsJdt j |j|jj|jjd} | | fS)NzXdata sould be an AnnData/MuData object with an .uns, and .obsm, .X, or .layers attributeunsobsmXrezuns_key must be specifiedzuns_key not found in data.unsz8stat_col must be one of the columns in the anova resultsTz2one of obsm_key, use_X, or layer must be specifiedz9only one of obsm_key, use_X, or layer should be specifiedz)the data does not have an .obsm attributezobsm_key not found in data.obsmX_rpcaPCmofaFactorumapUMAPcsg|] }dd|qS)z{0}{1}r)rt)rrP column_namerrrz-_check_assoc_plot_intputs..rrfrhz*the data does not have a .layers attributezlayer not found in data.layersz%the data does not have a .X attribute)hasattrrrfrhvariablerfactorr)rrror!replacer rrCrfrevarr) r associationscolsuns_keyobsm_keyuse_Xlayerstat_colristatsrrhrrr_check_assoc_plot_intputssF0  ( $"  rp_adjScoresStatsPurplesBrBGSet2c$ Cszddl}ddlm}ddlm}ddlm}Wn ty#tdwt||||||||\}}|d }|j dd|dj j dd|f}d|j _d|j_|dvrYt| }|j dd|jd| f}|j dd|jf}|j|| d d d d }|j| d d d||d|||j |jt|  k}|j|dd|t| d}|j|dd|j|j| dd dd }|j| dd d||d|D]}t|j|j}|j |||d}|j|dddq|!||j|d|} | jdddd| "|s|dur|#| j$d d d}!|d d d d!d"d|!j%d#}"|!&|"| j$ddd}!|d d d d!d"d|!j%d#}"|!&|"| j'}#|#(|d|#)|d |#*|t+|#d$||rj|#SdS)%a2 Create a composite plot displaying association results between scores (bottom) and summary statistics (top) using a clustermap. Requires PyComplexHeatmap to be installed. Parameters: ------------ data : AnnData or MuData The input data containing the association results in .uns[uns_key] and the underlying data in .obsm[obsm_key], .X or .layers[layer]. uns_key : str, optional Key in `data.uns` where the association statistics are stored. associations : list, optional List of association names to be plotted. If None, all associations will be plotted (default is None). cols : list, optional List of columns to be plotted (i.e. columns of .obsm/rows of .var). If None, all columns will be plotted (default is None). obsm_key : str, optional Key of `data.obsm` used to plot the bottom clustermap. Either `obsm_key`, `use_X`, or `layer` must be specified. use_X : bool, optional Boolean indicating whether to use the data in `.X` for the bottom clustermap. Either `obsm_key`, `use_X`, or `layer` must be specified. layer : str, optional Key of `data.layers` used to plot the bottom clustermap. Either `obsm_key`, `use_X`, or `layer` must be specified. stat_col : str, optional Name of the summary statistic column in `data.uns[uns_key]` to be shown in the top clustermap (default is 'p_adj'). titles : list, optional A list of two strings representing the titles for the ClusterMap for scores and statistics, respectively (default is ['Scores', 'Stats']). sign_thr : float Significance threshold for p-values. n_factors : int Number of factors (score features) to plot. cmap_stats : str Colormap to use for stats values. cmap_scores : str Colormap to use for scores values. cmap_cats : str Colormap to use for ``obs_annotation_cols`` categories. figsize : tuple Figure size. dpi : int DPI resolution of figure. return_fig : bool Whether to return a Figure object or not. save : str, None Path to where to save the plot. Infer the filetype if ending on {``.pdf``, ``.png``, ``.svg``}. Returns ------- fig : Figure, None If return_fig, returns Figure object. rN) RectanglezImarsilea is not installed. Please install it using "pip install marsilea"rrr)pvalrrrh1)rSrheightrcenter)ralignr*z {0} < {1})markerrr)rg?h2)rSrrrc)padr rrowrK)sidestack_by stack_size) board_nameax_name)rrFz.1)filleclw transformT),marsileamarsilea.plotterplottermatplotlib.patchesrr rrrrr rnr3rfrrhr0rirHeatmap add_title add_legendsadd_dendrogram add_rightLabelsr MarkerMeshrtr add_layerrDrCColors add_bottomrendercloseget_ax transAxes add_artistfigure set_figwidth set_figheightset_dpir-)$rrrrobs_annotation_colsrrrrtitlesr n_factors cmap_stats cmap_scores cmap_catsrGrHrr,rvmprrrr col_orderrsign_mskrcat_colr cat_colorsrRhaxborderr*rrrplot_associationssn;              r,cCsFtt|j|jg}ddt|D}t||jd}||fS)NcSsi|]\}}||qSrr)rrkrrr sz"get_dict_types..r)r0rhstackrh enumerater)actrCvsv_dicttypesrrrget_dict_typessr5cCsLg}|jD]}|j|df|j|df}}||||g}||q|S)Nr;r<)rfr3r|)r3redgesrr;r<edgerrrnet_to_edgelists   r8cCspt}t||\}}t||}|j|dd}|dj|jd<||jd<t||jd<t |dd|jd<|S) NT)r6directedr=rrcirclesquarer) r%r5r8Graphrrr2rDrfr0where)r1rCrr$r3r4r6grrrget_gs  r?cCsftdd}|j}|rtt|}|jj| |d}|St|}t|}|jj||d}|S)NTr)rTrU) rrr#r0rr2rrr)r4rrrPrUrrTrrrget_norms   r@cCst|trt|j|g}|St|trt|j|}|St|trD|r5tt|j d d|}|St|j d d|}|St d t |)NrzZn_sources needs to be a list of source names or an integer number, {0} was passed.) rrr0rrhrDrrr2rr)rtr) n_sourcesr1by_abss_idxrrrget_source_idxss    rDcst|trt|dt|g}|St|tr%t|dt|}|St|tr^fddt|d|dD|d<|rFt|d|d<|j ddgddgd j dgdd  |j j }|Std t|) Nr<cs0g|]\}}|jvrjd||ndqSr)rhry)rtrrCrrrs0z#get_target_idxs..r=prodr;TFr^rzZn_targets needs to be a list of target names or an integer number, {0} was passed.)rrr0rrrDrr{r2rxrheadrfrr)rtr) n_targetsrCrrBt_idxrrGrget_target_idxss.   $  rLcCst|j|jkr|jjdksJdt|||}|jdd|f}|jt|dt |j t  }d|j vr?d|d<t ||||}|j|}|jddt|j t |dt f}|jt|dt |j t ddf}|||fS)Nrz,obs and act need to have the same row index.r;r=r<) r0allrfr rDryr3rrrrhrrL)r1rCrrArJrBrCrKrrrget_obs_act_nets& (  ,, rNc stdd}|j}|vrS|vrS|j|jg} t|jdD](\} } |jd| rA| ||| jdq'| ||| jdq'd} nfdd|jdD} d} | |jd <| S) NTrrrrcsg|] }|dkr nqS)rr)rtyps_cmapt_cmaprrr rzadd_colors..Fr\)r colormapsrfrr0r2r|r) r>r1rCs_normt_normrQrRrcmapsr\rr-is_cmaprrPr add_colorss     rXdarkreddarkblueEnrichment scoreGene expressionkkrrc%stdd}tdd}t}|durE|durE|dd}tjtd|jfdg|d }|d d}tjtd|jfdg|d }t ||||||\}}}t || d }t || d }t |||}fd d |j dD|j d<t ||||||| }|j||dd} |jjddgdd}!| |!ddddf}"| |!d}#| |!d}$|j||"|||d|r| j|jj||d|#d| d| j|jj|| d|$d|dn |#d|$dt| d||r| SdS)a Plot results of enrichment analysis as network. Parameters ---------- net : DataFrame Network dataframe with ``source`` and ``target`` columns (``weight`` is optional). obs : None, DataFrame Input of enrichment analysis, needs to be a one row dataframe with targets as features. Used to filter net. act : None, DataFrame Ouput of enrichment analysis, needs to be a one row dataframe with sources as features. Used to filter net. n_sources : str, list, int Number of top sources to plot or list of source names. n_targets : str, list, int Number of top targets to plot or list of target names. by_abs : bool Whether to consider the absolute value when sorting for ``n_sources`` and ``n_targets``. node_size : float Size of the nodes in the plot. label_size : int Size of the labels in the plot. s_cmap : str Color or colormap to use to color sources. t_cmap : str Color or colormap to use to color targets. vcenter : bool Whether to center colors around 0. c_pos_w : str Color for edges with positive weights. If no weights are available, they are set to positive by default. c_neg_w : str Color for edges with negative weights. s_label : str Label to place in the source colorbar. t_label : str Label to place in the target colorbar. layout : str Layout to use to order the nodes. Check ``igraph`` documentation for more options. figsize : tuple Figure size. dpi : int DPI resolution of figure. return_fig : bool Whether to return a Figure object or not. save : str, None Path to where to save the plot. Infer the filetype if ending on {``.pdf``, ``.png``, ``.svg``}. Returns ------- fig : Figure, None If return_fig, returns Figure object. FrTNr;rr0rr<)rcsg|] }|dkr nqSrEr)rrc_neg_wc_pos_wrrr[rz plot_network..r=r\rAr:r)rrrrr)rr)rr)rr)r<layout vertex_sizevertex_label_size)rrS horizontal)caxrrr)rr%rrror!r0onesr rNr@r?rrXrgridspecGridSpec add_subplotrqrrrrMr-)%rrCr1rArJrB node_size label_sizerQrRrrcrbs_labelt_labelrdrGrHrr,rrr$sourcestargetsfactfobsfnetrTrUr>rWr*gsax1ax2ax3rrar plot_networksF 8         rz)F)NN)NNr:r;r<r=r>r?NNr@rArBrCrENFN) r:r>r?NNr@rArBrCrENFN) r;r<r=rrrENFN) NFFrCrENNNrAFN) rFrONrNrCrENFN)rrNTTTrrENNNNrYFN)rrNrrrEFN) NrrENNNNFN)rrENFN) NNNrrIrrJrKrLrENFN)rcrKrKrLrENFN)r;r<rorrEFN) rBNrYNrrENFN)r:rNrrENFN)NNr:rTr?r:rorKFrYrZr[r\r]r^r_FN)5numpyr0 scipy.sparserpandasroanndatarprerrr utils_anndatarr r r method_gsear rrrr%r-r9rrrrrrrrrr0r>rHrbrnrrrrr,r5r8r?r@rDrLrNrXrzrrrrs       b W N m  h Q mE Z C  G n +