pertpy 0.10.0__py3-none-any.whl → 0.11.0__py3-none-any.whl

pertpy/tools/_dialogue.py

@@ -33,9 +33,17 @@ if TYPE_CHECKING:
 
 
 class Dialogue:
-    """Python implementation of DIALOGUE"""
+    """Python implementation of DIALOGUE."""
 
-    def __init__(self, sample_id: str, celltype_key: str, n_counts_key: str, n_mpcs: int):
+    def __init__(
+        self,
+        sample_id: str,
+        celltype_key: str,
+        n_counts_key: str,
+        n_mpcs: int,
+        feature_space_key: str = "X_pca",
+        n_components: int = 50,
+    ):
         """Constructor for Dialogue.
 
         Args:
@@ -43,6 +51,8 @@ class Dialogue:
             celltype_key: The key in AnnData.obs which contains the cell type column.
             n_counts_key: The key of the number of counts in Anndata.obs . Also commonly the size factor.
             n_mpcs: Number of PMD components which corresponds to the number of determined MCPs.
+            feature_space_key: The key in adata.obsm for the feature space (e.g., "X_pca", "X_umap").
+            n_components: The number of components of the feature space to use, e.g. PCA components.
         """
         self.sample_id = sample_id
         self.celltype_key = celltype_key
@@ -53,6 +63,8 @@ class Dialogue:
             )
         self.n_counts_key = n_counts_key
         self.n_mcps = n_mpcs
+        self.feature_space_key = feature_space_key
+        self.n_components = n_components
 
     def _get_pseudobulks(
         self, adata: AnnData, groupby: str, strategy: Literal["median", "mean"] = "median"
@@ -62,6 +74,7 @@ class Dialogue:
         Copied from `https://github.com/schillerlab/sc-toolbox/blob/397e80dc5e8fb8017b75f6c3fa634a1e1213d484/sc_toolbox/tools/__init__.py#L458`
 
         Args:
+            adata: Annotated data matrix.
             groupby: The key to groupby for pseudobulks
             strategy: The pseudobulking strategy. One of "median" or "mean"
 
@@ -83,7 +96,9 @@ class Dialogue:
         return pseudobulk
 
     def _pseudobulk_feature_space(
-        self, adata: AnnData, groupby: str, n_components: int = 50, feature_space_key: str = "X_pca"
+        self,
+        adata: AnnData,
+        groupby: str,
     ) -> pd.DataFrame:
         """Return Cell-averaged components from a passed feature space.
 
@@ -91,9 +106,8 @@ class Dialogue:
         TODO: DIALOGUE recommends running PCA on each cell type separately before running PMD - this should be implemented as an option here.
 
         Args:
+            adata: Annotated data matrix.
             groupby: The key to groupby for pseudobulks.
-            n_components: The number of components to use.
-            feature_key: The key in adata.obsm for the feature space (e.g., "X_pca", "X_umap").
 
         Returns:
             A pseudobulk DataFrame of the averaged components.
@@ -101,7 +115,7 @@ class Dialogue:
         aggr = {}
         for category in adata.obs.loc[:, groupby].cat.categories:
             temp = adata.obs.loc[:, groupby] == category
-            aggr[category] = adata[temp].obsm[feature_space_key][:, :n_components].mean(axis=0)
+            aggr[category] = adata[temp].obsm[self.feature_space_key][:, : self.n_components].mean(axis=0)
         aggr = pd.DataFrame(aggr)
         return aggr
 
@@ -130,6 +144,7 @@ class Dialogue:
 
         Args:
             adata: The AnnData object to append mcp scores to.
+            ct_subs: cell type objects.
             mcp_scores: The MCP scores dictionary.
             celltype_key: Key of the cell type column in obs.
 
@@ -213,7 +228,7 @@ class Dialogue:
         sample_obs: str,
         return_all: bool = False,
     ):
-        """Applies a mixed linear model using the specified formula (MCP scores used for the dependent var) and returns the coefficient and p-value
+        """Applies a mixed linear model using the specified formula (MCP scores used for the dependent var) and returns the coefficient and p-value.
 
         TODO: reduce runtime? Maybe we can use an approximation or something that isn't statsmodels.
 
@@ -332,7 +347,7 @@ class Dialogue:
 
         Args:
             mcp_name: The name of the MCP to model.
-            scores: The MCP scores for a cell type. Number of MCPs x number of features.
+            scores_df: The MCP scores for a cell type. Number of MCPs x number of features.
             ct_data: The AnnData object containing the metadata and labels in obs.
             tme: Transcript mean expression in `x`.
             sig: DataFrame containing a series of up and downregulated MCPs.
@@ -418,11 +433,10 @@ class Dialogue:
         # Finally get corr coeff
         return np.dot(A_mA, B_mB.T) / np.sqrt(np.dot(ssA[:, None], ssB[None]))
 
+    # TODO: needs check for correctness and variable renaming
+    # TODO: Confirm that this doesn't return duplicate gene names.
     def _get_top_elements(self, m: pd.DataFrame, max_length: int, min_threshold: float):
-        """
-
-        TODO: needs check for correctness and variable renaming
-        TODO: Confirm that this doesn't return duplicate gene names
+        """Get top elements.
 
         Args:
             m: Any DataFrame of Gene name as index with variable columns.
@@ -457,12 +471,11 @@ class Dialogue:
         # TODO this whole function should be standalone
         # It will contain the calculation of up/down + calculation (new final mcp scores)
         # Ensure that it'll still fit/work with the hierarchical multilevel_modeling
-
         """Determine the up and down genes per MCP."""
         # TODO: something is slightly slow here
         cca_sig_results: dict[Any, dict[str, Any]] = {}
         new_mcp_scores: dict[Any, list[Any]] = {}
-        for ct in ct_subs.keys():
+        for ct in ct_subs:
             ct_adata = ct_subs[ct]
             conf_m = ct_adata.obs[n_counts_key].values
 
@@ -483,9 +496,7 @@ class Dialogue:
             from scipy.stats import spearmanr
 
             def _pcor_mat(v1, v2, v3, method="spearman"):
-                """
-                MAJOR TODO: I've only used normal correlation instead of partial correlation as we wait on the implementation
-                """
+                """MAJOR TODO: I've only used normal correlation instead of partial correlation as we wait on the implementation."""
                 correlations = []  # R
                 pvals = []  # P
                 for x2 in v2:
@@ -506,7 +517,7 @@ class Dialogue:
                 return np.array(correlations), np.array(pvals)  # pvals_adjusted
 
             C1, P1 = _pcor_mat(ct_adata[:, top_cor_genes_flattened].X.toarray().T, mcp_scores[ct].T, conf_m)
-            C1[P1 > (0.05 / ct_adata.shape[1])] = 0  # why?
+            C1[(0.05 / ct_adata.shape[1]) < P1] = 0  # why?
 
             cca_sig_unformatted = self._get_top_elements(  # 3 up, 3 dn, for each mcp
                 pd.DataFrame(C1.T, index=top_cor_genes_flattened), max_length=max_genes, min_threshold=0.05
@@ -514,7 +525,7 @@ class Dialogue:
 
             # TODO: probably format the up and down within get_top_elements
             cca_sig: dict[str, Any] = defaultdict(dict)
-            for i in range(0, int(len(cca_sig_unformatted) / 2)):
+            for i in range(int(len(cca_sig_unformatted) / 2)):
                 cca_sig[f"MCP{i}"]["up"] = cca_sig_unformatted[i * 2]
                 cca_sig[f"MCP{i}"]["down"] = cca_sig_unformatted[i * 2 + 1]
 
@@ -523,7 +534,7 @@ class Dialogue:
 
             # This is basically DIALOGUE 3 now
             pre_r_scores = {
-                ct: ct_subs[ct].obsm["X_pca"][:, :50] @ ws_dict[ct]
+                ct: ct_subs[ct].obsm[self.feature_space_key][:, : self.n_components] @ ws_dict[ct]
                 for i, ct in enumerate(ct_subs.keys())
                 # TODO This is a recalculation and not a new calculation
             }
@@ -591,8 +602,8 @@ class Dialogue:
     def calculate_multifactor_PMD(
         self,
         adata: AnnData,
-        penalties: list[int] = None,
-        ct_order: list[str] = None,
+        penalties: list[int] | None = None,
+        ct_order: list[str] | None = None,
         agg_feature: bool = True,
         solver: Literal["lp", "bs"] = "bs",
         normalize: bool = True,
@@ -603,10 +614,9 @@ class Dialogue:
 
         Args:
             adata: AnnData object to calculate PMD for.
-            sample_id: Key to use for pseudobulk determination.
             penalties: PMD penalties.
             ct_order: The order of cell types.
-            agg_features: Whether to calculate cell-averaged principal components.
+            agg_feature: Whether to calculate cell-averaged principal components.
             solver: Which solver to use for PMD. Must be one of "lp" (linear programming) or "bs" (binary search).
                     For differences between these to please refer to https://github.com/theislab/sparsecca/blob/main/examples/linear_programming_multicca.ipynb
             normalize: Whether to mimic DIALOGUE as close as possible
@@ -644,8 +654,6 @@ class Dialogue:
                     raise ValueError("Please ensure that every cell type is represented in every sample.") from e
                 else:
                     raise
-        else:
-            penalties = penalties
 
         if solver == "bs":
             ws, _ = multicca_pmd(mcca_in, penalties, K=self.n_mcps, standardize=True, niter=100, mimic_R=normalize)
@@ -656,8 +664,8 @@ class Dialogue:
         ws_dict = {ct: ws[i] for i, ct in enumerate(ct_order)}
 
         pre_r_scores = {
-            ct: ct_subs[ct].obsm["X_pca"][:, :50] @ ws[i]
-            for i, ct in enumerate(cell_types)  # TODO change from 50
+            ct: ct_subs[ct].obsm[self.feature_space_key][:, : self.n_components] @ ws[i]
+            for i, ct in enumerate(cell_types)
         }
 
         # TODO: output format needs some cleanup, even though each MCP score is matched to one cell, it's not at all
@@ -681,17 +689,17 @@ class Dialogue:
         ws_dict: dict,
         confounder: str | None,
         formula: str = None,
-    ):
+    ) -> pd.DataFrame:
         """Runs the multilevel modeling step to match genes to MCPs and generate p-values for MCPs.
 
         Args:
             ct_subs: The DIALOGUE cell type objects.
             mcp_scores: The determined MCP scores from the PMD step.
+            ws_dict: WS dictionary.
             confounder: Any modeling confounders.
             formula: The hierarchical modeling formula. Defaults to y ~ x + n_counts.
 
         Returns:
-            A Pandas DataFrame containing:
             - for each mcp: HLM_result_1, HLM_result_2, sig_genes_1, sig_genes_2
             - merged HLM_result_1, HLM_result_2, sig_genes_1, sig_genes_2 of all mcps
 
@@ -875,15 +883,15 @@ class Dialogue:
             if len(conditions_compare) != 2:
                 raise ValueError("Please specify conditions to compare or supply an object with only 2 conditions")
 
-        pvals = pd.DataFrame(1, adata.obs[celltype_label].unique(), ["mcp_" + str(n) for n in range(0, n_mcps)])
-        tstats = pd.DataFrame(1, adata.obs[celltype_label].unique(), ["mcp_" + str(n) for n in range(0, n_mcps)])
-        pvals_adj = pd.DataFrame(1, adata.obs[celltype_label].unique(), ["mcp_" + str(n) for n in range(0, n_mcps)])
+        pvals = pd.DataFrame(1, adata.obs[celltype_label].unique(), ["mcp_" + str(n) for n in range(n_mcps)])
+        tstats = pd.DataFrame(1, adata.obs[celltype_label].unique(), ["mcp_" + str(n) for n in range(n_mcps)])
+        pvals_adj = pd.DataFrame(1, adata.obs[celltype_label].unique(), ["mcp_" + str(n) for n in range(n_mcps)])
 
         response = adata.obs.groupby(sample_label)[condition_label].agg(pd.Series.mode)
         for celltype in adata.obs[celltype_label].unique():
             df = adata.obs[adata.obs[celltype_label] == celltype]
 
-            for mcpnum in ["mcp_" + str(n) for n in range(0, n_mcps)]:
+            for mcpnum in ["mcp_" + str(n) for n in range(n_mcps)]:
                 mns = df.groupby(sample_label)[mcpnum].mean()
                 mns = pd.concat([mns, response], axis=1)
                 res = stats.ttest_ind(
@@ -893,7 +901,7 @@ class Dialogue:
                 pvals.loc[celltype, mcpnum] = res[1]
                 tstats.loc[celltype, mcpnum] = res[0]
 
-        for mcpnum in ["mcp_" + str(n) for n in range(0, n_mcps)]:
+        for mcpnum in ["mcp_" + str(n) for n in range(n_mcps)]:
             pvals_adj[mcpnum] = multipletests(pvals[mcpnum], method="fdr_bh")[1]
 
         return {"pvals": pvals, "tstats": tstats, "pvals_adj": pvals_adj}
@@ -956,7 +964,7 @@ class Dialogue:
 
         genes_dict_up = {}  # type: ignore
         genes_dict_down = {}  # type: ignore
-        for celltype2 in mcp_dict.keys():
+        for celltype2 in mcp_dict:
             for gene in mcp_dict[celltype2][MCP + ".up"]:
                 if gene in genes_dict_up:
                     genes_dict_up[gene] += 1
@@ -1008,7 +1016,7 @@ class Dialogue:
             >>> genes_results = _get_extrema_MCP_genes_single(ct_subs, mcp="mcp_4", fraction=0.2)
         """
         genes = {}
-        for ct in ct_subs.keys():
+        for ct in ct_subs:
             mini = ct_subs[ct]
             mini.obs["extrema"] = pd.qcut(
                 mini.obs[mcp],
@@ -1056,13 +1064,13 @@ class Dialogue:
         for mcp in mcps:
             rank_dfs[mcp] = {}
             ct_ranked = self._get_extrema_MCP_genes_single(ct_subs, mcp=mcp, fraction=fraction)
-            for celltype in ct_ranked.keys():
+            for celltype in ct_ranked:
                 rank_dfs[mcp][celltype] = sc.get.rank_genes_groups_df(ct_ranked[celltype], group=None)
 
         return rank_dfs
 
     @_doc_params(common_plot_args=doc_common_plot_args)
-    def plot_split_violins(
+    def plot_split_violins(  # pragma: no cover # noqa: D417
         self,
         adata: AnnData,
         split_key: str,
@@ -1115,7 +1123,7 @@ class Dialogue:
         return None
 
     @_doc_params(common_plot_args=doc_common_plot_args)
-    def plot_pairplot(
+    def plot_pairplot(  # pragma: no cover # noqa: D417
         self,
         adata: AnnData,
         celltype_key: str,

pertpy/tools/_differential_gene_expression/_base.py

@@ -1,3 +1,4 @@
+import contextlib
 import math
 from abc import ABC, abstractmethod
 from collections.abc import Iterable, Mapping, Sequence
@@ -23,8 +24,7 @@ from pertpy.tools._differential_gene_expression._checks import check_is_numeric_
 
 class MethodBase(ABC):
     def __init__(self, adata, *, mask=None, layer=None, **kwargs):
-        """
-        Initialize the method.
+        """Initialize the method.
 
         Args:
             adata: AnnData object, usually pseudobulked.
@@ -62,8 +62,7 @@ class MethodBase(ABC):
         fit_kwargs=MappingProxyType({}),
         test_kwargs=MappingProxyType({}),
     ):
-        """
-        Compare between groups in a specified column.
+        """Compare between groups in a specified column.
 
         Args:
             adata: AnnData object.
@@ -100,7 +99,7 @@ class MethodBase(ABC):
         ...
 
     @_doc_params(common_plot_args=doc_common_plot_args)
-    def plot_volcano(
+    def plot_volcano(  # pragma: no cover # noqa: D417
         self,
         data: pd.DataFrame | ad.AnnData,
         *,
@@ -188,8 +187,7 @@ class MethodBase(ABC):
             colors = ["gray", "#D62728", "#1F77B4"]
 
         def _pval_reciprocal(lfc: float) -> float:
-            """
-            Function for relating -log10(pvalue) and logfoldchange in a reciprocal.
+            """Function for relating -log10(pvalue) and logfoldchange in a reciprocal.
 
             Used for plotting the S-curve
             """
@@ -197,7 +195,7 @@ class MethodBase(ABC):
 
         def _map_shape(symbol: str) -> str:
             if shape_dict is not None:
-                for k in shape_dict.keys():
+                for k in shape_dict:
                     if shape_dict[k] is not None and symbol in shape_dict[k]:
                         return k
             return "other"
@@ -211,8 +209,7 @@ class MethodBase(ABC):
             pval_thresh: float = None,
             s_curve: bool = False,
         ) -> str:
-            """
-            Map genes to categorize based on log2fc and pvalue.
+            """Map genes to categorize based on log2fc and pvalue.
 
             These categories are used for coloring the dots.
             Used when no color_dict is passed, sets up/down/nonsignificant.
@@ -229,14 +226,13 @@ class MethodBase(ABC):
                     return "Down"
                 else:
                     return "not DE"
+            # Standard condition for Up or Down categorization
+            elif log2fc > log2fc_thresh and nlog10 > pval_thresh:
+                return "Up"
+            elif log2fc < -log2fc_thresh and nlog10 > pval_thresh:
+                return "Down"
             else:
-                # Standard condition for Up or Down categorization
-                if log2fc > log2fc_thresh and nlog10 > pval_thresh:
-                    return "Up"
-                elif log2fc < -log2fc_thresh and nlog10 > pval_thresh:
-                    return "Down"
-                else:
-                    return "not DE"
+                return "not DE"
 
         def _map_genes_categories_highlight(
             row: pd.Series,
@@ -247,8 +243,7 @@ class MethodBase(ABC):
             s_curve: bool = False,
             symbol_col: str = None,
         ) -> str:
-            """
-            Map genes to categorize based on log2fc and pvalue.
+            """Map genes to categorize based on log2fc and pvalue.
 
             These categories are used for coloring the dots.
             Used when color_dict is passed, sets DE / not DE for background and user supplied highlight genes.
@@ -258,7 +253,7 @@ class MethodBase(ABC):
             symbol = row[symbol_col]
 
             if color_dict is not None:
-                for k in color_dict.keys():
+                for k in color_dict:
                     if symbol in color_dict[k]:
                         return k
 
@@ -489,7 +484,7 @@ class MethodBase(ABC):
         return None
 
     @_doc_params(common_plot_args=doc_common_plot_args)
-    def plot_paired(
+    def plot_paired(  # pragma: no cover # noqa: D417
         self,
         adata: ad.AnnData,
         results_df: pd.DataFrame,
@@ -581,14 +576,9 @@ class MethodBase(ABC):
                 adata, target_col=groupby, groups_col=pairedby, layer_key=layer, mode="sum", min_cells=1, min_counts=1
             )
 
-        if layer is not None:
-            X = adata.layers[layer]
-        else:
-            X = adata.X
-        try:
+        X = adata.layers[layer] if layer is not None else adata.X
+        with contextlib.suppress(AttributeError):
             X = X.toarray()
-        except AttributeError:
-            pass
 
         groupby_cols = [pairedby, groupby]
         df = adata.obs.loc[:, groupby_cols].join(pd.DataFrame(X, index=adata.obs_names, columns=var_names))
@@ -682,7 +672,7 @@ class MethodBase(ABC):
         return None
 
     @_doc_params(common_plot_args=doc_common_plot_args)
-    def plot_fold_change(
+    def plot_fold_change(  # pragma: no cover # noqa: D417
         self,
         results_df: pd.DataFrame,
         *,
@@ -763,7 +753,7 @@ class MethodBase(ABC):
         return None
 
     @_doc_params(common_plot_args=doc_common_plot_args)
-    def plot_multicomparison_fc(
+    def plot_multicomparison_fc(  # pragma: no cover # noqa: D417
         self,
         results_df: pd.DataFrame,
         *,
@@ -1013,7 +1003,7 @@ class LinearModelBase(MethodBase):
             )
         return self.formulaic_contrasts.cond(**kwargs)
 
-    def contrast(self, *args, **kwargs):
+    def contrast(self, *args, **kwargs):  # noqa: D417
         """Build a simple contrast for pairwise comparisons.
 
         Args:

pertpy/tools/_differential_gene_expression/_checks.py

@@ -16,9 +16,8 @@ def check_is_numeric_matrix(array: np.ndarray | spmatrix) -> None:
     if issparse(array):
         if np.any(~np.isfinite(array.data)):
             raise ValueError("Counts cannot contain negative, NaN or Inf values.")
-    else:
-        if np.any(~np.isfinite(array)):
-            raise ValueError("Counts cannot contain negative, NaN or Inf values.")
+    elif np.any(~np.isfinite(array)):
+        raise ValueError("Counts cannot contain negative, NaN or Inf values.")
 
 
 def check_is_integer_matrix(array: np.ndarray | spmatrix, tolerance: float = 1e-6) -> None:
@@ -34,8 +33,7 @@ def check_is_integer_matrix(array: np.ndarray | spmatrix, tolerance: float = 1e-
     if issparse(array):
         if not array.data.dtype.kind == "i" and not np.all(np.abs(array.data - np.round(array.data)) < tolerance):
             raise ValueError("Non-zero elements of the matrix must be close to integer values.")
-    else:
-        if not array.dtype.kind == "i" and not np.all(np.abs(array - np.round(array)) < tolerance):
-            raise ValueError("Matrix must be a count matrix.")
+    elif array.dtype.kind != "i" and not np.all(np.abs(array - np.round(array)) < tolerance):
+        raise ValueError("Matrix must be a count matrix.")
     if (array < 0).sum() > 0:
         raise ValueError("Non-zero elements of the matrix must be positive.")

pertpy/tools/_differential_gene_expression/_dge_comparison.py

@@ -36,16 +36,15 @@ class DGEEVAL:
             if not de_key1 or not de_key2:
                 raise ValueError("Both `de_key1` and `de_key2` must be provided together if using `adata`.")
 
-        else:  # use dfs
-            if de_df1 is None or de_df2 is None:
-                raise ValueError("Both `de_df1` and `de_df2` must be provided together if using DataFrames.")
+        elif de_df1 is None or de_df2 is None:
+            raise ValueError("Both `de_df1` and `de_df2` must be provided together if using DataFrames.")
 
         if de_key1:
             if not adata:
                 raise ValueError("`adata` should be provided with `de_key1` and `de_key2`. ")
-            assert all(
-                k in adata.uns for k in [de_key1, de_key2]
-            ), "Provided `de_key1` and `de_key2` must exist in `adata.uns`."
+            assert all(k in adata.uns for k in [de_key1, de_key2]), (
+                "Provided `de_key1` and `de_key2` must exist in `adata.uns`."
+            )
             vars = adata.var_names
 
         if de_df1 is not None:

pertpy/tools/_differential_gene_expression/_edger.py

@@ -10,7 +10,7 @@ from ._checks import check_is_integer_matrix
 
 
 class EdgeR(LinearModelBase):
-    """Differential expression test using EdgeR"""
+    """Differential expression test using EdgeR."""
 
     def _check_counts(self):
         check_is_integer_matrix(self.data)
@@ -39,17 +39,13 @@ class EdgeR(LinearModelBase):
             edger = importr("edgeR")
         except ImportError as e:
             raise ImportError(
-                "edgeR requires a valid R installation with the following packages:\n"
-                "edgeR, BiocParallel, RhpcBLASctl"
+                "edgeR requires a valid R installation with the following packages:\nedgeR, BiocParallel, RhpcBLASctl"
             ) from e
 
         # Convert dataframe
         with localconverter(get_conversion() + numpy2ri.converter):
             expr = self.adata.X if self.layer is None else self.adata.layers[self.layer]
-            if issparse(expr):
-                expr = expr.T.toarray()
-            else:
-                expr = expr.T
+            expr = expr.T.toarray() if issparse(expr) else expr.T
 
         with localconverter(get_conversion() + pandas2ri.converter):
             expr_r = ro.conversion.py2rpy(pd.DataFrame(expr, index=self.adata.var_names, columns=self.adata.obs_names))
@@ -72,8 +68,8 @@ class EdgeR(LinearModelBase):
         ro.globalenv["fit"] = fit
         self.fit = fit
 
-    def _test_single_contrast(self, contrast: Sequence[float], **kwargs) -> pd.DataFrame:
-        """Conduct test for each contrast and return a data frame
+    def _test_single_contrast(self, contrast: Sequence[float], **kwargs) -> pd.DataFrame:  # noqa: D417
+        """Conduct test for each contrast and return a data frame.
 
         Args:
             contrast: numpy array of integars indicating contrast i.e. [-1, 0, 1, 0, 0]
@@ -100,7 +96,7 @@ class EdgeR(LinearModelBase):
             importr("edgeR")
         except ImportError:
             raise ImportError(
-                "edgeR requires a valid R installation with the following packages: " "edgeR, BiocParallel, RhpcBLASctl"
+                "edgeR requires a valid R installation with the following packages: edgeR, BiocParallel, RhpcBLASctl"
             ) from None
 
         # Convert vector to R, which drops a category like `self.design_matrix` to use the intercept for the left out.

pertpy/tools/_differential_gene_expression/_pydeseq2.py

@@ -16,7 +16,7 @@ from ._checks import check_is_integer_matrix
 
 
 class PyDESeq2(LinearModelBase):
-    """Differential expression test using a PyDESeq2"""
+    """Differential expression test using a PyDESeq2."""
 
     def __init__(
         self, adata: AnnData, design: str | ndarray, *, mask: str | None = None, layer: str | None = None, **kwargs

pertpy/tools/_differential_gene_expression/_simple_tests.py

@@ -1,4 +1,4 @@
-"""Simple tests such as t-test, wilcoxon"""
+"""Simple tests such as t-test, wilcoxon."""
 
 import warnings
 from abc import abstractmethod
@@ -10,7 +10,7 @@ import pandas as pd
 import scipy.stats
 import statsmodels
 from anndata import AnnData
-from pandas.core.api import DataFrame as DataFrame
+from pandas.core.api import DataFrame
 from scipy.sparse import diags, issparse
 from tqdm.auto import tqdm
 
@@ -152,7 +152,7 @@ class WilcoxonTest(SimpleComparisonBase):
 
 
 class TTest(SimpleComparisonBase):
-    """Perform a unpaired or paired T-test"""
+    """Perform a unpaired or paired T-test."""
 
     @staticmethod
     def _test(x0: np.ndarray, x1: np.ndarray, paired: bool, **kwargs) -> float:

pertpy/tools/_differential_gene_expression/_statsmodels.py

@@ -6,14 +6,14 @@ import statsmodels.api as sm
 from tqdm.auto import tqdm
 
 from ._base import LinearModelBase
-from ._checks import check_is_integer_matrix
+from ._checks import check_is_numeric_matrix
 
 
 class Statsmodels(LinearModelBase):
-    """Differential expression test using a statsmodels linear regression"""
+    """Differential expression test using a statsmodels linear regression."""
 
     def _check_counts(self):
-        check_is_integer_matrix(self.data)
+        check_is_numeric_matrix(self.data)
 
     def fit(
         self,
@@ -55,7 +55,10 @@ class Statsmodels(LinearModelBase):
                     "t_value": t_test.tvalue.item(),
                     "sd": t_test.sd.item(),
                     "log_fc": t_test.effect.item(),
-                    "adj_p_value": statsmodels.stats.multitest.fdrcorrection(np.array([t_test.pvalue]))[1].item(),
                 }
             )
-        return pd.DataFrame(res).sort_values("p_value")
+        return (
+            pd.DataFrame(res)
+            .sort_values("p_value")
+            .assign(adj_p_value=lambda x: statsmodels.stats.multitest.fdrcorrection(x["p_value"])[1])
+        )

pertpy/tools/_distances/_distance_tests.py

@@ -83,8 +83,7 @@ class DistanceTest:
         contrast: str,
         show_progressbar: bool = True,
     ) -> pd.DataFrame:
-        """Run a permutation test using the specified distance metric, testing
-        all groups of cells against a specified contrast group ("control").
+        """Run a permutation test using the specified distance metric, testing all groups of cells against a specified contrast group ("control").
 
         Args:
             adata: Annotated data matrix.