ruff_mods

coomit

Author: Sergio Salas

notebooks/spatialdata_tutorials/.ipynb_checkpoints/4_quantify_exRNA-checkpoint.ipynb

@@ -23,7 +23,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 1,
+   "execution_count": null,
    "metadata": {},
    "outputs": [
     {
@@ -439,7 +439,7 @@
    "source": [
     "import pandas as pd\n",
     "\n",
-    "pd.crosstab(sdata[\"xrna_metadata\"].var[\"p_value_Poisson\"] < 0.00005, sdata[\"xrna_metadata\"].var[\"control_probe\"])"
+    "pd.crosstab(sdata['xrna_metadata'].var['p_value_Poisson']<0.00005,sdata['xrna_metadata'].var['control_probe'])"
    ]
   },
   {
@@ -524,7 +524,11 @@
     }
    ],
    "source": [
-    "troutpy.pl.metric_scatter(sdata, x_metric=\"proportion_of_colocalized\", y_metric=\"extracellular_proportion\", label_top_n_x=3, label_top_n_y=3)"
+
+    "troutpy.pl.metric_scatter(sdata,\n",
+    "x_metric=\"proportion_of_colocalized\",\n",
+    "y_metric=\"extracellular_proportion\",\n",
+    "label_top_n_x=3,label_top_n_y=3)"
    ]
   },
   {
@@ -719,9 +723,12 @@
     }
    ],
    "source": [
-    "troutpy.pl.global_distribution_from_source(\n",
-    "    sdata, cluster_key=\"kmeans_distribution\", feature_key=\"feature_name\", distance_key=\"distance\", n_bins=40, how=\"collapsed\"\n",
-    ")"
+    "troutpy.pl.global_distribution_from_source(sdata,\n",
+    "                               cluster_key= \"kmeans_distribution\",\n",
+    "                               feature_key = \"feature_name\",\n",
+    "                               distance_key= \"distance\",\n",
+    "                               n_bins= 40,\n",
+    "                               how= \"collapsed\")"
    ]
   },
   {
@@ -832,14 +839,10 @@
    ],
    "source": [
     "troutpy.pl.intra_extra_density(\n",
-    "    sdata,\n",
-    "    bottom_genes,\n",
-    "    layer=\"transcripts\",\n",
-    "    gene_key=\"feature_name\",\n",
-    "    coord_keys=[\"x\", \"y\"],\n",
-    "    intra_kde_kwargs={\"fill\": True, \"cmap\": \"Blues\", \"thresh\": 0.05, \"bw_adjust\": 0.2},\n",
-    "    extra_kde_kwargs={\"fill\": True, \"cmap\": \"Reds\", \"thresh\": 0.05, \"bw_adjust\": 0.2},\n",
-    "    figsize=(5, 7),\n",
+
+    "    sdata, bottom_genes, layer=\"transcripts\", gene_key=\"feature_name\", coord_keys=[\"x\", \"y\"],\n",
+    "    intra_kde_kwargs = {\"fill\": True, \"cmap\": \"Blues\", \"thresh\": 0.05,\"bw_adjust\":0.2},\n",
+    "    extra_kde_kwargs = {\"fill\": True, \"cmap\": \"Reds\", \"thresh\": 0.05,\"bw_adjust\":0.2},figsize=(5,7)\n",
     ")"
    ]
   },
@@ -861,14 +864,9 @@
    ],
    "source": [
     "troutpy.pl.intra_extra_density(\n",
-    "    sdata,\n",
-    "    top_genes,\n",
-    "    layer=\"transcripts\",\n",
-    "    gene_key=\"feature_name\",\n",
-    "    coord_keys=[\"x\", \"y\"],\n",
-    "    intra_kde_kwargs={\"fill\": True, \"cmap\": \"Blues\", \"thresh\": 0.05, \"bw_adjust\": 0.2},\n",
-    "    extra_kde_kwargs={\"fill\": True, \"cmap\": \"Reds\", \"thresh\": 0.05, \"bw_adjust\": 0.2},\n",
-    "    figsize=(5, 7),\n",
+    "    sdata, top_genes, layer=\"transcripts\", gene_key=\"feature_name\", coord_keys=[\"x\", \"y\"],\n",
+    "    intra_kde_kwargs = {\"fill\": True, \"cmap\": \"Blues\", \"thresh\": 0.05,\"bw_adjust\":0.2},\n",
+    "    extra_kde_kwargs = {\"fill\": True, \"cmap\": \"Reds\", \"thresh\": 0.05,\"bw_adjust\":0.2},figsize=(5,7)\n",
     ")"
    ]
   },
@@ -909,9 +907,10 @@
     }
    ],
    "source": [
-    "troutpy.pl.metric_scatter(\n",
-    "    sdata, x_metric=\"spatial_density_correlation\", y_metric=\"mean_displacement\", label_top_n_x=3, label_top_n_y=3, label_bottom_n_x=5\n",
-    ")"
+    "troutpy.pl.metric_scatter(sdata,\n",
+    "x_metric=\"spatial_density_correlation\",\n",
+    "y_metric=\"mean_displacement\",\n",
+    "label_top_n_x=3,label_top_n_y=3,label_bottom_n_x=5)"
    ]
   },
   {
@@ -929,6 +928,7 @@
    "source": [
     "import numpy as np\n",
     "\n",
+    "\n",
     "def compute_projection_score(sdata):\n",
     "    \"\"\"\n",
     "    Compute a segmentation score for each cell based on the expression of genes weighted by their intracellular proportion (1 - extracellular proportion).\n",
@@ -937,21 +937,17 @@
     "    ----------\n",
     "    sdata : dict\n",
     "        A spatialdata object with keys 'table' and 'xrna_metadata'.\n",
-    "        - sdata['table'] is an AnnData object containing expression data in layers['raw']\n",
-    "          and cell metadata in .obs.\n",
-    "        - sdata['xrna_metadata'].var is a DataFrame with gene names as the index and\n",
-    "          an 'extracellular proportion' column.\n",
     "\n",
     "    Returns\n",
     "    -------\n",
     "    sdata : dict\n",
     "        The same sdata object with a new column 'segmentation_score' in sdata['table'].obs.\n",
     "    \"\"\"\n",
     "    # Retrieve the AnnData object with cells in .obs and genes in .var\n",
-    "    adata = sdata[\"table\"]\n",
+    "    adata = sdata['table']\n",
     "\n",
     "    # Retrieve raw expression data; assume shape (n_cells, n_genes)\n",
-    "    raw_expr = adata.layers[\"raw\"]\n",
+    "    raw_expr = adata.layers['raw']\n",
     "\n",
     "    # If raw_expr is a sparse matrix, convert to a dense array\n",
     "    if hasattr(raw_expr, \"toarray\"):\n",
@@ -961,7 +957,7 @@
     "    genes = adata.var_names\n",
     "\n",
     "    # Retrieve gene metadata containing the extracellular proportions\n",
-    "    gene_meta = sdata[\"xrna_metadata\"].var\n",
+    "    gene_meta = sdata['xrna_metadata'].var\n",
     "\n",
     "    # Identify the genes common to both the expression data and the metadata\n",
     "    common_genes = gene_meta.index.intersection(genes)\n",
@@ -976,7 +972,7 @@
     "    # Reorder gene_meta so that it matches the ordering in the expression data.\n",
     "    # This assumes that the order of genes in adata.var_names is the desired order.\n",
     "    ordered_genes = [gene for gene in genes if gene in common_genes]\n",
-    "    gene_weights = gene_meta.loc[ordered_genes, \"extracellular_proportion\"]\n",
+    "    gene_weights = gene_meta.loc[ordered_genes, 'extracellular_proportion']\n",
     "\n",
     "    # Convert extracellular proportion to intracellular weight (1 - extracellular proportion)\n",
     "    intracellular_weights = 1 - gene_weights.values  # numpy array\n",
@@ -991,7 +987,7 @@
     "    score = np.divide(numerator, denominator, out=np.full_like(numerator, np.nan), where=denominator != 0)\n",
     "\n",
     "    # Store the score in the AnnData object under obs\n",
-    "    adata.obs[\"projection_score\"] = score\n",
+    "    adata.obs['projection_score'] = score\n",
     "\n",
     "#   return sdata\n",
     "\n",
@@ -2882,7 +2878,7 @@
    "source": [
     "import scanpy as sc\n",
     "\n",
-    "sc.pl.umap(sdata[\"table\"], color=[\"projection_score\", \"cell_type\", \"total_counts\"], vmax=\"p99.997\")"
+    "sc.pl.umap(sdata['table'],color=['projection_score','cell_type','total_counts'],vmax='p99.997')"
    ]
   },
   {
@@ -2891,39 +2887,36 @@
    "metadata": {},
    "outputs": [],
    "source": [
+    "\n",
     "def proportion_of_extracellularly_enriched_genes(sdata, threshold=0.5):\n",
-    "    \"\"\"For cell, compute the proportion of expressed genes that are extracellularly enriched, i.e. whose extracellular transcript proportion is above a given threshold.\n",
+    "    \"\"\"\n",
+    "    For each cell, compute the proportion of expressed genes that are extracellularly enriched, i.e. whose extracellular transcript proportion is above a given threshold.\n",
     "\n",
     "    Parameters\n",
     "    ----------\n",
     "    sdata : dict\n",
     "        A spatialdata object with keys 'table' and 'xrna_metadata'.\n",
-    "        - sdata['table'] is an AnnData object containing expression data in layers['raw']\n",
-    "          and cell metadata in .obs.\n",
-    "        - sdata['xrna_metadata'].var is a DataFrame with gene names as the index and\n",
-    "          an 'extracellular_proportion' column.\n",
     "\n",
-    "    threshold : float, default=0.5\n",
-    "        The minimum extracellular proportion required for a gene to be considered\n",
-    "        \"extracellularly enriched\".\n",
+    "    threshold : float\n",
+    "        The minimum extracellular proportion required for a gene to be considered \"extracellularly enriched\".\n",
     "\n",
     "    Returns\n",
     "    -------\n",
     "    sdata : dict\n",
-    "        The same sdata object with a new column 'extracellularly_enriched_proportion'\n",
-    "        in sdata['table'].obs, which contains the computed metric for each cell.\n",
+    "        The same sdata object with a new column 'extracellularly_enriched_proportion' in sdata['table'].obs, which contains the computed metric for each cell.\n",
+    "\n",
     "    \"\"\"\n",
     "    # Retrieve the AnnData object and raw expression data\n",
-    "    adata = sdata[\"table\"]\n",
-    "    raw_expr = adata.layers[\"raw\"]\n",
+    "    adata = sdata['table']\n",
+    "    raw_expr = adata.layers['raw']\n",
     "\n",
     "    # Convert to dense if necessary\n",
     "    if hasattr(raw_expr, \"toarray\"):\n",
     "        raw_expr = raw_expr.toarray()\n",
     "\n",
     "    # Get gene names from AnnData and the corresponding gene metadata\n",
     "    genes = adata.var_names\n",
-    "    gene_meta = sdata[\"xrna_metadata\"].var\n",
+    "    gene_meta = sdata['xrna_metadata'].var\n",
     "\n",
     "    # Find common genes between the expression data and metadata\n",
     "    common_genes = gene_meta.index.intersection(genes)\n",
@@ -2936,7 +2929,7 @@
     "\n",
     "    # Reorder the gene metadata to match the ordering in the expression data\n",
     "    ordered_genes = [gene for gene in genes if gene in common_genes]\n",
-    "    extracellular_props = gene_meta.loc[ordered_genes, \"extracellular_proportion\"].values\n",
+    "    extracellular_props = gene_meta.loc[ordered_genes, 'extracellular_proportion'].values\n",
     "\n",
     "    # Create a boolean mask for genes that are extracellularly enriched (above threshold)\n",
     "    enriched_mask = extracellular_props > threshold\n",
@@ -2951,7 +2944,12 @@
     "    enriched_count = (expressed & enriched_mask).sum(axis=1)\n",
     "\n",
     "    # Compute the proportion (handling potential division by zero)\n",
-    "    proportion = np.divide(enriched_count, expressed_count, out=np.full_like(enriched_count, np.nan, dtype=float), where=expressed_count != 0)\n",
+    "    proportion = np.divide(\n",
+    "        enriched_count,\n",
+    "        expressed_count,\n",
+    "        out=np.full_like(enriched_count, np.nan, dtype=float),\n",
+    "        where=expressed_count != 0\n",
+    "    )\n",
     "\n",
     "    # Store the computed metric in the AnnData object under .obs\n",
     "    adata.obs[\"extracellularly_enriched_proportion\"] = proportion"
@@ -3023,7 +3021,8 @@
     "import scanpy as sc\n",
     "\n",
     "# Ensure your AnnData object (sdata['table']) has a 'cell type' annotation in obs.\n",
-    "sc.pl.dotplot(sdata[\"table\"], extranscripts, groupby=\"cell_type\", standard_scale=\"var\", title=\"Extracellularly Enriched Genes by Cell Type\")"
+    "sc.pl.dotplot(sdata['table'], extranscripts, groupby='cell_type',\n",
+    "              standard_scale='var', title='Extracellularly Enriched Genes by Cell Type')\n"
    ]
   },
   {
@@ -3036,7 +3035,7 @@
  ],
  "metadata": {
   "kernelspec": {
-   "display_name": "exrna",
+   "display_name": "Python 3 (ipykernel)",
    "language": "python",
    "name": "python3"
   },
@@ -3050,9 +3049,9 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.10.15"
+   "version": "3.10.16"
   }
  },
  "nbformat": 4,
- "nbformat_minor": 2
+ "nbformat_minor": 4
 }