augur.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Augur - cell type prioritization prediction\n",
    "\n",
    "This is a short tutorial demonstrating Augur, a Python implementation of the [Augur R package](https://github.com/neurorestore/Augur) based on Skinnider, M.A., Squair, J.W., Kathe, C. et al. [Cell type prioritization in single-cell data](https://doi.org/10.1038/s41587-020-0605-1). Nat Biotechnol 39, 30–34 (2021). \n",
    "\n",
    "Augur aims to rank or prioritize cell types according to their response to experimental perturbations given high dimensional single-cell sequencing data. The basic idea is that in the space of molecular measurements cells reacting heavily to induced perturbations are more easily saperated into perturbed and unperturbed than cell types with little or no response. This separability is quantified by measuring how well experimental labels (eg. treatment and control) can be predicted within each cell type. Augur trains a machine learning model predicting experimental labels for each cell type in multiple cross validation runs and then prioritizes cell type response according to metric scores measuring the accuracy of the model. For categorical data the area under the curve is the default metric and for numerical data the concordance correlation coefficient is used as a proxy for how accurate the model is which in turn approximates perturbation response. \n",
    "\n",
    "The following tutorial runs through a simple analysis with Augur using a simulated dataset. This dataset consists of 600 cells, distributed evenly between three populations (cell types A, B, and C). Each of these cell types has approximately half of its cells in one of two conditions, treatment and control. The cell types also have different numbers of genes differentially expressed (DE) in response to the treatment. Cell type A has approximately 5% of genes DE in response to the treatment, while cell type B has 25% of its genes DE and cell type C has 50% of genes DE. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import warnings\n",
    "\n",
    "warnings.filterwarnings(\"ignore\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import scanpy as sc\n",
    "import pertpy as pt"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "First we import the data that we want to work with. This can either be an anndata object, a dataframe containing cell type labels as well as conditions for each cell or data contained in a dataframe with corresponding meta data containing cell type labels and conditions.\n",
    "\n",
    "Then, we create an Augur object from pertpy based on our estimator of interest to measure how predictable the perturbation labels for each cell type in the dataset are. Choose `random_forest_classifier` or `logistic_regression_classifier` for categorical data and `random_forest_regressor` for numerical data. Additional parameters can be added throught the `Params` class.\n",
    "\n",
    "Next, load this data into an anndata object adding dummie variables of labels and renaming cell type and label observation columns to `cell_type` and `label`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "adata = pt.dt.sc_sim_augur()\n",
    "\n",
    "ag_rfc = pt.tl.Augur(\"random_forest_classifier\")\n",
    "loaded_data = ag_rfc.load(adata)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "AnnData object with n_obs × n_vars = 600 × 15697\n",
       "    obs: 'label', 'cell_type', 'y_'\n",
       "    var: 'name'"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "loaded_data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Then we run Augur with the function `predict` and look at the results. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "5403dec69e3f44cfae95b307a8d37533",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Output()"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"></pre>\n"
      ],
      "text/plain": []
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                  CellTypeA  CellTypeB  CellTypeC\n",
      "mean_augur_score   0.598299   0.867948   0.918435\n",
      "mean_auc           0.598299   0.867948   0.918435\n",
      "mean_accuracy      0.542674   0.735971   0.783040\n",
      "mean_precision     0.539211   0.782898   0.790340\n",
      "mean_f1            0.419785   0.697024   0.795674\n",
      "mean_recall        0.418889   0.704921   0.855873\n"
     ]
    }
   ],
   "source": [
    "h_adata, h_results = ag_rfc.predict(loaded_data, subsample_size=20, n_threads=4)\n",
    "\n",
    "print(h_results[\"summary_metrics\"])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The default way to select features is based on `scanpy.pp.highly_variable_genes`. The results are shown above. In addition there is also the option of selecting features with `select_variance` activated by setting `select_variance_features=True`, an implementation that is very close to the original R Augur implementation."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #808000; text-decoration-color: #808000; font-weight: bold\">Set smaller span value in the case of a `segmentation fault` error.</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1;33mSet smaller span value in the case of a `segmentation fault` error.\u001b[0m\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #808000; text-decoration-color: #808000; font-weight: bold\">Set larger span in case of svddc or other near singularities error.</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1;33mSet larger span in case of svddc or other near singularities error.\u001b[0m\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7baee568218f4c7498b14d1e68237997",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Output()"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"></pre>\n"
      ],
      "text/plain": []
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "v_adata, v_results = ag_rfc.predict(\n",
    "    loaded_data, subsample_size=20, select_variance_features=True, n_threads=4\n",
    ")\n",
    "\n",
    "# to visualize the results\n",
    "lollipop = pt.pl.ag.lollipop(v_results)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Here we visualize the cell ranking and the corresponding augur scores using `select variance` feature selection in a lollipop plot. In this case the ranking it the same but the values themselves differ between the two methods. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To compare the two feature selection methods they can be plotted together in a scatterplot. Here you can see that CellTypeB has slightly higher values when using the method `select_highly_variable` compared to `select_variance`, where as CellTypeC has the same augur score."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "scatter = pt.pl.ag.scatterplot(v_results, h_results)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The corresponding `mean_augur_score` is also saved in `result_adata.obs` and can be plotted in a UMAP."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sc.pp.neighbors(adata, use_rep=\"X\")\n",
    "sc.tl.umap(adata)\n",
    "sc.pl.umap(adata=v_adata, color=\"augur_score\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Feature Importances\n",
    "\n",
    "The results object also returns feature importances. In the case of a random forest estimator the feature importances built into sci-kit learn were used as feature importances. For the logistic regression the agresti method was used (as in the R library Augur). This means the mean was subtracted from the coefficient values and then divided by the standard deviation. (See [blog post](https://think-lab.github.io/d/205/))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "important_features = pt.pl.ag.important_features(v_results)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Differential Prioritization\n",
    "\n",
    "Augur is also able to perform differential prioritization and executes a permutation test to identify cell types with statistically significant differences in AUC between two different rounds of cell type prioritization (eg. response to drugs A and B, compared to untreated control). \n",
    "\n",
    "In the following part, single-cell prefrontal cotrex data from adult mice under cocaine self-administration from  Bhattacherjee A, Djekidel MN, Chen R, Chen W, Tuesta LM, Zhang Y. [Cell type-specific transcriptional programs in mouse prefrontal cortex during adolescence and addiction](https://doi.org/10.1038/s41467-019-12054-3). Nat Commun. 2019 Sep 13. will be used. Adult mice were subject to cocaine self-administration, samples were collected a three time points: Maintance, 48h after cocaine withdrawal and 15 days after cocaine withdrawal.\n",
    "\n",
    "In the following data, `withdraw_15d_Cocaine` and `withdraw_48h_Cocaine` will be compared with respect to the difference to `Maintenance_Cocaine`. Each variation is run once in default mode and once in permute mode."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "# Load data and create classifier\n",
    "bhattacherjee_adata = pt.dt.bhattacherjee()\n",
    "ag_rfc = pt.tl.Augur(\"random_forest_classifier\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For this we first run Augur on `Maintenance_Cocaine` and `withdraw_15d_Cocaine` in `augur_mode`=`default`  and `augur_mode`=`permute` mode. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Filtering samples with Maintenance_Cocaine and withdraw_15d_Cocaine labels.\n",
       "</pre>\n"
      ],
      "text/plain": [
       "Filtering samples with Maintenance_Cocaine and withdraw_15d_Cocaine labels.\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "359a3c86a2054c60a0951e4ecc093949",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Output()"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #808000; text-decoration-color: #808000; font-weight: bold\">Data not normalized. Normalizing now using scanpy log1p normalize.</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1;33mData not normalized. Normalizing now using scanpy log1p normalize.\u001b[0m\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #808000; text-decoration-color: #808000; font-weight: bold\">Data not normalized. Normalizing now using scanpy log1p normalize.</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1;33mData not normalized. Normalizing now using scanpy log1p normalize.\u001b[0m\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #808000; text-decoration-color: #808000; font-weight: bold\">Data not normalized. Normalizing now using scanpy log1p normalize.</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1;33mData not normalized. Normalizing now using scanpy log1p normalize.\u001b[0m\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #808000; text-decoration-color: #808000; font-weight: bold\">Data not normalized. Normalizing now using scanpy log1p normalize.</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1;33mData not normalized. Normalizing now using scanpy log1p normalize.\u001b[0m\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #808000; text-decoration-color: #808000; font-weight: bold\">Data not normalized. Normalizing now using scanpy log1p normalize.</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1;33mData not normalized. Normalizing now using scanpy log1p normalize.\u001b[0m\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #808000; text-decoration-color: #808000; font-weight: bold\">Data not normalized. Normalizing now using scanpy log1p normalize.</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1;33mData not normalized. Normalizing now using scanpy log1p normalize.\u001b[0m\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #808000; text-decoration-color: #808000; font-weight: bold\">Data not normalized. Normalizing now using scanpy log1p normalize.</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1;33mData not normalized. Normalizing now using scanpy log1p normalize.\u001b[0m\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #808000; text-decoration-color: #808000; font-weight: bold\">Data not normalized. Normalizing now using scanpy log1p normalize.</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1;33mData not normalized. Normalizing now using scanpy log1p normalize.\u001b[0m\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"></pre>\n"
      ],
      "text/plain": []
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Oligo         0.916088\n",
      "Astro         0.912823\n",
      "Microglia     0.907143\n",
      "OPC           0.900397\n",
      "Endo          0.780590\n",
      "Excitatory    0.688617\n",
      "NF Oligo      0.678730\n",
      "Inhibitory    0.660023\n",
      "Name: mean_augur_score, dtype: float64\n"
     ]
    }
   ],
   "source": [
    "# default\n",
    "bhattacherjee_15 = ag_rfc.load(\n",
    "    bhattacherjee_adata, condition_label=\"Maintenance_Cocaine\", treatment_label=\"withdraw_15d_Cocaine\"\n",
    ")\n",
    "\n",
    "bhattacherjee_adata_15, bhattacherjee_results_15 = ag_rfc.predict(\n",
    "    bhattacherjee_15, random_state=None, n_threads=4\n",
    ")\n",
    "\n",
    "print(bhattacherjee_results_15[\"summary_metrics\"].loc[\"mean_augur_score\"].sort_values(ascending=False))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c79eeb548cdf4f4bb73acc5241483c3d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Output()"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #808000; text-decoration-color: #808000; font-weight: bold\">Data not normalized. Normalizing now using scanpy log1p normalize.</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1;33mData not normalized. Normalizing now using scanpy log1p normalize.\u001b[0m\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #808000; text-decoration-color: #808000; font-weight: bold\">Data not normalized. Normalizing now using scanpy log1p normalize.</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1;33mData not normalized. Normalizing now using scanpy log1p normalize.\u001b[0m\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #808000; text-decoration-color: #808000; font-weight: bold\">Data not normalized. Normalizing now using scanpy log1p normalize.</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1;33mData not normalized. Normalizing now using scanpy log1p normalize.\u001b[0m\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #808000; text-decoration-color: #808000; font-weight: bold\">Data not normalized. Normalizing now using scanpy log1p normalize.</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1;33mData not normalized. Normalizing now using scanpy log1p normalize.\u001b[0m\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #808000; text-decoration-color: #808000; font-weight: bold\">Data not normalized. Normalizing now using scanpy log1p normalize.</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1;33mData not normalized. Normalizing now using scanpy log1p normalize.\u001b[0m\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #808000; text-decoration-color: #808000; font-weight: bold\">Data not normalized. Normalizing now using scanpy log1p normalize.</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1;33mData not normalized. Normalizing now using scanpy log1p normalize.\u001b[0m\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #808000; text-decoration-color: #808000; font-weight: bold\">Data not normalized. Normalizing now using scanpy log1p normalize.</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1;33mData not normalized. Normalizing now using scanpy log1p normalize.\u001b[0m\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #808000; text-decoration-color: #808000; font-weight: bold\">Data not normalized. Normalizing now using scanpy log1p normalize.</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1;33mData not normalized. Normalizing now using scanpy log1p normalize.\u001b[0m\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"></pre>\n"
      ],
      "text/plain": []
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# permute\n",
    "bhattacherjee_adata_15_permute, bhattacherjee_results_15_permute = ag_rfc.predict(\n",
    "    bhattacherjee_15, augur_mode=\"permute\", n_subsamples=100, random_state=None, n_threads=4\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now lets do the same looking at `Maintenance_Cocaine` and `withdraw_48h_Cocaine`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Filtering samples with Maintenance_Cocaine and withdraw_48h_Cocaine labels.\n",
       "</pre>\n"
      ],
      "text/plain": [
       "Filtering samples with Maintenance_Cocaine and withdraw_48h_Cocaine labels.\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "8604b7aa957f4b5493751fe0ba7e95f7",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Output()"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #808000; text-decoration-color: #808000; font-weight: bold\">Data not normalized. Normalizing now using scanpy log1p normalize.</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1;33mData not normalized. Normalizing now using scanpy log1p normalize.\u001b[0m\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #808000; text-decoration-color: #808000; font-weight: bold\">Data not normalized. Normalizing now using scanpy log1p normalize.</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1;33mData not normalized. Normalizing now using scanpy log1p normalize.\u001b[0m\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #808000; text-decoration-color: #808000; font-weight: bold\">Data not normalized. Normalizing now using scanpy log1p normalize.</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1;33mData not normalized. Normalizing now using scanpy log1p normalize.\u001b[0m\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #808000; text-decoration-color: #808000; font-weight: bold\">Data not normalized. Normalizing now using scanpy log1p normalize.</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1;33mData not normalized. Normalizing now using scanpy log1p normalize.\u001b[0m\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #808000; text-decoration-color: #808000; font-weight: bold\">Data not normalized. Normalizing now using scanpy log1p normalize.</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1;33mData not normalized. Normalizing now using scanpy log1p normalize.\u001b[0m\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #808000; text-decoration-color: #808000; font-weight: bold\">Data not normalized. Normalizing now using scanpy log1p normalize.</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1;33mData not normalized. Normalizing now using scanpy log1p normalize.\u001b[0m\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #808000; text-decoration-color: #808000; font-weight: bold\">Data not normalized. Normalizing now using scanpy log1p normalize.</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1;33mData not normalized. Normalizing now using scanpy log1p normalize.\u001b[0m\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #808000; text-decoration-color: #808000; font-weight: bold\">Data not normalized. Normalizing now using scanpy log1p normalize.</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1;33mData not normalized. Normalizing now using scanpy log1p normalize.\u001b[0m\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"></pre>\n"
      ],
      "text/plain": []
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Inhibitory    0.675272\n",
      "Astro         0.664728\n",
      "OPC           0.628141\n",
      "Microglia     0.614116\n",
      "Endo          0.603107\n",
      "Oligo         0.589728\n",
      "NF Oligo      0.587596\n",
      "Excitatory    0.559229\n",
      "Name: mean_augur_score, dtype: float64\n"
     ]
    }
   ],
   "source": [
    "# default\n",
    "bhattacherjee_48 = ag_rfc.load(\n",
    "    bhattacherjee_adata, condition_label=\"Maintenance_Cocaine\", treatment_label=\"withdraw_48h_Cocaine\"\n",
    ")\n",
    "\n",
    "bhattacherjee_adata_48, bhattacherjee_results_48 = ag_rfc.predict(\n",
    "    bhattacherjee_48, random_state=None, n_threads=4\n",
    ")\n",
    "\n",
    "print(bhattacherjee_results_48[\"summary_metrics\"].loc[\"mean_augur_score\"].sort_values(ascending=False))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "9590d7299d694aa5895e5163b3c4f3a1",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Output()"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #808000; text-decoration-color: #808000; font-weight: bold\">Data not normalized. Normalizing now using scanpy log1p normalize.</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1;33mData not normalized. Normalizing now using scanpy log1p normalize.\u001b[0m\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #808000; text-decoration-color: #808000; font-weight: bold\">Data not normalized. Normalizing now using scanpy log1p normalize.</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1;33mData not normalized. Normalizing now using scanpy log1p normalize.\u001b[0m\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #808000; text-decoration-color: #808000; font-weight: bold\">Data not normalized. Normalizing now using scanpy log1p normalize.</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1;33mData not normalized. Normalizing now using scanpy log1p normalize.\u001b[0m\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #808000; text-decoration-color: #808000; font-weight: bold\">Data not normalized. Normalizing now using scanpy log1p normalize.</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1;33mData not normalized. Normalizing now using scanpy log1p normalize.\u001b[0m\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #808000; text-decoration-color: #808000; font-weight: bold\">Data not normalized. Normalizing now using scanpy log1p normalize.</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1;33mData not normalized. Normalizing now using scanpy log1p normalize.\u001b[0m\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #808000; text-decoration-color: #808000; font-weight: bold\">Data not normalized. Normalizing now using scanpy log1p normalize.</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1;33mData not normalized. Normalizing now using scanpy log1p normalize.\u001b[0m\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #808000; text-decoration-color: #808000; font-weight: bold\">Data not normalized. Normalizing now using scanpy log1p normalize.</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1;33mData not normalized. Normalizing now using scanpy log1p normalize.\u001b[0m\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #808000; text-decoration-color: #808000; font-weight: bold\">Data not normalized. Normalizing now using scanpy log1p normalize.</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1;33mData not normalized. Normalizing now using scanpy log1p normalize.\u001b[0m\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #800000; text-decoration-color: #800000; font-weight: bold\">Skipping NF Oligo cell type - </span><span style=\"color: #800000; text-decoration-color: #800000; font-weight: bold\">79</span><span style=\"color: #800000; text-decoration-color: #800000; font-weight: bold\"> samples is less than min_cells </span><span style=\"color: #800000; text-decoration-color: #800000; font-weight: bold\">100</span><span style=\"color: #800000; text-decoration-color: #800000; font-weight: bold\">.</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1;31mSkipping NF Oligo cell type - \u001b[0m\u001b[1;31m79\u001b[0m\u001b[1;31m samples is less than min_cells \u001b[0m\u001b[1;31m100\u001b[0m\u001b[1;31m.\u001b[0m\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"></pre>\n"
      ],
      "text/plain": []
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# permute\n",
    "bhattacherjee_adata_48_permute, bhattacherjee_results_48_permute = ag_rfc.predict(\n",
    "    bhattacherjee_48, augur_mode=\"permute\", n_subsamples=100, random_state=None, n_threads=4\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Lets also take a look at the augur scores of the two versions in a scatterplot. The diagonal line is the identity function. If the values were the same they would be on the line. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "scatter = pt.pl.ag.scatterplot(bhattacherjee_results_15, bhattacherjee_results_48)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To figure out which cell type was most affected in comparing `withdraw_48h_Cocaine` and `withdraw_15d_Cocaine` we can run differential prioritization."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "pvals = ag_rfc.predict_differential_prioritization(\n",
    "    augur_results1=bhattacherjee_results_15,\n",
    "    augur_results2=bhattacherjee_results_48,\n",
    "    permuted_results1=bhattacherjee_results_15_permute,\n",
    "    permuted_results2=bhattacherjee_results_48_permute,\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>cell_type</th>\n",
       "      <th>mean_augur_score1</th>\n",
       "      <th>mean_augur_score2</th>\n",
       "      <th>delta_augur</th>\n",
       "      <th>b</th>\n",
       "      <th>m</th>\n",
       "      <th>z</th>\n",
       "      <th>pval</th>\n",
       "      <th>padj</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Microglia</td>\n",
       "      <td>0.907143</td>\n",
       "      <td>0.614116</td>\n",
       "      <td>-0.293027</td>\n",
       "      <td>1000</td>\n",
       "      <td>1000</td>\n",
       "      <td>-20.449195</td>\n",
       "      <td>0.001998</td>\n",
       "      <td>0.002331</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Endo</td>\n",
       "      <td>0.780590</td>\n",
       "      <td>0.603107</td>\n",
       "      <td>-0.177483</td>\n",
       "      <td>1000</td>\n",
       "      <td>1000</td>\n",
       "      <td>-8.876841</td>\n",
       "      <td>0.001998</td>\n",
       "      <td>0.002331</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Astro</td>\n",
       "      <td>0.912823</td>\n",
       "      <td>0.664728</td>\n",
       "      <td>-0.248095</td>\n",
       "      <td>1000</td>\n",
       "      <td>1000</td>\n",
       "      <td>-16.247874</td>\n",
       "      <td>0.001998</td>\n",
       "      <td>0.002331</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>OPC</td>\n",
       "      <td>0.900397</td>\n",
       "      <td>0.628141</td>\n",
       "      <td>-0.272256</td>\n",
       "      <td>788</td>\n",
       "      <td>1000</td>\n",
       "      <td>-18.572921</td>\n",
       "      <td>0.425574</td>\n",
       "      <td>0.425574</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Inhibitory</td>\n",
       "      <td>0.660023</td>\n",
       "      <td>0.675272</td>\n",
       "      <td>0.015249</td>\n",
       "      <td>1000</td>\n",
       "      <td>1000</td>\n",
       "      <td>3.218067</td>\n",
       "      <td>0.001998</td>\n",
       "      <td>0.002331</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Excitatory</td>\n",
       "      <td>0.688617</td>\n",
       "      <td>0.559229</td>\n",
       "      <td>-0.129388</td>\n",
       "      <td>1000</td>\n",
       "      <td>1000</td>\n",
       "      <td>-9.034865</td>\n",
       "      <td>0.001998</td>\n",
       "      <td>0.002331</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Oligo</td>\n",
       "      <td>0.916088</td>\n",
       "      <td>0.589728</td>\n",
       "      <td>-0.326361</td>\n",
       "      <td>1000</td>\n",
       "      <td>1000</td>\n",
       "      <td>-19.445162</td>\n",
       "      <td>0.001998</td>\n",
       "      <td>0.002331</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    cell_type  mean_augur_score1  mean_augur_score2  delta_augur     b     m  \\\n",
       "0   Microglia           0.907143           0.614116    -0.293027  1000  1000   \n",
       "1        Endo           0.780590           0.603107    -0.177483  1000  1000   \n",
       "2       Astro           0.912823           0.664728    -0.248095  1000  1000   \n",
       "3         OPC           0.900397           0.628141    -0.272256   788  1000   \n",
       "4  Inhibitory           0.660023           0.675272     0.015249  1000  1000   \n",
       "5  Excitatory           0.688617           0.559229    -0.129388  1000  1000   \n",
       "6       Oligo           0.916088           0.589728    -0.326361  1000  1000   \n",
       "\n",
       "           z      pval      padj  \n",
       "0 -20.449195  0.001998  0.002331  \n",
       "1  -8.876841  0.001998  0.002331  \n",
       "2 -16.247874  0.001998  0.002331  \n",
       "3 -18.572921  0.425574  0.425574  \n",
       "4   3.218067  0.001998  0.002331  \n",
       "5  -9.034865  0.001998  0.002331  \n",
       "6 -19.445162  0.001998  0.002331  "
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pvals"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The p-value, following the R Augur implementation is calculated using b, the number of times permuted values are larger than original values and m, the number of permutations run. Since b is the same for all cells but OPC, the p-value is the same for these as well."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "diff = pt.pl.ag.dp_scatter(pvals)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In this case the cell type Microglia has the lowest z-score. When comparing the impact of `withdraw_48h_Cocaine` and `withdraw_15d_Cocaine`, this cell type was most perturbed. "
   ]
  }
 ],
 "metadata": {
  "interpreter": {
   "hash": "a24c21ce9f55a051e125ae55b4380093d8f4a9785cb7fd33812a3a228a0c8b4c"
  },
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}