docs/source/Modeling_options_and_result_analysis.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true,
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "# Modeling options and result analysis in scCODA\n",
    "\n",
    "This tutorial notebook serves as an extension to the general tutorial and presents ways to alternate the model and perform more in-depth result analysis and diagnostics.\n",
    "We will focus on:\n",
    "\n",
    "- Modifications of the model formula and reference cell type to perform different modeling tasks\n",
    "- Inference methods available in scCODA\n",
    "- Advanced interpretation and analysis of results\n",
    "- Alternative differential abundance testing using all references\n",
    "\n",
    "We will again analyze the small intestinal epithelium data of mice from [Haber et al., 2017](https://www.nature.com/articles/nature24489).\n",
    "First, we read in the data and perform the same preprocessing steps as in the general tutorial:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "outputs": [],
   "source": [
    "# Setup\n",
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\")\n",
    "\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import arviz as az\n",
    "\n",
    "from sccoda.util import comp_ana as mod\n",
    "from sccoda.util import cell_composition_data as dat\n",
    "from sccoda.util import data_visualization as viz\n"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Entire dataset: AnnData object with n_obs × n_vars = 10 × 8\n",
      "    obs: 'Mouse', 'Condition'\n",
      "Salmonella dataset: AnnData object with n_obs × n_vars = 6 × 8\n",
      "    obs: 'Mouse', 'Condition'\n"
     ]
    },
    {
     "data": {
      "text/plain": "<Figure size 600x400 with 1 Axes>",
      "image/png": "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\n"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Load data\n",
    "cell_counts = pd.read_csv(\"../data/haber_counts.csv\")\n",
    "\n",
    "# Convert data to anndata object\n",
    "data_all = dat.from_pandas(cell_counts, covariate_columns=[\"Mouse\"])\n",
    "\n",
    "# Extract condition from mouse name and add it as an extra column to the covariates\n",
    "data_all.obs[\"Condition\"] = data_all.obs[\"Mouse\"].str.replace(r\"_[0-9]\", \"\")\n",
    "print(f\"Entire dataset: {data_all}\")\n",
    "\n",
    "# Select control and salmonella data\n",
    "data_salm = data_all[data_all.obs[\"Condition\"].isin([\"Control\", \"Salm\"])].copy()\n",
    "print(f\"Salmonella dataset: {data_salm}\")\n",
    "\n",
    "viz.boxplots(data_all, feature_name=\"Condition\")\n",
    "plt.show()\n"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "markdown",
   "source": [
    "## Tweaking the model formula and reference cell type\n",
    "\n",
    "First, we take a closer look at how changing the `formula` parameter of the scCODA model influences the results.\n",
    "Internally, the formula string is converted into a linear model-like design matrix via [patsy](https://patsy.readthedocs.io/en/latest/),\n",
    "which has a similar syntax to the `lm` function in the R language.\n",
    "\n",
    "### Multi-level categories\n",
    "\n",
    "Patsy allows us to automatically handle categorical covariates, even with multiple levels.\n",
    "For example, we can model the effect of all three diseases at once:"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%% md\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2021-11-28 19:15:21.670317: I tensorflow/core/platform/cpu_feature_guard.cc:142] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations:  AVX2 FMA\n",
      "To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From /Users/johannes.ostner/opt/anaconda3/envs/scCODA_3/lib/python3.8/site-packages/tensorflow/python/ops/array_ops.py:5043: calling gather (from tensorflow.python.ops.array_ops) with validate_indices is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "The `validate_indices` argument has no effect. Indices are always validated on CPU and never validated on GPU.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2021-11-28 19:15:23.943628: I tensorflow/compiler/mlir/mlir_graph_optimization_pass.cc:176] None of the MLIR Optimization Passes are enabled (registered 2)\n",
      "  0%|          | 0/20000 [00:00<?, ?it/s]2021-11-28 19:15:25.649198: I tensorflow/compiler/xla/service/service.cc:169] XLA service 0x7fd21d1c63a0 initialized for platform Host (this does not guarantee that XLA will be used). Devices:\n",
      "2021-11-28 19:15:25.649348: I tensorflow/compiler/xla/service/service.cc:177]   StreamExecutor device (0): Host, Default Version\n",
      "2021-11-28 19:15:28.022894: I tensorflow/compiler/jit/xla_compilation_cache.cc:337] Compiled cluster using XLA!  This line is logged at most once for the lifetime of the process.\n",
      "100%|██████████| 20000/20000 [01:38<00:00, 203.36it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MCMC sampling finished. (123.099 sec)\n",
      "Acceptance rate: 78.2%\n",
      "Compositional Analysis summary:\n",
      "\n",
      "Data: 10 samples, 8 cell types\n",
      "Reference index: 0\n",
      "Formula: Condition\n",
      "\n",
      "Intercepts:\n",
      "                       Final Parameter  Expected Sample\n",
      "Cell Type                                              \n",
      "Endocrine                        0.986        46.634633\n",
      "Enterocyte                       1.905       116.902874\n",
      "Enterocyte.Progenitor            2.348       182.061302\n",
      "Goblet                           1.474        75.970374\n",
      "Stem                             2.427       197.027528\n",
      "TA                               1.878       113.788726\n",
      "TA.Early                         2.534       219.278684\n",
      "Tuft                             0.626        32.535879\n",
      "\n",
      "\n",
      "Effects:\n",
      "                                                 Final Parameter  \\\n",
      "Covariate                 Cell Type                                \n",
      "Condition[T.H.poly.Day10] Endocrine                     0.000000   \n",
      "                          Enterocyte                    0.000000   \n",
      "                          Enterocyte.Progenitor         0.000000   \n",
      "                          Goblet                        0.000000   \n",
      "                          Stem                          0.000000   \n",
      "                          TA                            0.000000   \n",
      "                          TA.Early                      0.000000   \n",
      "                          Tuft                          0.000000   \n",
      "Condition[T.H.poly.Day3]  Endocrine                     0.000000   \n",
      "                          Enterocyte                    0.000000   \n",
      "                          Enterocyte.Progenitor         0.000000   \n",
      "                          Goblet                        0.000000   \n",
      "                          Stem                          0.000000   \n",
      "                          TA                            0.000000   \n",
      "                          TA.Early                      0.000000   \n",
      "                          Tuft                          0.000000   \n",
      "Condition[T.Salm]         Endocrine                     0.000000   \n",
      "                          Enterocyte                    1.554932   \n",
      "                          Enterocyte.Progenitor         0.000000   \n",
      "                          Goblet                        0.000000   \n",
      "                          Stem                          0.000000   \n",
      "                          TA                            0.000000   \n",
      "                          TA.Early                      0.000000   \n",
      "                          Tuft                          0.000000   \n",
      "\n",
      "                                                 Expected Sample  \\\n",
      "Covariate                 Cell Type                                \n",
      "Condition[T.H.poly.Day10] Endocrine                    46.634633   \n",
      "                          Enterocyte                  116.902874   \n",
      "                          Enterocyte.Progenitor       182.061302   \n",
      "                          Goblet                       75.970374   \n",
      "                          Stem                        197.027528   \n",
      "                          TA                          113.788726   \n",
      "                          TA.Early                    219.278684   \n",
      "                          Tuft                         32.535879   \n",
      "Condition[T.H.poly.Day3]  Endocrine                    46.634633   \n",
      "                          Enterocyte                  116.902874   \n",
      "                          Enterocyte.Progenitor       182.061302   \n",
      "                          Goblet                       75.970374   \n",
      "                          Stem                        197.027528   \n",
      "                          TA                          113.788726   \n",
      "                          TA.Early                    219.278684   \n",
      "                          Tuft                         32.535879   \n",
      "Condition[T.Salm]         Endocrine                    32.304092   \n",
      "                          Enterocyte                  383.418044   \n",
      "                          Enterocyte.Progenitor       126.114963   \n",
      "                          Goblet                       52.625137   \n",
      "                          Stem                        136.482158   \n",
      "                          TA                           78.822138   \n",
      "                          TA.Early                    151.895669   \n",
      "                          Tuft                         22.537800   \n",
      "\n",
      "                                                 log2-fold change  \n",
      "Covariate                 Cell Type                                \n",
      "Condition[T.H.poly.Day10] Endocrine                      0.000000  \n",
      "                          Enterocyte                     0.000000  \n",
      "                          Enterocyte.Progenitor          0.000000  \n",
      "                          Goblet                         0.000000  \n",
      "                          Stem                           0.000000  \n",
      "                          TA                             0.000000  \n",
      "                          TA.Early                       0.000000  \n",
      "                          Tuft                           0.000000  \n",
      "Condition[T.H.poly.Day3]  Endocrine                      0.000000  \n",
      "                          Enterocyte                     0.000000  \n",
      "                          Enterocyte.Progenitor          0.000000  \n",
      "                          Goblet                         0.000000  \n",
      "                          Stem                           0.000000  \n",
      "                          TA                             0.000000  \n",
      "                          TA.Early                       0.000000  \n",
      "                          Tuft                           0.000000  \n",
      "Condition[T.Salm]         Endocrine                     -0.529685  \n",
      "                          Enterocyte                     1.713608  \n",
      "                          Enterocyte.Progenitor         -0.529685  \n",
      "                          Goblet                        -0.529685  \n",
      "                          Stem                          -0.529685  \n",
      "                          TA                            -0.529685  \n",
      "                          TA.Early                      -0.529685  \n",
      "                          Tuft                          -0.529685  \n"
     ]
    }
   ],
   "source": [
    "# model all three diseases at once\n",
    "model_all = mod.CompositionalAnalysis(data_all, formula=\"Condition\", reference_cell_type=\"Endocrine\")\n",
    "all_results = model_all.sample_hmc()\n",
    "all_results.summary()"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "markdown",
   "source": [
    "### Different reference levels\n",
    "\n",
    "Per default, categorical variables are encoded via full-rank treatment coding. Hereby, the value of the first sample in the dataset is used as the default (control) category.\n",
    "We can select the default level by changing the model formula to `\"C(<CovariateName>, Treatment('<ReferenceLevelName>'))\"`:\n",
    "\n",
    "For example, we can switch the salmonella model to test diseased versus healthy samples, which switches the sign of the only credible effect (Enterocytes)."
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%% md\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 20000/20000 [01:02<00:00, 317.86it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MCMC sampling finished. (82.254 sec)\n",
      "Acceptance rate: 49.8%\n",
      "Compositional Analysis summary:\n",
      "\n",
      "Data: 6 samples, 8 cell types\n",
      "Reference index: 3\n",
      "Formula: C(Condition, Treatment('Salm'))\n",
      "\n",
      "Intercepts:\n",
      "                       Final Parameter  Expected Sample\n",
      "Cell Type                                              \n",
      "Endocrine                        1.269        28.559849\n",
      "Enterocyte                       3.708       327.340808\n",
      "Enterocyte.Progenitor            2.566       104.480647\n",
      "Goblet                           1.777        47.465439\n",
      "Stem                             2.615       109.727702\n",
      "TA                               2.048        62.240247\n",
      "TA.Early                         2.877       142.594061\n",
      "Tuft                             0.450        12.591247\n",
      "\n",
      "\n",
      "Effects:\n",
      "                                                                  Final Parameter  \\\n",
      "Covariate                                  Cell Type                                \n",
      "C(Condition, Treatment('Salm'))[T.Control] Endocrine                     0.000000   \n",
      "                                           Enterocyte                   -1.365487   \n",
      "                                           Enterocyte.Progenitor         0.000000   \n",
      "                                           Goblet                        0.000000   \n",
      "                                           Stem                          0.000000   \n",
      "                                           TA                            0.000000   \n",
      "                                           TA.Early                      0.000000   \n",
      "                                           Tuft                          0.000000   \n",
      "\n",
      "                                                                  Expected Sample  \\\n",
      "Covariate                                  Cell Type                                \n",
      "C(Condition, Treatment('Salm'))[T.Control] Endocrine                    40.336384   \n",
      "                                           Enterocyte                  118.009650   \n",
      "                                           Enterocyte.Progenitor       147.562807   \n",
      "                                           Goblet                       67.037616   \n",
      "                                           Stem                        154.973463   \n",
      "                                           TA                           87.904755   \n",
      "                                           TA.Early                    201.392129   \n",
      "                                           Tuft                         17.783195   \n",
      "\n",
      "                                                                  log2-fold change  \n",
      "Covariate                                  Cell Type                                \n",
      "C(Condition, Treatment('Salm'))[T.Control] Endocrine                      0.498093  \n",
      "                                           Enterocyte                    -1.471889  \n",
      "                                           Enterocyte.Progenitor          0.498093  \n",
      "                                           Goblet                         0.498093  \n",
      "                                           Stem                           0.498093  \n",
      "                                           TA                             0.498093  \n",
      "                                           TA.Early                       0.498093  \n",
      "                                           Tuft                           0.498093  \n"
     ]
    }
   ],
   "source": [
    "# Set salmonella infection as \"default\" category\n",
    "\n",
    "model_salm_switch_cond = mod.CompositionalAnalysis(data_salm, formula=\"C(Condition, Treatment('Salm'))\", reference_cell_type=\"Goblet\")\n",
    "switch_results = model_salm_switch_cond.sample_hmc()\n",
    "switch_results.summary()"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "markdown",
   "source": [
    "### Switching the reference cell type\n",
    "\n",
    "Compositional analysis generally does not allow statements on absolute abundance changes, but only in relation to a reference category,\n",
    "which is assumed to be unchanged in absolute abundance.\n",
    "The reference cell type fixes this category in scCODA. Thus, an interpretation of scCODA's effects should always be formulated like:\n",
    "\n",
    "\"Using cell type xy as a reference, cell types (a, b, c) were found to credibly change in abundance\"\n",
    "\n",
    "Switching the reference cell type might thus produce different results.\n",
    "For example, if we choose a different cell type as the reference (such as Enterocytes in the salmonella infection data),\n",
    "scCODA can find other credible effects on the other cell types."
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%% md\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 20000/20000 [01:02<00:00, 321.43it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MCMC sampling finished. (80.035 sec)\n",
      "Acceptance rate: 54.2%\n",
      "Compositional Analysis summary:\n",
      "\n",
      "Data: 6 samples, 8 cell types\n",
      "Reference index: 1\n",
      "Formula: Condition\n",
      "\n",
      "Intercepts:\n",
      "                       Final Parameter  Expected Sample\n",
      "Cell Type                                              \n",
      "Endocrine                        0.563        34.927176\n",
      "Enterocyte                       2.088       160.495387\n",
      "Enterocyte.Progenitor            1.870       129.058424\n",
      "Goblet                           1.100        59.755737\n",
      "Stem                             2.095       161.622796\n",
      "TA                               1.475        86.944084\n",
      "TA.Early                         2.220       183.142621\n",
      "Tuft                            -0.043        19.053774\n",
      "\n",
      "\n",
      "Effects:\n",
      "                                         Final Parameter  Expected Sample  \\\n",
      "Covariate         Cell Type                                                 \n",
      "Condition[T.Salm] Endocrine                          0.0        34.927176   \n",
      "                  Enterocyte                         0.0       160.495387   \n",
      "                  Enterocyte.Progenitor              0.0       129.058424   \n",
      "                  Goblet                             0.0        59.755737   \n",
      "                  Stem                               0.0       161.622796   \n",
      "                  TA                                 0.0        86.944084   \n",
      "                  TA.Early                           0.0       183.142621   \n",
      "                  Tuft                               0.0        19.053774   \n",
      "\n",
      "                                         log2-fold change  \n",
      "Covariate         Cell Type                                \n",
      "Condition[T.Salm] Endocrine                           0.0  \n",
      "                  Enterocyte                          0.0  \n",
      "                  Enterocyte.Progenitor               0.0  \n",
      "                  Goblet                              0.0  \n",
      "                  Stem                                0.0  \n",
      "                  TA                                  0.0  \n",
      "                  TA.Early                            0.0  \n",
      "                  Tuft                                0.0  \n"
     ]
    }
   ],
   "source": [
    "model_salm_ref = mod.CompositionalAnalysis(data_salm, formula=\"Condition\", reference_cell_type=\"Enterocyte\")\n",
    "reference_results = model_salm_ref.sample_hmc()\n",
    "reference_results.summary()"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "markdown",
   "source": [
    "## Inference algorithms in scCODA\n",
    "\n",
    "Currently, scCODA performs parameter inference via Markov-chain Monte Carlo (MCMC) methods.\n",
    "There are three different MCMC sampling methods available for scCODA:\n",
    "\n",
    "- Hamiltonian Monte Carlo (HMC) sampling: `sample_hmc()`\n",
    "- HMC sampling with Dual-averaging step size adaptation ([Nesterov, 2009](https://link.springer.com/article/10.1007/s10107-007-0149-x)): `sample_hmc_da()`\n",
    "- No-U-Turn sampling ([Hoffman and Gelman, 2014](https://jmlr.org/papers/volume15/hoffman14a/hoffman14a.pdf)): `sample_nuts()`\n",
    "\n",
    "Generally, it is recommended to use the standard HMC sampling. Other methods, such as variational inference, are in consideration.\n",
    "\n",
    "For all MCMC sampling methods, properties such as the MCMC chain length and the number of burn-in samples are directly adjustable."
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%% md\n"
    }
   }
  },
  {
   "cell_type": "markdown",
   "source": [
    "## Result analysis and diagnostics\n",
    "\n",
    "The \"getting started\" tutorial explains how to do interpret the basic output of scCODA.\n",
    "To follow this up, we now take a look at how MCMC diagnostics and more advanced result analysis in scCODA can be performed.\n",
    "\n",
    "For this section, we again use the model of salmonella infection versus control group, with a reference cell type of Goblet cells."
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%% md\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 20000/20000 [01:03<00:00, 314.08it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MCMC sampling finished. (80.894 sec)\n",
      "Acceptance rate: 55.8%\n"
     ]
    }
   ],
   "source": [
    "model_salm = mod.CompositionalAnalysis(data_salm, formula=\"Condition\", reference_cell_type=\"Goblet\")\n",
    "salm_results = model_salm.sample_hmc(num_results=20000)"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "markdown",
   "source": [
    "### Extended model summary\n",
    "\n",
    "`result.summary_extended()` gives us, apart from the properties already explained in the basic tutorial, more information about the posterior inferred by the model.\n",
    "The extended summary also includes some information on the MCMC sampling procedure (chain length, burn-in, acceptance rate, duration).\n",
    "\n",
    "For both effects and intercepts, we also get the standard deviation (SD) and high density interval endpoints of the posterior density of the generated Markov chain.\n",
    "\n",
    "The effects summary also includes the spike-and-slab inclusion probability for each effect, i.e. the share of MCMC samples, for which this effect was not set to 0 by the spike-and-slab prior.\n",
    "A threshold on this value serves as the deciding factor whether an effect is considered statistically credible\n",
    "\n",
    "We can also use the summary tables from `summary_extended()` as pandas DataFrames to tweak them further.\n",
    "They are also accessible as `result.intercept_df` and `result.effect_df`, respectively.\n",
    "Furthermore, the tables a direct result of the `summary()` function in [arviz](https://arviz-devs.github.io/arviz/index.html) and support all its functionality.\n",
    "This means that we can, for example, change the credible interval:"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%% md\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Compositional Analysis summary (extended):\n",
      "\n",
      "Data: 6 samples, 8 cell types\n",
      "Reference index: 3\n",
      "Formula: Condition\n",
      "Spike-and-slab threshold: 1.000\n",
      "\n",
      "MCMC Sampling: Sampled 20000 chain states (5000 burnin samples) in 80.894 sec. Acceptance rate: 55.8%\n",
      "\n",
      "Intercepts:\n",
      "                       Final Parameter  HDI 5%  HDI 95%     SD  \\\n",
      "Cell Type                                                        \n",
      "Endocrine                        1.105   0.548    1.750  0.369   \n",
      "Enterocyte                       2.330   1.842    2.844  0.312   \n",
      "Enterocyte.Progenitor            2.517   2.055    3.071  0.310   \n",
      "Goblet                           1.749   1.253    2.277  0.316   \n",
      "Stem                             2.710   2.229    3.235  0.306   \n",
      "TA                               2.114   1.576    2.658  0.330   \n",
      "TA.Early                         2.858   2.361    3.341  0.297   \n",
      "Tuft                             0.433  -0.210    1.065  0.394   \n",
      "\n",
      "                       Expected Sample  \n",
      "Cell Type                               \n",
      "Endocrine                    34.199308  \n",
      "Enterocyte                  116.420125  \n",
      "Enterocyte.Progenitor       140.359279  \n",
      "Goblet                       65.118287  \n",
      "Stem                        170.239355  \n",
      "TA                           93.803805  \n",
      "TA.Early                    197.394727  \n",
      "Tuft                         17.465114  \n",
      "\n",
      "\n",
      "Effects:\n",
      "                                         Final Parameter  HDI 5%  HDI 95%  \\\n",
      "Covariate         Cell Type                                                 \n",
      "Condition[T.Salm] Endocrine                     0.000000  -0.432    1.033   \n",
      "                  Enterocyte                    1.347912   0.912    1.768   \n",
      "                  Enterocyte.Progenitor         0.000000  -0.255    0.581   \n",
      "                  Goblet                        0.000000   0.000    0.000   \n",
      "                  Stem                          0.000000  -0.720    0.129   \n",
      "                  TA                            0.000000  -0.789    0.249   \n",
      "                  TA.Early                      0.000000  -0.331    0.422   \n",
      "                  Tuft                          0.000000  -0.772    0.866   \n",
      "\n",
      "                                            SD  Inclusion probability  \\\n",
      "Covariate         Cell Type                                             \n",
      "Condition[T.Salm] Endocrine              0.348               0.506933   \n",
      "                  Enterocyte             0.265               1.000000   \n",
      "                  Enterocyte.Progenitor  0.158               0.334800   \n",
      "                  Goblet                 0.000               0.000000   \n",
      "                  Stem                   0.214               0.398667   \n",
      "                  TA                     0.247               0.418933   \n",
      "                  TA.Early               0.119               0.267467   \n",
      "                  Tuft                   0.325               0.417733   \n",
      "\n",
      "                                         Expected Sample  log2-fold change  \n",
      "Covariate         Cell Type                                                 \n",
      "Condition[T.Salm] Endocrine                    24.475708         -0.482617  \n",
      "                  Enterocyte                  320.727867          1.462009  \n",
      "                  Enterocyte.Progenitor       100.452112         -0.482617  \n",
      "                  Goblet                       46.603755         -0.482617  \n",
      "                  Stem                        121.836638         -0.482617  \n",
      "                  TA                           67.133362         -0.482617  \n",
      "                  TA.Early                    141.271153         -0.482617  \n",
      "                  Tuft                         12.499406         -0.482617  \n"
     ]
    }
   ],
   "source": [
    "salm_results.summary_extended(hdi_prob=0.9)\n"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "markdown",
   "source": [
    "\n",
    "### Diagnostics and plotting\n",
    "\n",
    "Similarly to the summary dataframes being compatible with [arviz](https://arviz-devs.github.io/arviz/index.html),\n",
    "the result class itself is an extension of arviz's `Inference Data` class. This means that we can use all its MCMC diagnostic and plotting functionality.\n",
    "As an example, looking at the MCMC trace plots and kernel density estimates, we see that they are indicative of a well sampled MCMC chain:\n",
    "\n",
    "Note: Due to the spike-and-slab priors, the `beta` parameters have many values at 0, which looks like a convergence issue, but is actually not.\n",
    "\n",
    "**Caution: Trying to plot a kernel density estimate for an effect on the reference cell type results in an error, since it is constant at 0 for the entire chain.\n",
    "To avoid this, add** `coords={\"cell_type\": salm_results.posterior.coords[\"cell_type_nb\"]}` **as an argument to `az.plot_trace`, which causes the plots for the reference cell type to be skipped.**"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%% md\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "outputs": [
    {
     "data": {
      "text/plain": "<Figure size 864x288 with 4 Axes>",
      "image/png": "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\n"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "az.plot_trace(\n",
    "    salm_results,\n",
    "    divergences=False,\n",
    "    var_names=[\"alpha\", \"beta\"],\n",
    "    coords={\"cell_type\": salm_results.posterior.coords[\"cell_type_nb\"]},\n",
    ")\n",
    "plt.show()"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "markdown",
   "source": [
    "### Using all cell types as reference alternatively to reference selection\n",
    "\n",
    "scCODA uses a reference cell type that is considered to be unchanged over the experiment to guarantee the unique identifiability of results.\n",
    "If no such cell type is known beforehand, setting `reference_cell_type=\"automatic\"` will find a suited reference.\n",
    "Alternatively, it is possible to find credible effects on cell types that are mostly independent of the reference.\n",
    "By sequentially running scCODA and selecting each cell type as the reference once, we can then use a majority vote to find the cell types that were credibly changing more than half of the time.\n",
    "\n",
    "Below, an example code for this procedure on the Salmonella infection data shows that only Enterocytes were found to be credible more than half of the time.\n",
    "Indeed, they re credibly changing for every reference cell type except themselves.\n",
    "All other cell types were not found to change with any reference."
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%% md\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Reference: Endocrine\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 20000/20000 [01:03<00:00, 313.42it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:5 out of the last 5 calls to <function CompositionalModel.sampling.<locals>.sample_mcmc at 0x7fd1ed75e3a0> triggered tf.function retracing. Tracing is expensive and the excessive number of tracings could be due to (1) creating @tf.function repeatedly in a loop, (2) passing tensors with different shapes, (3) passing Python objects instead of tensors. For (1), please define your @tf.function outside of the loop. For (2), @tf.function has experimental_relax_shapes=True option that relaxes argument shapes that can avoid unnecessary retracing. For (3), please refer to https://www.tensorflow.org/guide/function#controlling_retracing and https://www.tensorflow.org/api_docs/python/tf/function for  more details.\n",
      "MCMC sampling finished. (81.026 sec)\n",
      "Acceptance rate: 50.1%\n",
      "Reference: Enterocyte\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 20000/20000 [01:04<00:00, 309.59it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:6 out of the last 6 calls to <function CompositionalModel.sampling.<locals>.sample_mcmc at 0x7fd1ed75e3a0> triggered tf.function retracing. Tracing is expensive and the excessive number of tracings could be due to (1) creating @tf.function repeatedly in a loop, (2) passing tensors with different shapes, (3) passing Python objects instead of tensors. For (1), please define your @tf.function outside of the loop. For (2), @tf.function has experimental_relax_shapes=True option that relaxes argument shapes that can avoid unnecessary retracing. For (3), please refer to https://www.tensorflow.org/guide/function#controlling_retracing and https://www.tensorflow.org/api_docs/python/tf/function for  more details.\n",
      "MCMC sampling finished. (85.399 sec)\n",
      "Acceptance rate: 56.5%\n",
      "Reference: Enterocyte.Progenitor\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 20000/20000 [01:08<00:00, 291.32it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MCMC sampling finished. (93.835 sec)\n",
      "Acceptance rate: 60.7%\n",
      "Reference: Goblet\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 20000/20000 [01:13<00:00, 271.64it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MCMC sampling finished. (90.710 sec)\n",
      "Acceptance rate: 58.2%\n",
      "Reference: Stem\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 20000/20000 [01:07<00:00, 295.02it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MCMC sampling finished. (86.994 sec)\n",
      "Acceptance rate: 51.8%\n",
      "Reference: TA\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 20000/20000 [01:22<00:00, 242.56it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MCMC sampling finished. (103.528 sec)\n",
      "Acceptance rate: 53.4%\n",
      "Reference: TA.Early\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 20000/20000 [01:21<00:00, 244.23it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MCMC sampling finished. (105.065 sec)\n",
      "Acceptance rate: 54.0%\n",
      "Reference: Tuft\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 20000/20000 [01:23<00:00, 240.64it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MCMC sampling finished. (104.704 sec)\n",
      "Acceptance rate: 49.7%\n"
     ]
    }
   ],
   "source": [
    "# Run scCODA with each cell type as the reference\n",
    "cell_types = data_salm.var.index\n",
    "results_cycle = pd.DataFrame(index=cell_types, columns=[\"times_credible\"]).fillna(0)\n",
    "\n",
    "for ct in cell_types:\n",
    "    print(f\"Reference: {ct}\")\n",
    "\n",
    "    # Run inference\n",
    "    model_temp = mod.CompositionalAnalysis(data_salm, formula=\"Condition\", reference_cell_type=ct)\n",
    "    temp_results = model_temp.sample_hmc(num_results=20000)\n",
    "\n",
    "    # Select credible effects\n",
    "    cred_eff = temp_results.credible_effects()\n",
    "    cred_eff.index = cred_eff.index.droplevel(level=0)\n",
    "\n",
    "    # add up credible effects\n",
    "    results_cycle[\"times_credible\"] += cred_eff.astype(\"int\")"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                       times_credible  pct_credible  is_credible\n",
      "Endocrine                           0         0.000        False\n",
      "Enterocyte                          7         0.875         True\n",
      "Enterocyte.Progenitor               0         0.000        False\n",
      "Goblet                              0         0.000        False\n",
      "Stem                                0         0.000        False\n",
      "TA                                  0         0.000        False\n",
      "TA.Early                            0         0.000        False\n",
      "Tuft                                0         0.000        False\n"
     ]
    }
   ],
   "source": [
    "# Calculate percentages\n",
    "results_cycle[\"pct_credible\"] = results_cycle[\"times_credible\"]/len(cell_types)\n",
    "results_cycle[\"is_credible\"] = results_cycle[\"pct_credible\"] > 0.5\n",
    "print(results_cycle)\n"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "outputs": [],
   "source": [],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  }
 ],
 "metadata": {
  "kernelspec": {
   "name": "python3",
   "language": "python",
   "display_name": "Python 3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}