04_Tutorial_Downstream_Analysis_Python.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "cb792b1d-d41e-4280-afa5-a5780d622d17",
   "metadata": {},
   "outputs": [],
   "source": [
    "# all_slow\n",
    "# hide\n",
    "%load_ext autoreload\n",
    "%autoreload 2"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "36cc2299-0b20-435e-9b35-79fc0be6c0e3",
   "metadata": {},
   "source": [
    "# Downstream Analysis - Python"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b201d968-9137-41af-8bb3-2330de4dd9e7",
   "metadata": {},
   "source": [
    "**Note**: This tutorial was originally written as an Jupyter Notebook (ipynb). If you would like to run or edit the notebook, it can be found [here](https://github.com/morriso1/clonedetective/blob/master/04_Tutorial_Downstream_Analysis_Python.ipynb)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d7dc4c65-90c2-4fc4-869e-a30d4af3a54e",
   "metadata": {},
   "source": [
    "Now for the fun bit, wrangling the data and plotting our results! Below, we do this using pandas, matplotlib and statsmodels, and some convenience functions from `clonedetective.clone_analysis`."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "dc070a91-838e-4e08-82a1-c8ba0c30efc7",
   "metadata": {},
   "source": [
    "## Recap - background and experimental aims"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0c833e11-33c7-49b6-891b-981a79fd2459",
   "metadata": {},
   "source": [
    "Our original demo dataset consisted of 67 images of the *Drosophila* intestine from a MARCM genetic cell lineage experiment. GFP (C1 channel) marks genetic recombination events, and so clusters of GFP positive cells define \"clones\" of daughter cells that over time have grown out from a single parent cell. The remaining images include DAPI (C0 channel), which ubiquitously marks all cell nuclei; PDM1 (C2 channel), which labels EnteroCyte (EC) cells; and Prospero (C3 channel), which labels EnteroEndocrine cells.\n",
    "\n",
    "We aimed to investigate:\n",
    "1) Does gene of interest regulate **cell proliferation**?  \n",
    "2) Does our gene of interest control **cell differentiation**?   \n",
    "3) Does gene of interest effect **local tissue cell arrangement and structure**?\n",
    "\n",
    "We will tackle each of these experimental aims using data [previously generated by clonedetective](https://morriso1.github.io/clonedetective/Tutorial_Walkthrough.html)."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "76ee2b69-2553-48be-9419-d979bb02147d",
   "metadata": {},
   "source": [
    "## Loading data and updating column metadata"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0d19afd5-964a-466a-9f21-e5e0567f6d45",
   "metadata": {},
   "source": [
    "Reading in our previous results:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "4ff44905-e674-4d68-8ab9-e398e35a00f0",
   "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>int_img</th>\n",
       "      <th>label</th>\n",
       "      <th>GFP_clone</th>\n",
       "      <th>GFPneg_nc</th>\n",
       "      <th>GFPpos_nc</th>\n",
       "      <th>tot_nc</th>\n",
       "      <th>ECneg_nc</th>\n",
       "      <th>ECpos_nc</th>\n",
       "      <th>EEneg_nc</th>\n",
       "      <th>EEpos_nc</th>\n",
       "      <th>...</th>\n",
       "      <th>centroid-1</th>\n",
       "      <th>eccentricity</th>\n",
       "      <th>area_um2</th>\n",
       "      <th>GFP_clonepos</th>\n",
       "      <th>total_intensity</th>\n",
       "      <th>EC_pos</th>\n",
       "      <th>EE_pos</th>\n",
       "      <th>ECEE_pos</th>\n",
       "      <th>ISCorEB_pos</th>\n",
       "      <th>preEC_pos</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>a1g01p1</td>\n",
       "      <td>11</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>89.552116</td>\n",
       "      <td>0.440881</td>\n",
       "      <td>73.280625</td>\n",
       "      <td>False</td>\n",
       "      <td>2196457.0</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>a1g01p1</td>\n",
       "      <td>11</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>89.552116</td>\n",
       "      <td>0.440881</td>\n",
       "      <td>73.280625</td>\n",
       "      <td>False</td>\n",
       "      <td>52757.0</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>a1g01p1</td>\n",
       "      <td>11</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>89.552116</td>\n",
       "      <td>0.440881</td>\n",
       "      <td>73.280625</td>\n",
       "      <td>False</td>\n",
       "      <td>2262648.0</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>a1g01p1</td>\n",
       "      <td>11</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>89.552116</td>\n",
       "      <td>0.440881</td>\n",
       "      <td>73.280625</td>\n",
       "      <td>False</td>\n",
       "      <td>67331.0</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>a1g01p1</td>\n",
       "      <td>12</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>486.798467</td>\n",
       "      <td>0.465227</td>\n",
       "      <td>98.690625</td>\n",
       "      <td>False</td>\n",
       "      <td>4113516.0</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 32 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   int_img  label  GFP_clone  GFPneg_nc  GFPpos_nc  tot_nc  ECneg_nc  \\\n",
       "0  a1g01p1     11          0          2          2       4         1   \n",
       "1  a1g01p1     11          0          2          2       4         1   \n",
       "2  a1g01p1     11          0          2          2       4         1   \n",
       "3  a1g01p1     11          0          2          2       4         1   \n",
       "4  a1g01p1     12          0          4          0       4         1   \n",
       "\n",
       "   ECpos_nc  EEneg_nc  EEpos_nc  ...  centroid-1  eccentricity   area_um2  \\\n",
       "0         3         4         0  ...   89.552116      0.440881  73.280625   \n",
       "1         3         4         0  ...   89.552116      0.440881  73.280625   \n",
       "2         3         4         0  ...   89.552116      0.440881  73.280625   \n",
       "3         3         4         0  ...   89.552116      0.440881  73.280625   \n",
       "4         3         4         0  ...  486.798467      0.465227  98.690625   \n",
       "\n",
       "   GFP_clonepos  total_intensity  EC_pos EE_pos ECEE_pos ISCorEB_pos  \\\n",
       "0         False        2196457.0    True  False    False       False   \n",
       "1         False          52757.0    True  False    False       False   \n",
       "2         False        2262648.0    True  False    False       False   \n",
       "3         False          67331.0    True  False    False       False   \n",
       "4         False        4113516.0    True  False    False       False   \n",
       "\n",
       "   preEC_pos  \n",
       "0      False  \n",
       "1      False  \n",
       "2      False  \n",
       "3      False  \n",
       "4      False  \n",
       "\n",
       "[5 rows x 32 columns]"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "\n",
    "df = pd.read_csv(\"data/example_results.csv\")\n",
    "\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "59bc9244-85c1-4fe5-9e6c-2ee766ffbeed",
   "metadata": {},
   "source": [
    "To ease downstream wrangling and plotting, we next update some of the column metadata.\n",
    "\n",
    "For example, when plotting, it will be helpful to label whether a cell is inside a clone (IC) or outside a clone (OC). This information is already present within the `df['GFP_clonepos']` column that states if a cell is GFP positive. We, therefore, recode this column and rename it to `clone_status`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "420ec7f7-e539-4cdc-833e-06134956b813",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "\n",
    "df[\"GFP_clonepos\"] = np.where(df[\"GFP_clonepos\"], \"IC\", \"OC\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "0c5754a5-29d8-4f3d-9016-600cb7238e85",
   "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>int_img</th>\n",
       "      <th>label</th>\n",
       "      <th>GFP_clone</th>\n",
       "      <th>GFPneg_nc</th>\n",
       "      <th>GFPpos_nc</th>\n",
       "      <th>tot_nc</th>\n",
       "      <th>ECneg_nc</th>\n",
       "      <th>ECpos_nc</th>\n",
       "      <th>EEneg_nc</th>\n",
       "      <th>EEpos_nc</th>\n",
       "      <th>...</th>\n",
       "      <th>centroid-1</th>\n",
       "      <th>eccentricity</th>\n",
       "      <th>area_um2</th>\n",
       "      <th>clone_status</th>\n",
       "      <th>total_intensity</th>\n",
       "      <th>EC_pos</th>\n",
       "      <th>EE_pos</th>\n",
       "      <th>ECEE_pos</th>\n",
       "      <th>ISCorEB_pos</th>\n",
       "      <th>preEC_pos</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>a1g01p1</td>\n",
       "      <td>11</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>89.552116</td>\n",
       "      <td>0.440881</td>\n",
       "      <td>73.280625</td>\n",
       "      <td>OC</td>\n",
       "      <td>2196457.0</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>a1g01p1</td>\n",
       "      <td>11</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>89.552116</td>\n",
       "      <td>0.440881</td>\n",
       "      <td>73.280625</td>\n",
       "      <td>OC</td>\n",
       "      <td>52757.0</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>a1g01p1</td>\n",
       "      <td>11</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>89.552116</td>\n",
       "      <td>0.440881</td>\n",
       "      <td>73.280625</td>\n",
       "      <td>OC</td>\n",
       "      <td>2262648.0</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>a1g01p1</td>\n",
       "      <td>11</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>89.552116</td>\n",
       "      <td>0.440881</td>\n",
       "      <td>73.280625</td>\n",
       "      <td>OC</td>\n",
       "      <td>67331.0</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>a1g01p1</td>\n",
       "      <td>12</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>486.798467</td>\n",
       "      <td>0.465227</td>\n",
       "      <td>98.690625</td>\n",
       "      <td>OC</td>\n",
       "      <td>4113516.0</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 32 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   int_img  label  GFP_clone  GFPneg_nc  GFPpos_nc  tot_nc  ECneg_nc  \\\n",
       "0  a1g01p1     11          0          2          2       4         1   \n",
       "1  a1g01p1     11          0          2          2       4         1   \n",
       "2  a1g01p1     11          0          2          2       4         1   \n",
       "3  a1g01p1     11          0          2          2       4         1   \n",
       "4  a1g01p1     12          0          4          0       4         1   \n",
       "\n",
       "   ECpos_nc  EEneg_nc  EEpos_nc  ...  centroid-1  eccentricity   area_um2  \\\n",
       "0         3         4         0  ...   89.552116      0.440881  73.280625   \n",
       "1         3         4         0  ...   89.552116      0.440881  73.280625   \n",
       "2         3         4         0  ...   89.552116      0.440881  73.280625   \n",
       "3         3         4         0  ...   89.552116      0.440881  73.280625   \n",
       "4         3         4         0  ...  486.798467      0.465227  98.690625   \n",
       "\n",
       "   clone_status  total_intensity  EC_pos EE_pos ECEE_pos ISCorEB_pos  \\\n",
       "0            OC        2196457.0    True  False    False       False   \n",
       "1            OC          52757.0    True  False    False       False   \n",
       "2            OC        2262648.0    True  False    False       False   \n",
       "3            OC          67331.0    True  False    False       False   \n",
       "4            OC        4113516.0    True  False    False       False   \n",
       "\n",
       "   preEC_pos  \n",
       "0      False  \n",
       "1      False  \n",
       "2      False  \n",
       "3      False  \n",
       "4      False  \n",
       "\n",
       "[5 rows x 32 columns]"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = df.rename(columns={\"GFP_clonepos\": \"clone_status\"})\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2a1921f4-74b9-4142-8aa4-c58ff9ad7f7a",
   "metadata": {},
   "source": [
    "We also can extract metadata stored within the image names. We know that images are named 'a1' if from a wild-type (ctrl) animal, and 'a2' if from a mutant (mut) animal. Using this information, we can add an 'genotype' columns.  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "728dd6be-45cd-4d25-9b88-f03b55f2a343",
   "metadata": {},
   "outputs": [],
   "source": [
    "df[\"genotype\"] = np.where(df[\"int_img\"].str.match(r\"a1\"), \"ctrl\", \"mut\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8de41f85-464c-4212-abde-9f170e4fc9aa",
   "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>int_img</th>\n",
       "      <th>label</th>\n",
       "      <th>GFP_clone</th>\n",
       "      <th>GFPneg_nc</th>\n",
       "      <th>GFPpos_nc</th>\n",
       "      <th>tot_nc</th>\n",
       "      <th>ECneg_nc</th>\n",
       "      <th>ECpos_nc</th>\n",
       "      <th>EEneg_nc</th>\n",
       "      <th>EEpos_nc</th>\n",
       "      <th>...</th>\n",
       "      <th>eccentricity</th>\n",
       "      <th>area_um2</th>\n",
       "      <th>clone_status</th>\n",
       "      <th>total_intensity</th>\n",
       "      <th>EC_pos</th>\n",
       "      <th>EE_pos</th>\n",
       "      <th>ECEE_pos</th>\n",
       "      <th>ISCorEB_pos</th>\n",
       "      <th>preEC_pos</th>\n",
       "      <th>genotype</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>a1g01p1</td>\n",
       "      <td>11</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.440881</td>\n",
       "      <td>73.280625</td>\n",
       "      <td>OC</td>\n",
       "      <td>2196457.0</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>ctrl</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>a1g01p1</td>\n",
       "      <td>11</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.440881</td>\n",
       "      <td>73.280625</td>\n",
       "      <td>OC</td>\n",
       "      <td>52757.0</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>ctrl</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>a1g01p1</td>\n",
       "      <td>11</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.440881</td>\n",
       "      <td>73.280625</td>\n",
       "      <td>OC</td>\n",
       "      <td>2262648.0</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>ctrl</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>a1g01p1</td>\n",
       "      <td>11</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.440881</td>\n",
       "      <td>73.280625</td>\n",
       "      <td>OC</td>\n",
       "      <td>67331.0</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>ctrl</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>a1g01p1</td>\n",
       "      <td>12</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.465227</td>\n",
       "      <td>98.690625</td>\n",
       "      <td>OC</td>\n",
       "      <td>4113516.0</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>ctrl</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 33 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   int_img  label  GFP_clone  GFPneg_nc  GFPpos_nc  tot_nc  ECneg_nc  \\\n",
       "0  a1g01p1     11          0          2          2       4         1   \n",
       "1  a1g01p1     11          0          2          2       4         1   \n",
       "2  a1g01p1     11          0          2          2       4         1   \n",
       "3  a1g01p1     11          0          2          2       4         1   \n",
       "4  a1g01p1     12          0          4          0       4         1   \n",
       "\n",
       "   ECpos_nc  EEneg_nc  EEpos_nc  ...  eccentricity   area_um2  clone_status  \\\n",
       "0         3         4         0  ...      0.440881  73.280625            OC   \n",
       "1         3         4         0  ...      0.440881  73.280625            OC   \n",
       "2         3         4         0  ...      0.440881  73.280625            OC   \n",
       "3         3         4         0  ...      0.440881  73.280625            OC   \n",
       "4         3         4         0  ...      0.465227  98.690625            OC   \n",
       "\n",
       "   total_intensity  EC_pos  EE_pos ECEE_pos ISCorEB_pos preEC_pos  genotype  \n",
       "0        2196457.0    True   False    False       False     False      ctrl  \n",
       "1          52757.0    True   False    False       False     False      ctrl  \n",
       "2        2262648.0    True   False    False       False     False      ctrl  \n",
       "3          67331.0    True   False    False       False     False      ctrl  \n",
       "4        4113516.0    True   False    False       False     False      ctrl  \n",
       "\n",
       "[5 rows x 33 columns]"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "358f786b-1093-427a-960f-944abbecb7f6",
   "metadata": {},
   "source": [
    "## Cell Proliferation - number of cells per clone"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "13cb626a-73cd-400e-b171-f5211a8ba816",
   "metadata": {},
   "source": [
    "Our first aim is to determine whether there is a difference in the number of cells per clone in mutant vs control animals. `count_df` currently contains four rows for each cell - one for each of the four intensity image channels. Since we are presently interested in cell types and numbers, rather than the intensity values of individual cells, we can query count_df for a single intensity image channel. This ensures that `count_data` contains only a single row for each cell.\n",
    "\n",
    " For simplicity, we also filter out columns not required for counting the number and type of cells per clone."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "dd9387d9-b468-4546-bc93-0161342aecaf",
   "metadata": {},
   "outputs": [],
   "source": [
    "count_df = df.query('int_img_ch == \"C1\"').filter(\n",
    "    regex=r\"clone$|pos$|int_img|clone_status|genotype\"\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8a9fce9a-24ce-4ef4-bede-eaf06d928d53",
   "metadata": {},
   "source": [
    "To help with later calculations, we add a 'total' column:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e6715f36-6bea-423f-8352-6fbbf99a99d2",
   "metadata": {},
   "outputs": [],
   "source": [
    "count_df[\"total\"] = count_df.filter(regex=\"_pos\").sum(axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "304c9d14-a8a0-4ae0-9626-5452b47a97ae",
   "metadata": {},
   "source": [
    "And quickly simplify the cell type column names:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "79f86fed-f750-45da-928e-eab96ba24198",
   "metadata": {},
   "outputs": [],
   "source": [
    "import re\n",
    "\n",
    "count_df.columns = [re.sub(\"_pos\", \"\", col) for col in count_df.columns]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8b3b5c8f-c167-48c3-b45f-96fa8a6be609",
   "metadata": {},
   "source": [
    "Next, we groupby clones per image and sum to get the number of cells of each type per clone."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "7cdc60e6-e362-42e4-acd7-9e09b64750d9",
   "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>int_img</th>\n",
       "      <th>GFP_clone</th>\n",
       "      <th>clone_status</th>\n",
       "      <th>genotype</th>\n",
       "      <th>EC</th>\n",
       "      <th>EE</th>\n",
       "      <th>preEC</th>\n",
       "      <th>ECEE</th>\n",
       "      <th>ISCorEB</th>\n",
       "      <th>total</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>a1g01p1</td>\n",
       "      <td>0</td>\n",
       "      <td>OC</td>\n",
       "      <td>ctrl</td>\n",
       "      <td>70</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>98</td>\n",
       "      <td>175</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>a1g01p1</td>\n",
       "      <td>1</td>\n",
       "      <td>IC</td>\n",
       "      <td>ctrl</td>\n",
       "      <td>9</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>a1g01p1</td>\n",
       "      <td>2</td>\n",
       "      <td>IC</td>\n",
       "      <td>ctrl</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>a1g01p1</td>\n",
       "      <td>3</td>\n",
       "      <td>IC</td>\n",
       "      <td>ctrl</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>a1g01p1</td>\n",
       "      <td>4</td>\n",
       "      <td>IC</td>\n",
       "      <td>ctrl</td>\n",
       "      <td>12</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>21</td>\n",
       "      <td>34</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   int_img  GFP_clone clone_status genotype  EC  EE  preEC  ECEE  ISCorEB  \\\n",
       "0  a1g01p1          0           OC     ctrl  70   2      1     4       98   \n",
       "1  a1g01p1          1           IC     ctrl   9   0      0     0        5   \n",
       "2  a1g01p1          2           IC     ctrl   0   0      0     0        1   \n",
       "3  a1g01p1          3           IC     ctrl   1   0      0     0        3   \n",
       "4  a1g01p1          4           IC     ctrl  12   0      0     1       21   \n",
       "\n",
       "   total  \n",
       "0    175  \n",
       "1     14  \n",
       "2      1  \n",
       "3      4  \n",
       "4     34  "
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "count_df = count_df.groupby([\"int_img\", \"GFP_clone\"], as_index=False).agg(\n",
    "    {\n",
    "        \"clone_status\": \"first\",\n",
    "        \"genotype\": \"first\",\n",
    "        \"EC\": \"sum\",\n",
    "        \"EE\": \"sum\",\n",
    "        \"preEC\": \"sum\",\n",
    "        \"ECEE\": \"sum\",\n",
    "        \"ISCorEB\": \"sum\",\n",
    "        \"total\": \"sum\",\n",
    "    }\n",
    ")\n",
    "count_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "daeff346-6715-4fdb-bdbb-5f5a00f12336",
   "metadata": {},
   "source": [
    "We then groupby only the genotype column and generate a stacked bar plot displaying cell numbers and types per clone. Here, we only interested in plotting cells inside clones, and so we first query on 'clone_status':"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "cc387fe3-0e18-4d63-b65e-fe5b090d2ef8",
   "metadata": {},
   "outputs": [],
   "source": [
    "count_df_inside = count_df.query('clone_status == \"IC\"')\n",
    "\n",
    "inside_clone_mean = (\n",
    "    count_df_inside[[\"genotype\", \"ISCorEB\", \"preEC\", \"EC\", \"EE\", \"ECEE\"]]\n",
    "    .groupby(\"genotype\")\n",
    "    .mean()\n",
    ")\n",
    "\n",
    "inside_clone_sem = (\n",
    "    count_df_inside[[\"genotype\", \"ISCorEB\", \"preEC\", \"EC\", \"EE\", \"ECEE\"]]\n",
    "    .groupby(\"genotype\")\n",
    "    .sem()\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "0914fbd9-46c0-409e-852c-163be70a553a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 144x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "\n",
    "from clonedetective import clone_analysis as ca\n",
    "\n",
    "with sns.plotting_context(\"notebook\", font_scale=1.2):\n",
    "    ca.create_stack_bar_plot(\n",
    "        inside_clone_mean,\n",
    "        inside_clone_sem,\n",
    "        x_figSize=2,\n",
    "        y_figSize=4,\n",
    "        color_pal=[\"#008b68\", \"#539DC2\", \"#F57171\", \"#95a5a6\", \"#000000\"],\n",
    "        bar_width=0.8,\n",
    "        y_axis_limit=16,\n",
    "        y_label=\"Cells per clone\",\n",
    "    )"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "38f4190f-7bd6-4ece-9ea1-64d06e926e8b",
   "metadata": {},
   "source": [
    "From this plot we can see that 'mutant' clones are on average bigger than 'ctrl' clones. This suggests that mutation of our gene of interest may affect **cell proliferation**. Since we have count data, let's fit a negative binomial regression using the statsmodels library test and if this difference is significant:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "3eb52d72-707c-4894-b6a9-706d3bd9773b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>Generalized Linear Model Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>         <td>total</td>      <th>  No. Observations:  </th>  <td>   429</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                  <td>GLM</td>       <th>  Df Residuals:      </th>  <td>   427</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model Family:</th>    <td>NegativeBinomial</td> <th>  Df Model:          </th>  <td>     1</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Link Function:</th>          <td>log</td>       <th>  Scale:             </th> <td>  1.0000</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>                <td>IRLS</td>       <th>  Log-Likelihood:    </th> <td> -1297.0</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>            <td>Sat, 18 Sep 2021</td> <th>  Deviance:          </th> <td>  445.65</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                <td>19:15:46</td>     <th>  Pearson chi2:      </th>  <td>  577.</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Iterations:</th>          <td>6</td>        <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>     <td>nonrobust</td>    <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "         <td></td>            <th>coef</th>     <th>std err</th>      <th>z</th>      <th>P>|z|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Intercept</th>       <td>    2.5930</td> <td>    0.074</td> <td>   34.926</td> <td> 0.000</td> <td>    2.447</td> <td>    2.738</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>genotype[T.mut]</th> <td>   -1.1887</td> <td>    0.104</td> <td>  -11.420</td> <td> 0.000</td> <td>   -1.393</td> <td>   -0.985</td>\n",
       "</tr>\n",
       "</table>"
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                 Generalized Linear Model Regression Results                  \n",
       "==============================================================================\n",
       "Dep. Variable:                  total   No. Observations:                  429\n",
       "Model:                            GLM   Df Residuals:                      427\n",
       "Model Family:        NegativeBinomial   Df Model:                            1\n",
       "Link Function:                    log   Scale:                          1.0000\n",
       "Method:                          IRLS   Log-Likelihood:                -1297.0\n",
       "Date:                Sat, 18 Sep 2021   Deviance:                       445.65\n",
       "Time:                        19:15:46   Pearson chi2:                     577.\n",
       "No. Iterations:                     6                                         \n",
       "Covariance Type:            nonrobust                                         \n",
       "===================================================================================\n",
       "                      coef    std err          z      P>|z|      [0.025      0.975]\n",
       "-----------------------------------------------------------------------------------\n",
       "Intercept           2.5930      0.074     34.926      0.000       2.447       2.738\n",
       "genotype[T.mut]    -1.1887      0.104    -11.420      0.000      -1.393      -0.985\n",
       "===================================================================================\n",
       "\"\"\""
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import statsmodels.api as sm\n",
    "from statsmodels.formula.api import glm\n",
    "\n",
    "glm_nb = glm(\n",
    "    \"total ~ genotype\", data=count_df_inside, family=sm.families.NegativeBinomial()\n",
    ").fit()\n",
    "\n",
    "glm_nb.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "03105ab9-6f8a-4265-b0d0-44921dff23d5",
   "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>coef</th>\n",
       "      <th>std err</th>\n",
       "      <th>z</th>\n",
       "      <th>P&gt;|z|</th>\n",
       "      <th>Conf. Int. Low</th>\n",
       "      <th>Conf. Int. Upp.</th>\n",
       "      <th>pvalue-hs</th>\n",
       "      <th>reject-hs</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>mut-ctrl</th>\n",
       "      <td>-1.188662</td>\n",
       "      <td>0.104089</td>\n",
       "      <td>-11.419623</td>\n",
       "      <td>3.336787e-30</td>\n",
       "      <td>-1.392674</td>\n",
       "      <td>-0.984651</td>\n",
       "      <td>3.336787e-30</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              coef   std err          z         P>|z|  Conf. Int. Low  \\\n",
       "mut-ctrl -1.188662  0.104089 -11.419623  3.336787e-30       -1.392674   \n",
       "\n",
       "          Conf. Int. Upp.     pvalue-hs  reject-hs  \n",
       "mut-ctrl        -0.984651  3.336787e-30       True  "
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "glm_nb.t_test_pairwise(\"genotype\").result_frame"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "087e3d91-9830-41c0-81bd-e725d23dbcce",
   "metadata": {},
   "source": [
    "Yes, the total number of cells per clone is significantly different in \"ctrl\" vs \"mutant\" animals.\n",
    "\n",
    "**Conclusion:**  \n",
    "These data are consistent with a role for our gene of interest in controlling cell proliferation. **However**, complementary studies are be required to test this hypothesis further and eliminate alternative theories, e.g. does our gene of interest enhance cell death, leading to few cells per clone."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "67cde493-71d7-4cf0-8a97-efadaf673608",
   "metadata": {},
   "source": [
    "## Cell Differentiation - percentage cell type inside and outside clones"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "be69defe-43b3-4eff-bc94-9bfeed4a5f49",
   "metadata": {},
   "source": [
    "Next, let's investigate whether the types of cells inside and outside clones are different in ctrl and mutant animals. This information will help us determine whether our gene of interest affects **cell differentiation**.\n",
    "\n",
    "First, since we already know that 'mutant' clones possess fewer cells than 'ctrl' clones, it is helpful to summarise cell types as a percentage."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "feb73191-1c89-4608-a444-0eb6cccfd4f2",
   "metadata": {},
   "outputs": [],
   "source": [
    "percentage_df = count_df.eval(\n",
    "    \"\"\"EC = EC / total * 100 \n",
    "    EE = EE / total * 100\n",
    "    ECEE = ECEE / total * 100 \n",
    "    preEC = preEC / total * 100 \n",
    "    ISCorEB = ISCorEB / total * 100\"\"\"\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "117712dd-8e77-4947-97b4-a5709f7c7469",
   "metadata": {},
   "source": [
    "Next, we plot a stacked barplot as before. However, this time we plot percentage cell types:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "12e6654c-68ac-461e-abcc-39a34402d647",
   "metadata": {},
   "outputs": [],
   "source": [
    "percentage_df_mean = (\n",
    "    percentage_df[[\"genotype\", \"clone_status\", \"ISCorEB\", \"preEC\", \"EC\", \"EE\", \"ECEE\"]]\n",
    "    .groupby([\"clone_status\", \"genotype\"])\n",
    "    .mean()\n",
    ")\n",
    "\n",
    "percentage_df_sem = (\n",
    "    percentage_df[[\"genotype\", \"clone_status\", \"ISCorEB\", \"preEC\", \"EC\", \"EE\", \"ECEE\"]]\n",
    "    .groupby([\"clone_status\", \"genotype\"])\n",
    "    .sem()\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "49297640-918b-49f7-a4c0-cce21597bcb4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 144x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "with sns.plotting_context(\"notebook\", font_scale=1.2):\n",
    "    ca.create_stack_bar_plot(\n",
    "        percentage_df_mean,\n",
    "        percentage_df_sem,\n",
    "        x_figSize=2,\n",
    "        y_figSize=4,\n",
    "        color_pal=[\"#008b68\", \"#539DC2\", \"#F57171\", \"#95a5a6\", \"#000000\"],\n",
    "        bar_width=0.8,\n",
    "        y_label=\"% cell type per clone\",\n",
    "    )"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7f03a807-d767-47a8-b134-97963fffd0ff",
   "metadata": {},
   "source": [
    "This plot suggests there might be a difference in percentage cell-types of mutant clones vs ctrl. Let's check if these differences are statistically significant.\n",
    "\n",
    "To do this, we reshape our percentage data into a [tidy](https://cran.r-project.org/web/packages/tidyr/vignettes/tidy-data.html) format:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "57972ef8-4592-4c49-9c3c-cf372efc01ad",
   "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>genotype</th>\n",
       "      <th>clone_status</th>\n",
       "      <th>cell_type</th>\n",
       "      <th>perc_cell_num</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>ctrl</td>\n",
       "      <td>OC</td>\n",
       "      <td>ISCorEB</td>\n",
       "      <td>56.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>ctrl</td>\n",
       "      <td>IC</td>\n",
       "      <td>ISCorEB</td>\n",
       "      <td>35.714286</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>ctrl</td>\n",
       "      <td>IC</td>\n",
       "      <td>ISCorEB</td>\n",
       "      <td>100.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>ctrl</td>\n",
       "      <td>IC</td>\n",
       "      <td>ISCorEB</td>\n",
       "      <td>75.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>ctrl</td>\n",
       "      <td>IC</td>\n",
       "      <td>ISCorEB</td>\n",
       "      <td>61.764706</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  genotype clone_status cell_type  perc_cell_num\n",
       "0     ctrl           OC   ISCorEB      56.000000\n",
       "1     ctrl           IC   ISCorEB      35.714286\n",
       "2     ctrl           IC   ISCorEB     100.000000\n",
       "3     ctrl           IC   ISCorEB      75.000000\n",
       "4     ctrl           IC   ISCorEB      61.764706"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tidy_percentage_df = pd.melt(\n",
    "    percentage_df,\n",
    "    id_vars=[\"genotype\", \"clone_status\"],\n",
    "    value_vars=[\"ISCorEB\", \"preEC\", \"EC\", \"EE\", \"ECEE\"],\n",
    "    var_name=\"cell_type\",\n",
    "    value_name=\"perc_cell_num\",\n",
    ")\n",
    "tidy_percentage_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6c238b73-05b9-43db-b1a0-95bdcdec43f5",
   "metadata": {},
   "source": [
    "We then create a composite column that contains genotype (g), clone_status (cs) and cell_type (ct) information."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "53c9e174-c238-46f9-8d2d-758973bc6263",
   "metadata": {},
   "outputs": [],
   "source": [
    "tidy_percentage_df[\"g_cs_ct\"] = (\n",
    "    tidy_percentage_df[\"genotype\"]\n",
    "    + \"_\"\n",
    "    + tidy_percentage_df[\"clone_status\"]\n",
    "    + \"_\"\n",
    "    + tidy_percentage_df[\"cell_type\"]\n",
    ")\n",
    "tidy_percentage_df.head()\n",
    "\n",
    "tidy_percentage_df[\"g_cs_ct\"] = tidy_percentage_df[\"g_cs_ct\"].astype(\"category\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fc871c57-e111-4a85-b685-e5e9a6c30b92",
   "metadata": {},
   "source": [
    "Since we interested in specific comparisons, we setup a contrast matrix.\n",
    "\n",
    "Our first set of contrasts compares the percentage of each cell type <u>inside mutant vs ctrl clones</u>:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "02c50e6e-d1f8-45f9-ab15-5cb3266f3879",
   "metadata": {},
   "outputs": [],
   "source": [
    "contrasts = [\n",
    "    \"mut_IC_EC-ctrl_IC_EC\",\n",
    "    \"mut_IC_EE-ctrl_IC_EE\",\n",
    "    \"mut_IC_preEC-ctrl_IC_preEC\",\n",
    "    \"mut_IC_ISCorEB-ctrl_IC_ISCorEB\",\n",
    "    \"mut_IC_ECEE-ctrl_IC_ECEE\",\n",
    "]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9a105017-8055-4a22-beef-d40ecf335dc7",
   "metadata": {},
   "source": [
    "Next, we would like to compare the percentage of each cell type <u>outside mutant vs ctrl clones</u>:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "aec424b2-e6a6-4518-9ea0-4b5ce6d1f9a5",
   "metadata": {},
   "outputs": [],
   "source": [
    "contrasts = contrasts + [\n",
    "    \"mut_OC_EC-ctrl_OC_EC\",\n",
    "    \"mut_OC_EE-ctrl_OC_EE\",\n",
    "    \"mut_OC_preEC-ctrl_OC_preEC\",\n",
    "    \"mut_OC_ISCorEB-ctrl_OC_ISCorEB\",\n",
    "    \"mut_OC_ECEE-ctrl_OC_ECEE\",\n",
    "]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "10a42003-0b25-4476-a820-f8dc330b31c1",
   "metadata": {},
   "source": [
    "Finally, we would  like to compare the percentage of each cell type <u>outside mutant vs inside mutant clones</u>."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "941c2e41-4128-4bc0-b5ca-5e851c67dd88",
   "metadata": {},
   "outputs": [],
   "source": [
    "contrasts = contrasts + [\n",
    "    \"mut_OC_EC-mut_IC_EC\",\n",
    "    \"mut_OC_EE-mut_IC_EE\",\n",
    "    \"mut_OC_preEC-mut_IC_preEC\",\n",
    "    \"mut_OC_ISCorEB-mut_IC_ISCorEB\",\n",
    "    \"mut_OC_ECEE-mut_IC_ECEE\",\n",
    "]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "69986f0d-24bb-4064-91e6-bf4854db6d2b",
   "metadata": {},
   "source": [
    "Below is a helper function to construct the values of contrast matrix:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b1bd2dd8-76b9-41f1-afd6-871245142cf0",
   "metadata": {},
   "outputs": [],
   "source": [
    "def create_contrast_matrix(desired_contrasts, df_column):\n",
    "    l = list()\n",
    "    levels = pd.Series(df_column.cat.categories.values)\n",
    "    for contrast in desired_contrasts:\n",
    "        group_A, group_B = contrast.split(\"-\")\n",
    "        l.append(\n",
    "            levels.replace({f\"^{group_A}$\": -1, f\"^{group_B}$\": 1, \".\": 0}, regex=True)\n",
    "        )\n",
    "    df = pd.DataFrame(l)\n",
    "    df.columns = levels.values\n",
    "    df.index = desired_contrasts\n",
    "    return df"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fc3b937e-9e6f-4f57-b5ca-17279d76457c",
   "metadata": {},
   "source": [
    "We can now setup a custom contrast matrix:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "fb94f1df-a711-4779-82b7-fdb3042d897c",
   "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>ctrl_IC_EC</th>\n",
       "      <th>ctrl_IC_ECEE</th>\n",
       "      <th>ctrl_IC_EE</th>\n",
       "      <th>ctrl_IC_ISCorEB</th>\n",
       "      <th>ctrl_IC_preEC</th>\n",
       "      <th>ctrl_OC_EC</th>\n",
       "      <th>ctrl_OC_ECEE</th>\n",
       "      <th>ctrl_OC_EE</th>\n",
       "      <th>ctrl_OC_ISCorEB</th>\n",
       "      <th>ctrl_OC_preEC</th>\n",
       "      <th>mut_IC_EC</th>\n",
       "      <th>mut_IC_ECEE</th>\n",
       "      <th>mut_IC_EE</th>\n",
       "      <th>mut_IC_ISCorEB</th>\n",
       "      <th>mut_IC_preEC</th>\n",
       "      <th>mut_OC_EC</th>\n",
       "      <th>mut_OC_ECEE</th>\n",
       "      <th>mut_OC_EE</th>\n",
       "      <th>mut_OC_ISCorEB</th>\n",
       "      <th>mut_OC_preEC</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>mut_IC_EC-ctrl_IC_EC</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mut_IC_EE-ctrl_IC_EE</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mut_IC_preEC-ctrl_IC_preEC</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mut_IC_ISCorEB-ctrl_IC_ISCorEB</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mut_IC_ECEE-ctrl_IC_ECEE</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mut_OC_EC-ctrl_OC_EC</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mut_OC_EE-ctrl_OC_EE</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mut_OC_preEC-ctrl_OC_preEC</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mut_OC_ISCorEB-ctrl_OC_ISCorEB</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mut_OC_ECEE-ctrl_OC_ECEE</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mut_OC_EC-mut_IC_EC</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mut_OC_EE-mut_IC_EE</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mut_OC_preEC-mut_IC_preEC</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mut_OC_ISCorEB-mut_IC_ISCorEB</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mut_OC_ECEE-mut_IC_ECEE</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                ctrl_IC_EC  ctrl_IC_ECEE  ctrl_IC_EE  \\\n",
       "mut_IC_EC-ctrl_IC_EC                     1             0           0   \n",
       "mut_IC_EE-ctrl_IC_EE                     0             0           1   \n",
       "mut_IC_preEC-ctrl_IC_preEC               0             0           0   \n",
       "mut_IC_ISCorEB-ctrl_IC_ISCorEB           0             0           0   \n",
       "mut_IC_ECEE-ctrl_IC_ECEE                 0             1           0   \n",
       "mut_OC_EC-ctrl_OC_EC                     0             0           0   \n",
       "mut_OC_EE-ctrl_OC_EE                     0             0           0   \n",
       "mut_OC_preEC-ctrl_OC_preEC               0             0           0   \n",
       "mut_OC_ISCorEB-ctrl_OC_ISCorEB           0             0           0   \n",
       "mut_OC_ECEE-ctrl_OC_ECEE                 0             0           0   \n",
       "mut_OC_EC-mut_IC_EC                      0             0           0   \n",
       "mut_OC_EE-mut_IC_EE                      0             0           0   \n",
       "mut_OC_preEC-mut_IC_preEC                0             0           0   \n",
       "mut_OC_ISCorEB-mut_IC_ISCorEB            0             0           0   \n",
       "mut_OC_ECEE-mut_IC_ECEE                  0             0           0   \n",
       "\n",
       "                                ctrl_IC_ISCorEB  ctrl_IC_preEC  ctrl_OC_EC  \\\n",
       "mut_IC_EC-ctrl_IC_EC                          0              0           0   \n",
       "mut_IC_EE-ctrl_IC_EE                          0              0           0   \n",
       "mut_IC_preEC-ctrl_IC_preEC                    0              1           0   \n",
       "mut_IC_ISCorEB-ctrl_IC_ISCorEB                1              0           0   \n",
       "mut_IC_ECEE-ctrl_IC_ECEE                      0              0           0   \n",
       "mut_OC_EC-ctrl_OC_EC                          0              0           1   \n",
       "mut_OC_EE-ctrl_OC_EE                          0              0           0   \n",
       "mut_OC_preEC-ctrl_OC_preEC                    0              0           0   \n",
       "mut_OC_ISCorEB-ctrl_OC_ISCorEB                0              0           0   \n",
       "mut_OC_ECEE-ctrl_OC_ECEE                      0              0           0   \n",
       "mut_OC_EC-mut_IC_EC                           0              0           0   \n",
       "mut_OC_EE-mut_IC_EE                           0              0           0   \n",
       "mut_OC_preEC-mut_IC_preEC                     0              0           0   \n",
       "mut_OC_ISCorEB-mut_IC_ISCorEB                 0              0           0   \n",
       "mut_OC_ECEE-mut_IC_ECEE                       0              0           0   \n",
       "\n",
       "                                ctrl_OC_ECEE  ctrl_OC_EE  ctrl_OC_ISCorEB  \\\n",
       "mut_IC_EC-ctrl_IC_EC                       0           0                0   \n",
       "mut_IC_EE-ctrl_IC_EE                       0           0                0   \n",
       "mut_IC_preEC-ctrl_IC_preEC                 0           0                0   \n",
       "mut_IC_ISCorEB-ctrl_IC_ISCorEB             0           0                0   \n",
       "mut_IC_ECEE-ctrl_IC_ECEE                   0           0                0   \n",
       "mut_OC_EC-ctrl_OC_EC                       0           0                0   \n",
       "mut_OC_EE-ctrl_OC_EE                       0           1                0   \n",
       "mut_OC_preEC-ctrl_OC_preEC                 0           0                0   \n",
       "mut_OC_ISCorEB-ctrl_OC_ISCorEB             0           0                1   \n",
       "mut_OC_ECEE-ctrl_OC_ECEE                   1           0                0   \n",
       "mut_OC_EC-mut_IC_EC                        0           0                0   \n",
       "mut_OC_EE-mut_IC_EE                        0           0                0   \n",
       "mut_OC_preEC-mut_IC_preEC                  0           0                0   \n",
       "mut_OC_ISCorEB-mut_IC_ISCorEB              0           0                0   \n",
       "mut_OC_ECEE-mut_IC_ECEE                    0           0                0   \n",
       "\n",
       "                                ctrl_OC_preEC  mut_IC_EC  mut_IC_ECEE  \\\n",
       "mut_IC_EC-ctrl_IC_EC                        0         -1            0   \n",
       "mut_IC_EE-ctrl_IC_EE                        0          0            0   \n",
       "mut_IC_preEC-ctrl_IC_preEC                  0          0            0   \n",
       "mut_IC_ISCorEB-ctrl_IC_ISCorEB              0          0            0   \n",
       "mut_IC_ECEE-ctrl_IC_ECEE                    0          0           -1   \n",
       "mut_OC_EC-ctrl_OC_EC                        0          0            0   \n",
       "mut_OC_EE-ctrl_OC_EE                        0          0            0   \n",
       "mut_OC_preEC-ctrl_OC_preEC                  1          0            0   \n",
       "mut_OC_ISCorEB-ctrl_OC_ISCorEB              0          0            0   \n",
       "mut_OC_ECEE-ctrl_OC_ECEE                    0          0            0   \n",
       "mut_OC_EC-mut_IC_EC                         0          1            0   \n",
       "mut_OC_EE-mut_IC_EE                         0          0            0   \n",
       "mut_OC_preEC-mut_IC_preEC                   0          0            0   \n",
       "mut_OC_ISCorEB-mut_IC_ISCorEB               0          0            0   \n",
       "mut_OC_ECEE-mut_IC_ECEE                     0          0            1   \n",
       "\n",
       "                                mut_IC_EE  mut_IC_ISCorEB  mut_IC_preEC  \\\n",
       "mut_IC_EC-ctrl_IC_EC                    0               0             0   \n",
       "mut_IC_EE-ctrl_IC_EE                   -1               0             0   \n",
       "mut_IC_preEC-ctrl_IC_preEC              0               0            -1   \n",
       "mut_IC_ISCorEB-ctrl_IC_ISCorEB          0              -1             0   \n",
       "mut_IC_ECEE-ctrl_IC_ECEE                0               0             0   \n",
       "mut_OC_EC-ctrl_OC_EC                    0               0             0   \n",
       "mut_OC_EE-ctrl_OC_EE                    0               0             0   \n",
       "mut_OC_preEC-ctrl_OC_preEC              0               0             0   \n",
       "mut_OC_ISCorEB-ctrl_OC_ISCorEB          0               0             0   \n",
       "mut_OC_ECEE-ctrl_OC_ECEE                0               0             0   \n",
       "mut_OC_EC-mut_IC_EC                     0               0             0   \n",
       "mut_OC_EE-mut_IC_EE                     1               0             0   \n",
       "mut_OC_preEC-mut_IC_preEC               0               0             1   \n",
       "mut_OC_ISCorEB-mut_IC_ISCorEB           0               1             0   \n",
       "mut_OC_ECEE-mut_IC_ECEE                 0               0             0   \n",
       "\n",
       "                                mut_OC_EC  mut_OC_ECEE  mut_OC_EE  \\\n",
       "mut_IC_EC-ctrl_IC_EC                    0            0          0   \n",
       "mut_IC_EE-ctrl_IC_EE                    0            0          0   \n",
       "mut_IC_preEC-ctrl_IC_preEC              0            0          0   \n",
       "mut_IC_ISCorEB-ctrl_IC_ISCorEB          0            0          0   \n",
       "mut_IC_ECEE-ctrl_IC_ECEE                0            0          0   \n",
       "mut_OC_EC-ctrl_OC_EC                   -1            0          0   \n",
       "mut_OC_EE-ctrl_OC_EE                    0            0         -1   \n",
       "mut_OC_preEC-ctrl_OC_preEC              0            0          0   \n",
       "mut_OC_ISCorEB-ctrl_OC_ISCorEB          0            0          0   \n",
       "mut_OC_ECEE-ctrl_OC_ECEE                0           -1          0   \n",
       "mut_OC_EC-mut_IC_EC                    -1            0          0   \n",
       "mut_OC_EE-mut_IC_EE                     0            0         -1   \n",
       "mut_OC_preEC-mut_IC_preEC               0            0          0   \n",
       "mut_OC_ISCorEB-mut_IC_ISCorEB           0            0          0   \n",
       "mut_OC_ECEE-mut_IC_ECEE                 0           -1          0   \n",
       "\n",
       "                                mut_OC_ISCorEB  mut_OC_preEC  \n",
       "mut_IC_EC-ctrl_IC_EC                         0             0  \n",
       "mut_IC_EE-ctrl_IC_EE                         0             0  \n",
       "mut_IC_preEC-ctrl_IC_preEC                   0             0  \n",
       "mut_IC_ISCorEB-ctrl_IC_ISCorEB               0             0  \n",
       "mut_IC_ECEE-ctrl_IC_ECEE                     0             0  \n",
       "mut_OC_EC-ctrl_OC_EC                         0             0  \n",
       "mut_OC_EE-ctrl_OC_EE                         0             0  \n",
       "mut_OC_preEC-ctrl_OC_preEC                   0            -1  \n",
       "mut_OC_ISCorEB-ctrl_OC_ISCorEB              -1             0  \n",
       "mut_OC_ECEE-ctrl_OC_ECEE                     0             0  \n",
       "mut_OC_EC-mut_IC_EC                          0             0  \n",
       "mut_OC_EE-mut_IC_EE                          0             0  \n",
       "mut_OC_preEC-mut_IC_preEC                    0            -1  \n",
       "mut_OC_ISCorEB-mut_IC_ISCorEB               -1             0  \n",
       "mut_OC_ECEE-mut_IC_ECEE                      0             0  "
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "matrix = create_contrast_matrix(contrasts, tidy_percentage_df[\"g_cs_ct\"])\n",
    "matrix"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0bd0c7ea-0b88-40be-ace3-a2625a75c8ec",
   "metadata": {},
   "source": [
    "Finally, since we have proportional data, we use statsmodels to construct a quasi-binomal regression model and perform tests for our desired contrasts:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "0bbb6a37-3f2e-4a05-a86d-550576e3d647",
   "metadata": {},
   "outputs": [],
   "source": [
    "# quasi-binomal regression requires the dependent variable be between 0 and 1. Convert percentages to proportions.\n",
    "tidy_percentage_df[\"pro_cell_num\"] = tidy_percentage_df[\"perc_cell_num\"] / 100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e6f0e681-93a4-4dc5-9030-e31b288bfcac",
   "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>coef</th>\n",
       "      <th>std err</th>\n",
       "      <th>z</th>\n",
       "      <th>P&gt;|z|</th>\n",
       "      <th>Conf. Int. Low</th>\n",
       "      <th>Conf. Int. Upp.</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>mut_IC_EC-ctrl_IC_EC</th>\n",
       "      <td>0.7649</td>\n",
       "      <td>0.1978</td>\n",
       "      <td>3.8681</td>\n",
       "      <td>0.0001</td>\n",
       "      <td>0.3773</td>\n",
       "      <td>1.1525</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mut_IC_EE-ctrl_IC_EE</th>\n",
       "      <td>-0.8064</td>\n",
       "      <td>0.2862</td>\n",
       "      <td>-2.8173</td>\n",
       "      <td>0.0048</td>\n",
       "      <td>-1.3675</td>\n",
       "      <td>-0.2454</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mut_IC_preEC-ctrl_IC_preEC</th>\n",
       "      <td>-1.3296</td>\n",
       "      <td>1.4032</td>\n",
       "      <td>-0.9476</td>\n",
       "      <td>0.3433</td>\n",
       "      <td>-4.0797</td>\n",
       "      <td>1.4205</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mut_IC_ISCorEB-ctrl_IC_ISCorEB</th>\n",
       "      <td>-0.8487</td>\n",
       "      <td>0.1217</td>\n",
       "      <td>-6.9731</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-1.0873</td>\n",
       "      <td>-0.6102</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mut_IC_ECEE-ctrl_IC_ECEE</th>\n",
       "      <td>0.0260</td>\n",
       "      <td>0.7414</td>\n",
       "      <td>0.0351</td>\n",
       "      <td>0.9720</td>\n",
       "      <td>-1.4271</td>\n",
       "      <td>1.4792</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mut_OC_EC-ctrl_OC_EC</th>\n",
       "      <td>0.3803</td>\n",
       "      <td>0.2990</td>\n",
       "      <td>1.2719</td>\n",
       "      <td>0.2034</td>\n",
       "      <td>-0.2057</td>\n",
       "      <td>0.9663</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mut_OC_EE-ctrl_OC_EE</th>\n",
       "      <td>-0.2120</td>\n",
       "      <td>0.6093</td>\n",
       "      <td>-0.3479</td>\n",
       "      <td>0.7279</td>\n",
       "      <td>-1.4062</td>\n",
       "      <td>0.9823</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mut_OC_preEC-ctrl_OC_preEC</th>\n",
       "      <td>-1.2323</td>\n",
       "      <td>1.8048</td>\n",
       "      <td>-0.6828</td>\n",
       "      <td>0.4947</td>\n",
       "      <td>-4.7695</td>\n",
       "      <td>2.3050</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mut_OC_ISCorEB-ctrl_OC_ISCorEB</th>\n",
       "      <td>-0.2758</td>\n",
       "      <td>0.2887</td>\n",
       "      <td>-0.9552</td>\n",
       "      <td>0.3395</td>\n",
       "      <td>-0.8417</td>\n",
       "      <td>0.2901</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mut_OC_ECEE-ctrl_OC_ECEE</th>\n",
       "      <td>0.2190</td>\n",
       "      <td>1.6601</td>\n",
       "      <td>0.1319</td>\n",
       "      <td>0.8950</td>\n",
       "      <td>-3.0346</td>\n",
       "      <td>3.4727</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mut_OC_EC-mut_IC_EC</th>\n",
       "      <td>-0.7743</td>\n",
       "      <td>0.2403</td>\n",
       "      <td>-3.2226</td>\n",
       "      <td>0.0013</td>\n",
       "      <td>-1.2452</td>\n",
       "      <td>-0.3034</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mut_OC_EE-mut_IC_EE</th>\n",
       "      <td>0.0324</td>\n",
       "      <td>0.4411</td>\n",
       "      <td>0.0734</td>\n",
       "      <td>0.9415</td>\n",
       "      <td>-0.8321</td>\n",
       "      <td>0.8968</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mut_OC_preEC-mut_IC_preEC</th>\n",
       "      <td>-1.2188</td>\n",
       "      <td>1.0590</td>\n",
       "      <td>-1.1509</td>\n",
       "      <td>0.2498</td>\n",
       "      <td>-3.2943</td>\n",
       "      <td>0.8567</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mut_OC_ISCorEB-mut_IC_ISCorEB</th>\n",
       "      <td>0.6880</td>\n",
       "      <td>0.2252</td>\n",
       "      <td>3.0544</td>\n",
       "      <td>0.0023</td>\n",
       "      <td>0.2465</td>\n",
       "      <td>1.1294</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mut_OC_ECEE-mut_IC_ECEE</th>\n",
       "      <td>-0.1520</td>\n",
       "      <td>1.3415</td>\n",
       "      <td>-0.1133</td>\n",
       "      <td>0.9098</td>\n",
       "      <td>-2.7813</td>\n",
       "      <td>2.4773</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                  coef  std err       z   P>|z|  \\\n",
       "mut_IC_EC-ctrl_IC_EC            0.7649   0.1978  3.8681  0.0001   \n",
       "mut_IC_EE-ctrl_IC_EE           -0.8064   0.2862 -2.8173  0.0048   \n",
       "mut_IC_preEC-ctrl_IC_preEC     -1.3296   1.4032 -0.9476  0.3433   \n",
       "mut_IC_ISCorEB-ctrl_IC_ISCorEB -0.8487   0.1217 -6.9731  0.0000   \n",
       "mut_IC_ECEE-ctrl_IC_ECEE        0.0260   0.7414  0.0351  0.9720   \n",
       "mut_OC_EC-ctrl_OC_EC            0.3803   0.2990  1.2719  0.2034   \n",
       "mut_OC_EE-ctrl_OC_EE           -0.2120   0.6093 -0.3479  0.7279   \n",
       "mut_OC_preEC-ctrl_OC_preEC     -1.2323   1.8048 -0.6828  0.4947   \n",
       "mut_OC_ISCorEB-ctrl_OC_ISCorEB -0.2758   0.2887 -0.9552  0.3395   \n",
       "mut_OC_ECEE-ctrl_OC_ECEE        0.2190   1.6601  0.1319  0.8950   \n",
       "mut_OC_EC-mut_IC_EC            -0.7743   0.2403 -3.2226  0.0013   \n",
       "mut_OC_EE-mut_IC_EE             0.0324   0.4411  0.0734  0.9415   \n",
       "mut_OC_preEC-mut_IC_preEC      -1.2188   1.0590 -1.1509  0.2498   \n",
       "mut_OC_ISCorEB-mut_IC_ISCorEB   0.6880   0.2252  3.0544  0.0023   \n",
       "mut_OC_ECEE-mut_IC_ECEE        -0.1520   1.3415 -0.1133  0.9098   \n",
       "\n",
       "                                Conf. Int. Low  Conf. Int. Upp.  \n",
       "mut_IC_EC-ctrl_IC_EC                    0.3773           1.1525  \n",
       "mut_IC_EE-ctrl_IC_EE                   -1.3675          -0.2454  \n",
       "mut_IC_preEC-ctrl_IC_preEC             -4.0797           1.4205  \n",
       "mut_IC_ISCorEB-ctrl_IC_ISCorEB         -1.0873          -0.6102  \n",
       "mut_IC_ECEE-ctrl_IC_ECEE               -1.4271           1.4792  \n",
       "mut_OC_EC-ctrl_OC_EC                   -0.2057           0.9663  \n",
       "mut_OC_EE-ctrl_OC_EE                   -1.4062           0.9823  \n",
       "mut_OC_preEC-ctrl_OC_preEC             -4.7695           2.3050  \n",
       "mut_OC_ISCorEB-ctrl_OC_ISCorEB         -0.8417           0.2901  \n",
       "mut_OC_ECEE-ctrl_OC_ECEE               -3.0346           3.4727  \n",
       "mut_OC_EC-mut_IC_EC                    -1.2452          -0.3034  \n",
       "mut_OC_EE-mut_IC_EE                    -0.8321           0.8968  \n",
       "mut_OC_preEC-mut_IC_preEC              -3.2943           0.8567  \n",
       "mut_OC_ISCorEB-mut_IC_ISCorEB           0.2465           1.1294  \n",
       "mut_OC_ECEE-mut_IC_ECEE                -2.7813           2.4773  "
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "glm_quasi = glm(\n",
    "    \"pro_cell_num ~ g_cs_ct\", data=tidy_percentage_df, family=sm.families.Binomial(),\n",
    ").fit(scale=\"X2\")\n",
    "\n",
    "t_test_df = glm_quasi.t_test(matrix).summary_frame()\n",
    "t_test_df.index = matrix.index\n",
    "t_test_df.round(4)  # round for easier reading"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "003041bc-5aeb-4161-ada8-844313dceb63",
   "metadata": {},
   "source": [
    "However, as we are performing multiple comparisons, we need to adjust our p values. We can do that using the Benjamini/Hochberg false-discovery rate procedure:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "4e19bd30-2c0e-459d-9db5-31424370c4da",
   "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>coef</th>\n",
       "      <th>std err</th>\n",
       "      <th>z</th>\n",
       "      <th>P&gt;|z|</th>\n",
       "      <th>Conf. Int. Low</th>\n",
       "      <th>Conf. Int. Upp.</th>\n",
       "      <th>pval_fdr_bh</th>\n",
       "      <th>reject_fdr_bh</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>mut_IC_EC-ctrl_IC_EC</th>\n",
       "      <td>0.7649</td>\n",
       "      <td>0.1978</td>\n",
       "      <td>3.8681</td>\n",
       "      <td>0.0001</td>\n",
       "      <td>0.3773</td>\n",
       "      <td>1.1525</td>\n",
       "      <td>0.0008</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mut_IC_EE-ctrl_IC_EE</th>\n",
       "      <td>-0.8064</td>\n",
       "      <td>0.2862</td>\n",
       "      <td>-2.8173</td>\n",
       "      <td>0.0048</td>\n",
       "      <td>-1.3675</td>\n",
       "      <td>-0.2454</td>\n",
       "      <td>0.0145</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mut_IC_preEC-ctrl_IC_preEC</th>\n",
       "      <td>-1.3296</td>\n",
       "      <td>1.4032</td>\n",
       "      <td>-0.9476</td>\n",
       "      <td>0.3433</td>\n",
       "      <td>-4.0797</td>\n",
       "      <td>1.4205</td>\n",
       "      <td>0.5722</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mut_IC_ISCorEB-ctrl_IC_ISCorEB</th>\n",
       "      <td>-0.8487</td>\n",
       "      <td>0.1217</td>\n",
       "      <td>-6.9731</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-1.0873</td>\n",
       "      <td>-0.6102</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mut_IC_ECEE-ctrl_IC_ECEE</th>\n",
       "      <td>0.0260</td>\n",
       "      <td>0.7414</td>\n",
       "      <td>0.0351</td>\n",
       "      <td>0.9720</td>\n",
       "      <td>-1.4271</td>\n",
       "      <td>1.4792</td>\n",
       "      <td>0.9720</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mut_OC_EC-ctrl_OC_EC</th>\n",
       "      <td>0.3803</td>\n",
       "      <td>0.2990</td>\n",
       "      <td>1.2719</td>\n",
       "      <td>0.2034</td>\n",
       "      <td>-0.2057</td>\n",
       "      <td>0.9663</td>\n",
       "      <td>0.5085</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mut_OC_EE-ctrl_OC_EE</th>\n",
       "      <td>-0.2120</td>\n",
       "      <td>0.6093</td>\n",
       "      <td>-0.3479</td>\n",
       "      <td>0.7279</td>\n",
       "      <td>-1.4062</td>\n",
       "      <td>0.9823</td>\n",
       "      <td>0.9720</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mut_OC_preEC-ctrl_OC_preEC</th>\n",
       "      <td>-1.2323</td>\n",
       "      <td>1.8048</td>\n",
       "      <td>-0.6828</td>\n",
       "      <td>0.4947</td>\n",
       "      <td>-4.7695</td>\n",
       "      <td>2.3050</td>\n",
       "      <td>0.7421</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mut_OC_ISCorEB-ctrl_OC_ISCorEB</th>\n",
       "      <td>-0.2758</td>\n",
       "      <td>0.2887</td>\n",
       "      <td>-0.9552</td>\n",
       "      <td>0.3395</td>\n",
       "      <td>-0.8417</td>\n",
       "      <td>0.2901</td>\n",
       "      <td>0.5722</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mut_OC_ECEE-ctrl_OC_ECEE</th>\n",
       "      <td>0.2190</td>\n",
       "      <td>1.6601</td>\n",
       "      <td>0.1319</td>\n",
       "      <td>0.8950</td>\n",
       "      <td>-3.0346</td>\n",
       "      <td>3.4727</td>\n",
       "      <td>0.9720</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mut_OC_EC-mut_IC_EC</th>\n",
       "      <td>-0.7743</td>\n",
       "      <td>0.2403</td>\n",
       "      <td>-3.2226</td>\n",
       "      <td>0.0013</td>\n",
       "      <td>-1.2452</td>\n",
       "      <td>-0.3034</td>\n",
       "      <td>0.0064</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mut_OC_EE-mut_IC_EE</th>\n",
       "      <td>0.0324</td>\n",
       "      <td>0.4411</td>\n",
       "      <td>0.0734</td>\n",
       "      <td>0.9415</td>\n",
       "      <td>-0.8321</td>\n",
       "      <td>0.8968</td>\n",
       "      <td>0.9720</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mut_OC_preEC-mut_IC_preEC</th>\n",
       "      <td>-1.2188</td>\n",
       "      <td>1.0590</td>\n",
       "      <td>-1.1509</td>\n",
       "      <td>0.2498</td>\n",
       "      <td>-3.2943</td>\n",
       "      <td>0.8567</td>\n",
       "      <td>0.5352</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mut_OC_ISCorEB-mut_IC_ISCorEB</th>\n",
       "      <td>0.6880</td>\n",
       "      <td>0.2252</td>\n",
       "      <td>3.0544</td>\n",
       "      <td>0.0023</td>\n",
       "      <td>0.2465</td>\n",
       "      <td>1.1294</td>\n",
       "      <td>0.0085</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mut_OC_ECEE-mut_IC_ECEE</th>\n",
       "      <td>-0.1520</td>\n",
       "      <td>1.3415</td>\n",
       "      <td>-0.1133</td>\n",
       "      <td>0.9098</td>\n",
       "      <td>-2.7813</td>\n",
       "      <td>2.4773</td>\n",
       "      <td>0.9720</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                  coef  std err       z   P>|z|  \\\n",
       "mut_IC_EC-ctrl_IC_EC            0.7649   0.1978  3.8681  0.0001   \n",
       "mut_IC_EE-ctrl_IC_EE           -0.8064   0.2862 -2.8173  0.0048   \n",
       "mut_IC_preEC-ctrl_IC_preEC     -1.3296   1.4032 -0.9476  0.3433   \n",
       "mut_IC_ISCorEB-ctrl_IC_ISCorEB -0.8487   0.1217 -6.9731  0.0000   \n",
       "mut_IC_ECEE-ctrl_IC_ECEE        0.0260   0.7414  0.0351  0.9720   \n",
       "mut_OC_EC-ctrl_OC_EC            0.3803   0.2990  1.2719  0.2034   \n",
       "mut_OC_EE-ctrl_OC_EE           -0.2120   0.6093 -0.3479  0.7279   \n",
       "mut_OC_preEC-ctrl_OC_preEC     -1.2323   1.8048 -0.6828  0.4947   \n",
       "mut_OC_ISCorEB-ctrl_OC_ISCorEB -0.2758   0.2887 -0.9552  0.3395   \n",
       "mut_OC_ECEE-ctrl_OC_ECEE        0.2190   1.6601  0.1319  0.8950   \n",
       "mut_OC_EC-mut_IC_EC            -0.7743   0.2403 -3.2226  0.0013   \n",
       "mut_OC_EE-mut_IC_EE             0.0324   0.4411  0.0734  0.9415   \n",
       "mut_OC_preEC-mut_IC_preEC      -1.2188   1.0590 -1.1509  0.2498   \n",
       "mut_OC_ISCorEB-mut_IC_ISCorEB   0.6880   0.2252  3.0544  0.0023   \n",
       "mut_OC_ECEE-mut_IC_ECEE        -0.1520   1.3415 -0.1133  0.9098   \n",
       "\n",
       "                                Conf. Int. Low  Conf. Int. Upp.  pval_fdr_bh  \\\n",
       "mut_IC_EC-ctrl_IC_EC                    0.3773           1.1525       0.0008   \n",
       "mut_IC_EE-ctrl_IC_EE                   -1.3675          -0.2454       0.0145   \n",
       "mut_IC_preEC-ctrl_IC_preEC             -4.0797           1.4205       0.5722   \n",
       "mut_IC_ISCorEB-ctrl_IC_ISCorEB         -1.0873          -0.6102       0.0000   \n",
       "mut_IC_ECEE-ctrl_IC_ECEE               -1.4271           1.4792       0.9720   \n",
       "mut_OC_EC-ctrl_OC_EC                   -0.2057           0.9663       0.5085   \n",
       "mut_OC_EE-ctrl_OC_EE                   -1.4062           0.9823       0.9720   \n",
       "mut_OC_preEC-ctrl_OC_preEC             -4.7695           2.3050       0.7421   \n",
       "mut_OC_ISCorEB-ctrl_OC_ISCorEB         -0.8417           0.2901       0.5722   \n",
       "mut_OC_ECEE-ctrl_OC_ECEE               -3.0346           3.4727       0.9720   \n",
       "mut_OC_EC-mut_IC_EC                    -1.2452          -0.3034       0.0064   \n",
       "mut_OC_EE-mut_IC_EE                    -0.8321           0.8968       0.9720   \n",
       "mut_OC_preEC-mut_IC_preEC              -3.2943           0.8567       0.5352   \n",
       "mut_OC_ISCorEB-mut_IC_ISCorEB           0.2465           1.1294       0.0085   \n",
       "mut_OC_ECEE-mut_IC_ECEE                -2.7813           2.4773       0.9720   \n",
       "\n",
       "                                reject_fdr_bh  \n",
       "mut_IC_EC-ctrl_IC_EC                     True  \n",
       "mut_IC_EE-ctrl_IC_EE                     True  \n",
       "mut_IC_preEC-ctrl_IC_preEC              False  \n",
       "mut_IC_ISCorEB-ctrl_IC_ISCorEB           True  \n",
       "mut_IC_ECEE-ctrl_IC_ECEE                False  \n",
       "mut_OC_EC-ctrl_OC_EC                    False  \n",
       "mut_OC_EE-ctrl_OC_EE                    False  \n",
       "mut_OC_preEC-ctrl_OC_preEC              False  \n",
       "mut_OC_ISCorEB-ctrl_OC_ISCorEB          False  \n",
       "mut_OC_ECEE-ctrl_OC_ECEE                False  \n",
       "mut_OC_EC-mut_IC_EC                      True  \n",
       "mut_OC_EE-mut_IC_EE                     False  \n",
       "mut_OC_preEC-mut_IC_preEC               False  \n",
       "mut_OC_ISCorEB-mut_IC_ISCorEB            True  \n",
       "mut_OC_ECEE-mut_IC_ECEE                 False  "
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from statsmodels.stats.multitest import fdrcorrection\n",
    "\n",
    "fdr_results = fdrcorrection(t_test_df[\"P>|z|\"])\n",
    "\n",
    "t_test_df[\"pval_fdr_bh\"] = fdr_results[1]\n",
    "t_test_df[\"reject_fdr_bh\"] = fdr_results[0]\n",
    "\n",
    "t_test_df.round(4)  # round for easier reading"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e0b7749c-c39a-4d71-abc0-08c72c4f856c",
   "metadata": {},
   "source": [
    "To aid visualization, let's add these results to our stacked percentage barplot from earlier. We annotations only statistically significant comparisons."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "001013db-9517-4cb9-b812-b1c98dc2af6b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 144x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# stat annotation functions from clonedetective:\n",
    "from clonedetective.clone_analysis import plot_stat_annotation, pvals_to_stat_anots\n",
    "\n",
    "# convert t_test_results to annotations:\n",
    "stat_anots = pvals_to_stat_anots(t_test_df[\"pval_fdr_bh\"])\n",
    "\n",
    "with sns.plotting_context(\"notebook\", font_scale=1.2):\n",
    "    ca.create_stack_bar_plot(\n",
    "        percentage_df_mean,\n",
    "        percentage_df_sem,\n",
    "        x_figSize=2,\n",
    "        y_figSize=5,\n",
    "        color_pal=[\"#008b68\", \"#539DC2\", \"#F57171\", \"#95a5a6\", \"#000000\"],\n",
    "        bar_width=0.8,\n",
    "        y_label=\"% cell type per clone\",\n",
    "    )\n",
    "    plot_stat_annotation(\n",
    "        (0, 1),\n",
    "        105,\n",
    "        stat_anots[\n",
    "            [\n",
    "                \"mut_IC_EC-ctrl_IC_EC\",\n",
    "                \"mut_IC_ISCorEB-ctrl_IC_ISCorEB\",\n",
    "                \"mut_IC_EE-ctrl_IC_EE\",\n",
    "            ]\n",
    "        ],\n",
    "        sep=1,\n",
    "        text_colors=[\"#008b68\", \"#F57171\", \"#95a5a6\"],\n",
    "    )\n",
    "    plot_stat_annotation(\n",
    "        (1, 3),\n",
    "        120,\n",
    "        stat_anots[[\"mut_OC_EC-mut_IC_EC\", \"mut_OC_ISCorEB-mut_IC_ISCorEB\"]],\n",
    "        sep=1,\n",
    "        text_colors=[\"#008b68\", \"#F57171\"],\n",
    "    )"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "da398447-ef6c-44f2-b5e3-061a953d46e6",
   "metadata": {},
   "source": [
    "**Summary of results:**  \n",
    "* comparing the percentage makeup of control and mutant clones, mutant clones possess significantly fewer ECs and significantly more progenitor cells (e.g. ISCs or EBs) and EEs.\n",
    "* comparing the percentage makeup of cells inside mutant clones and outside mutant clones, cells inside clones similarly possess significantly fewer ECs and more progenitor cells.\n",
    "* no significant difference is seen in the percentage makeup of cells outside clones in control vs mutant animals, suggesting our gene of interest may be haploinsufficient.\n",
    "\n",
    "\n",
    "**Conclusion:**  \n",
    "Together, these data suggest mutation of our gene of interest may hinder ISC -> EC differentiation. **However**, complementary studies are be required to further test this hypothesis and eliminate alternative theories, e.g. does our gene of interest lead to enhanced EC cell death, resulting in higher percentage progenitor and EE cells."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "060c418b-377b-4658-8834-ddaee0999df4",
   "metadata": {},
   "source": [
    "## Number of neighbours"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ca361e66-7288-4cdb-8f19-4954143ff72a",
   "metadata": {},
   "source": [
    "We have found that mutant clones are smaller than control (wildtype) clones, and may therefore regulate **cell proliferation**. We also found that gene of interest may regulate **cell differentiation**, as mutant clones possess a lower percentage of ECs and a higher percentage of progenitor cells (ISCs or EBs). We now wish to find out whether our gene of interest regulates **local tissue cell arrangement and structure** by comparing the number and type of cell neighbours in control vs mutant clones.\n",
    "\n",
    "First, we initially perform a 'sanity check' to make sure our 'neighbour counts' are reporting what we would expect.\n",
    "\n",
    "Since mutant clones are smaller than controls clones, we would expect that compared to cells within control clones, each cell within mutant clones will have (on average) a lower proportion GFP positive neighbours and a higher proportion of GFP negative neighbours."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "be6235bb-49d9-449b-9a33-6072c48f6ffb",
   "metadata": {},
   "outputs": [],
   "source": [
    "gfp_df = (\n",
    "    df.query('int_img_ch == \"C0\" & clone_status == \"IC\"')\n",
    "    .set_index([\"int_img\", \"label\", \"GFP_clone\"])\n",
    "    .filter(regex=\"GFP|tot_nc|genotype|clone_status\")\n",
    "    .groupby([\"int_img\", \"GFP_clone\"])\n",
    "    .agg(\n",
    "        {\n",
    "            \"genotype\": \"first\",\n",
    "            \"GFPneg_nc\": \"mean\",\n",
    "            \"GFPpos_nc\": \"mean\",\n",
    "            \"tot_nc\": \"mean\",\n",
    "        }\n",
    "    )\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "ac8ea8ec-9139-49bf-a058-d214acbfb131",
   "metadata": {},
   "outputs": [],
   "source": [
    "tidy_gfp_df = pd.melt(\n",
    "    gfp_df,\n",
    "    id_vars=[\"genotype\"],\n",
    "    value_vars=[\"GFPneg_nc\", \"GFPpos_nc\", \"tot_nc\"],\n",
    "    var_name=\"neigh_type\",\n",
    "    value_name=\"neighbour_count\",\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "6fc5e087-6eea-4c3f-aad1-acbb0f052b11",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAREAAAERCAYAAABclJOOAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8rg+JYAAAACXBIWXMAAAsTAAALEwEAmpwYAAAmq0lEQVR4nO3deVhUZf8G8BuGYVFDkFXAJa3QN0xAkiXNXRTDtTQjocy03LVF0RIiRVIzUzRfUhPLsq7MEFTKVMQ3RcVI30pNwwhE9mDYZxjO7w9f5+fE4oFhNrk/1+VFc+ac83yP5H2d5TnPYyIIggAiolYy1XcBRGTcGCJEpBGGCBFphCFCRBox03cBzampqcEvv/wCBwcHSCQSfZdDraBUKlFYWAgPDw9YWlrquxzSAoMOkV9++QUhISH6LoPawN69e+Hj46PvMkgLDDpEHBwcANz+H9DZ2VnP1VBr5OXlISQkRPW7pPuPQYfInUsYZ2dnuLm56bka0gQvR+9fvLFKRBphiBCRRhgiRKQRg74nQve3+vp65OTkoLKyUt+lUDOkUikcHR1hbW3d6PcMEdKboqIimJiYwN3dHaamPCk2RIIgoLq6Gjdv3gSARoOEvznSm9LSUjg5OTFADJiJiQk6dOgAV1dXFBQUNLoOf3ukN0qlElKpVN9lkAhWVlZQKBSNfscQaSNpaWlYunQp0tLS9F2KUTExMdF3CSRCc78n3hNpI7t378a1a9dQVVUFPz8/fZdDpDMMkTZSVVWl9pNark6phJmWe7a2tI3jx49j586duHr1Kurr6+Hp6YmFCxfC09MTy5cvR1JSUoNLsm+++QYZGRl46623YGFhoVrerVs3vPnmmxg0aFCbHY8hYIiQwTCTSLD9s71abeOV58W/0HngwAFs2rQJERER8PPzQ319PY4cOYKXX34Z+/fvBwAsXLgQs2fPbrBtRkYG/P39sXPnTtWykydPYsGCBTh27Bi6dOmi+cEYCIYIUSMUCgViYmKwdetWtbePn3nmGUgkEpSXl7d4n0OGDIGlpSVycnKwd+9eFBQU4Nq1a/j9998xYMAArFu3Dra2tpDL5Vi/fj0OHz4MCwsLzJkzB9OmTQNwO5xWrVqFgoICBAcH4/jx49izZ0+z75YtX74ctra2+M9//oNbt25hyJAhiI6OhoWFBYqKivD222/j7NmzsLe3x8qVKzFkyJAWHRdvrBI1IiMjAx07dmx0+ILJkyfj0UcfbdH+FAoFvv32WwiCgIceeggAcOjQIURGRiI1NRUymQxffPEFACAuLg7Xr1/HkSNHsHv3bsTFxSE9PR21tbVYsGABXnjhBZw6dQpmZmaq/hv3kpycjI8++gjJycn4+eefkZycDAB466230KVLF5w5cwZr167Fa6+9hoqKihYdG89EiBpRUFAAR0dH1eeqqio8+eSTAG4/mh43bhwAIDY2FnFxcar1XnjhBcyfPx/A7Sd2d0LIxMQEvXv3xubNm9GhQwcAwKBBg9CnTx8At89SsrOzAQAHDx5ETEwMrK2tYW1tjWnTpuHgwYNQKBTo2LEjpkyZAgBYvHgxPv30U1HHM2bMGNXZip+fH7Kzs1FTU4NTp07hxIkTsLCwwIABA7B79+4WP3ZniLSQok4JqVnrbv5psi3pVpcuXVBUVKT63KFDB6SnpwO4faaQmZkJAJg/f36j90SA2/9Y774n8k82Njaq/zYzM8Od2Vvy8vLw8ssvqx6rKpVK+Pr6ori4GE5OTqptLC0tYWtrK+p47l5PIpFAEATIZDLU19erjfXi4eEhan93Y4i0kNRMgpn/PtRgeX5ZpepnY98DwK4547RaG7Udb29vlJaW4ueff4anp6dO27a3t8dHH30Ed3d3AEBxcTEAIDMzE/n5+ar15HI5SktLW92OnZ0dTE1NUVRUpAqSXbt2YeTIkejevbvo/fCeCFEjLC0tER4ejsWLF+PEiRNQKBSora1FYmIi9uzZAzs7O621PW7cOMTGxqKiogIlJSWYPXs29u/fD09PT8jlciQkJEChUGDr1q1N9iIVQyKRYMSIEYiNjYVcLseFCxcQFxeHzp07t2g/PBMhUdLS0vDVV19h6tSpWutMV6dUtugRbGvbENtPZMqUKXBwcMDOnTuxbNkyKBQKuLu74/XXX8eECRMQHh6ulRrnzZuHdevWITAwEHV1dQgKCsLMmTNhZmaGDRs2YOXKlVizZg2mTJkCqVSq0asDkZGRiIiIwKBBg2BnZ4cPPvigxSFiYsjTaObk5GDEiBE4duyYQQ2P2Njlys3Du1BXUQqzTjZwDZrZ6HbGfDnzyiuv4Nq1a3j44Yexfft20ds19zu8fPky+vbt29al3rcqKytx/fp19O/fH8DtyxlPT09cuHABVlZWWm+/qd+XTi5n3nvvPfTr1w9eXl7w8vLC008/rYtmqQ2xR67+SSQSvPTSS7h8+TLq6+vxySefoF+/fjoJkObo5HLm6tWriI2NbXEnFiL6f5aWloiOjsbChQtRVFQEDw8PrFu3DidOnMDSpUsb3cbX17dFZ46tobMQeeSRR5pdRyaTQSaTqS3Ly8vTZlltylRqrvaTSBtGjx6N0aNHqy3r0aMHMjIy9FSRDkKkpKQEf//9NyIjI3Hx4kW4u7sjMjISDz74oNp68fHxiI2N1XY5WtP50QDIrqbD2p0TNFH7opMQGThwIObPnw93d3d8/PHHmDt3LpKSktTmIgkLC8OkSZPUtr0z8ZEx6ODSCx1ceum7DCKd03qIPPTQQ9i9e7fq89y5c7Fr1y789ddfamcjd7r4kn6xRy61lNZD5Oeff8aVK1fw7LPPArg9wrdSqYS5Oe8dGCL2yKWW0vojXktLS6xbtw6XLl2CQqHApk2b0LdvX7i6umq7aSLSAa2fifTp0werVq3CkiVLUFJSAm9vb2zcuFHbzZIRqlHIYanlp1stbYMjm92bTh7xTpw4ERMnTtRFU2TELKXm6LLyBa22UbJmt+h1ObKZOHx3hkRpb/1gdDGyWW5uLq5cuYI///wTw4YNw5o1a2BlZYXS0lJERUXhxx9/xAMPPIDZs2dj6tSpAIBPP/0UcXFxUCqV8Pb2xurVq9WGFPins2fPYvPmzXB1dcXRo0fRrVs3REdHq175j4uLw549e6BQKBAcHIwVK1a0eB4gvsVLonR+NAAWDm7o/GiAvkvRCV2MbJaUlITw8HCcOnUKJSUl2LJlCwBgzZo1MDU1RUpKCj766CNs3rwZp0+fRnFxMbZs2YL9+/cjJSUFSqUSX3/99T3bTk9Ph5+fH86ePQtfX19s2LABAPDdd99h37592Lt3L44dO4ZLly7hm2++adFxATwTIZHaWz8YXYxsNmrUKAwcOBAAMGfOHKxatQqvv/46vvvuO3z//fewsrLCww8/jOnTpyMpKQmPPfYYFAoFvv76awQFBWHbtm2i5u2xsbHB5MmTAQCBgYE4fvw4AODo0aOYNm0aevToAQDYuHFjq94IZogQNUIXI5vd/YTSyckJxcXFKCkpgUKhUBvBrGvXrvjpp5/QqVMnxMXFYevWrdi2bRt69+6N9957TzXEYlPuHtXs7hHU/v77b7V2WvumPC9niBpx98hm2nJ3SOXn58PJyQm2traQSCRqI5jl5ubC1tYWZWVlsLCwwO7du5GWlgZ/f39ERka2un1HR0e1+XXPnTuHw4cPt3g/DBGiRuhiZLPvvvsOv/32GyoqKrB9+3aMGzcOEokEgYGB2LBhA6qrq3H9+nXs27cPY8aMQUlJCWbNmoXr16+jY8eO6NChQ4sHELpbYGAgvvzyS+Tk5KC8vBwbN25s1VAPvJwhg1GjkLfoEWxr2xDbT0TbI5t5eXnh7bffRlZWFsaNG6e6LHr77bcRFRWFoUOHwtLSErNnz8bIkSMBAEuWLMGsWbNQVlYGDw8PREdHt7r9oUOHIjMzE88//zxqamowefJk1UjyLcGRzVqhqW7f92Is3cLb8vg4slnjtmzZgsLCQkRFRem7FNH0OrIZEd2/eDlDZOSefPLJJju/JScnqz2B0QaGCJEeLFiwoM32lZqa2mb7ag1ezhCRRhgiRKQRhggRaYQhQkQaYYgQkUYYImQwFHVKg2qjpKQEtbW1za5TWVmJsrIyTcsyanzESwajqUGi21JLeg2PHTsWSUlJcHBwaHKdGTNmYNWqVfD09GyD6owTz0SImlBaWtom69zvGCJEjQgNDQVwe+Cgc+fO4eWXX4aPjw+CgoJw7NgxAMDy5cuRm5uLsLAwnD59utn9/etf/8KuXbvg7++PIUOG4MCBA6rvUlJSEBQUBG9vb8yaNQslJSXaOzAtYIgQNWLPnj0Abo/+tWnTJri7u+P06dOIiorCsmXL8McffyAmJgYuLi6Ij49HQEDzw0YqlUpkZmbi5MmTWL58OaKiolBbW4v8/HwsWbIE4eHhOH/+PLp3746YmBhdHGKbYYgQNaO+vh6//fYbFi5cCHNzc/j4+GD06NFITk5u8b5CQ0Nhbm6OsWPHQi6Xo7i4GKmpqfD29sbgwYMhkUiwePHiJkdKM1S8sUrUjKysLHTp0kVtxsauXbuqjTwm1t3DFJqamqK+vh4lJSVqY7ka43SyPBMhaoabmxtKSkogl8tVy+4MV9gW/jlEYUFBAbZu3dom+9YVhghRE6RSKeRyOfr06YPNmzdDLpfjwoULSE5OxujRo1XrVFZWtrqNwYMH49KlSzhz5gzq6uoQFxeHrKystjoEneDlDBkMRZ1S66O/KeqUkJpJRK0bHByMiRMn4uOPP0ZcXBwCAgLQpUsXrF69WjXvzFNPPYV58+YhJiYGY8aMaXE99vb22LRpE6Kjo5GbmwtfX1+NhjzUB4YIGQyx/7h11cbatWuxdu1aAMDjjz/e6DoLFiwQNTbI1atX1T7/97//Vf33E088gcTERNF1GRpezhCRRngmQtQGli5dihMnTjT63Zo1axAUFKTjinRHpyGSlpaGOXPm4OLFi7pslkjrNm7cqO8S9EZnlzPV1dWIiIhAfX29rpokI2DAM5bQXZr7PeksRDZu3IjBgwfrqjkyAhKJBAqFQt9lkAjV1dVNTvYtKkSOHDnS6PJ9+/aJKiAjIwOXLl1CWFhYk+vIZDLk5OSo/cnLyxO1fzJONjY2yM/P59mpARMEAVVVVbh586Zaz9q7NXlPpLy8HDdv3gQArFixAr169VI7pamoqMB7772HZ599ttki5HI5IiIisH79epiYmDS5Xnx8PGJjY5vdF91f7O3tkZOT0+DxJxkWqVQKJyenJrvjN3tjdebMmarXkidMmNBgx08//fQ9C9iyZQuGDx8Od3d35OTkNLleWFgYJk2apLYsLy8PISEh92yDjJOpqSm6d++u7zJIQ02GyAMPPKAaI2Hy5Mn45ptvWtXA0aNHUVhYiM8++wz19fWQy+Xw8fHBwYMH4eLiolrPGF88IiKRj3hbGyAA1F6ZzsnJwdixY5Gent7q/RGRYREVIhkZGYiJiUFWVlaDm2Dnzp3TSmFEZBxEhci7776Lvn37YtGiRTAza33/NDc3N7V3BojI+IlKhBs3bmDfvn1qA7MQEQEi+4l4eHjg999/13YtRGSERJ2JPPLIIwgLC8PQoUNhb2+v9l14eLhWCqPGpaWl4auvvsLUqVPh5+en73KIxIVIVVWVaiQnmUym1YKoebt378a1a9dQVVXFECGDICpE7gzMQvpXVVWl9pNI3zQOEV7OELVvokLkn5cwf//9N86ePdugmzoRtT+tPhNJT0/Hxx9/3OYFEZFxafV4Iv3798f58+fbshYiMkKizkSuXLmi9lmhUCAxMRHdunXTSlFEZDxEhcjEiRPVPpuamqJnz5549913tVETAahTKmEmad0UCppsS9RSrToTIe0zk0iw/bO9DZaXlZerfjb2PQC88jzHYCHdEf02XXZ2Ng4fPoxbt27B3t4eY8eORe/evbVZGxEZAVE3Vs+dO4fx48fj/PnzEAQBGRkZmDx5Mn788Udt10dEBk7UmciGDRsaTMBz6NAhvP/++3jiiSe0VhwRGT5RZyI3btxoMFnx2LFj8eeff2qjJiIyIqJCxNnZGRcuXFBblp6erjZGKhG1T6IuZ1555RXMmTMHEyZMgIuLC27evInExEQ+4tUDs/9NIGTWxERCRLomKkTGjRsHGxsbJCQkIC0tDa6urvj3v/8NHx8fbddH/+Dh5Y2rv/wX7h799F0KEYAWPOItLS3Fm2++CXt7exw9ehQFBQXarIua0LVbN3RlT2EyIKLuicTGxmLjxo2oqKgAcHsO1Q8++ABxcXFaLY6IDJ+oEPnyyy/x2WefoWfPngCA4cOHIz4+Hnv3Nt5jkojaD1EhUl1dDVtbW7VldnZ2qK2t1UpRRGQ8RIVIQEAAIiIiUFpaCgAoKyvD6tWrMXDgQG3WRkRGQNSN1bfffhsLFiyAv78/LCwsUFtbCx8fH2zcuFHb9RGRgRMVIg4ODti3bx9ycnJQXFwMJycnODs7q61z5MgRjB07VitFEpHhatHIZm5ubujfv3+DAAGAlStXtllRRGQ8Wj084j8JgtBWu9JIWloali5dirS0NH2XQtQutH527n8wMTFpq11phJM7EelWm52JNCc+Ph5PPvkkvL298eqrr6K4uFhrbXFyJyLd0nqIXLhwAbt27cLnn3+Os2fPwsbGhk91iO4jWg+RAQMG4MiRI3Bzc0N1dTUqKysbdFwjIuPVZvdEmrux2qFDByQmJuKNN96Ak5MTVqxY0WAdmUzWYKa9vLy8tiqPiLRE1JlIaGio6uW7pqxfv77Z7wMDA3Hx4kWMHj0aixYtavB9fHw8RowYofYnJISjlhMZOlFnIpmZmfdcZ+TIkc1+b25uDgBYvHgxBgwYgL///lvtsiYsLKzB3L55eXkMEiIDJypEBg8ejGnTpmHo0KFwdHRUe5wbGhra7LYJCQm4cOECoqKiAAB1dXUwNTVFx44d1daztraGtbV1S+snIj0TFSI5OTno0qULLl26pLbcxMTkniHSr18/REVFYfLkyejbty/Wr1+PwMBA1ZkJERk3USHy6aeftrqBXr16Yc2aNXjjjTcgk8kwePBgvPPOO63eH6D5NJGcZpKo7YgKkT179jT53b3ORABgzJgxDaac0ERTU0wCnGaSSNdEhcjRo0fVPpeWliIzMxOjR48WFSJEdP9q9eXMkSNHcPz48TYviIiMS6t7rAYGBuLEiRNtWQsRGSFRZyL/7Ggml8uRkJAAOzs7rRRFRMZDVIj4+Pio9Q0RBAHW1taIjo7WWmFEZBxEhcixY8fUPkskEtjZ2UHKqRyJ2j1RIeLq6oqKigqkpKTg1q1bsLe3x9ChQw3ybVzOVUukW6JC5OrVq3jppZfwwAMPwNXVFTk5OYiJicGuXbvw6KOParvGFuFctUS6JSpEoqOjMXPmTMycOVO1bOfOnYiOjja4WfA4Vy2Rbol6xHv58mWEhYWpLQsLC8Ply5e1UhQRGQ9RIdK5c2dcu3ZNbdn169dhb2+vlaKIyHiIupwJDQ3F7NmzERoaChcXF9y8eROffvopZs2ape36iMjAiQqRGTNmwMLCAgkJCSguLoaLiwtef/11jB8/Xtv16VRaWhq++uorTJ06ldNNEIkkeozVqVOn4qmnnkJhYSG6du16X44HwjlriFpOVIiUlJRg5cqVSElJAXB7MKKnnnoKq1atQqdOnbRZn05xzhqilhN1YzUyMhLm5uZITk7GxYsXkZSUhIqKCqxevVrb9RGRgRN1JnL27FmkpqbCwsICwO3RytatW4cRI0ZotTgiMnyizkS6deuGv/76S21ZYWGhQXZ7JyLdavZM5M6wiH369MGLL76IkJAQuLi4oKioCJ9//jmGDRumkyKJyHA1GyJ3D4v44IMP4vTp06rPLi4u+P3337VXGREZhWZDRJNR3omofRDdTyQ1NRU5OTmoq6tTW26MAzXXKOSwlN5//VyI9EFUiERFReHAgQNwd3dXG4hIzORVhshSao4uK19osLxvcT4sAfxRnN/o9wBQsma3NksjMjqiQuTgwYPYv38/evXqpe16iMjIiHrEa2NjAzc3N23XQkRGqNkQqaioQEVFBaZMmYIVK1bg5s2bqmV3/hBR+9bs5cydUd4FQQAAJCUlqUZ9FwQBJiYmHJiIqJ1rNkT+Oco7EdE/NRsirq6uAIDc3NxGv5dKpZDL5fflsABEJI6opzOjRo1CfX09gP+/jLnD1NQUAwcOxHvvvQdHR0ftVElEBkvU05nXX38dw4YNw5EjR3Dp0iUkJycjMDAQs2fPxsGDB+Hk5IR33nmnye2PHz+OoKAgDBgwACEhIfjjjz/a7ADaUr3ERO0nEd2bqBDZs2cPNmzYgJ49e8Lc3Bw9evRAdHQ0EhIS0Lt3b0RGRuLcuXONbpuXl4fly5fj3Xffxblz5zBs2DAsWrSoTQ+irdxy64xyawvccuus71KIjIaoy5na2loUFRWhe/fuqmWFhYWorq4GACiVyian1Lx16xamTJmCAQMGAACmT5+O9evXo6qqCh06dFCtJ5PJIJPJ1LbNy8tr2dFoSGZrBZmtlU7bJDJ2okLkueeew4svvogZM2bAyckJeXl52Lt3L6ZPn47CwkK88cYbTQ5Q5OXlBS8vL9Xn1NRUuLq6qgUIAMTHxyM2NlaDQyEifRAVIvPnz4ezszMSExORn5+Prl27YtGiRQgODsaVK1cQEBCAF1544Z77uXLlCiIjI/Huu+82+C4sLAyTJk1SW5aXl4eQkBBxR0JEeiH6Ld6nn34aTz/9dIPlffr0QZ8+fe65fXp6OubNm4clS5Zg9OjRDb63traGtbW12HKIyEA0GyLTp0/HF198gYkTJ6o91r3bgQMH7tlIamoqli5diqioKAQFBbWuUiIySM2GyHPPPQcADebhbYns7GwsXrwYGzZswPDhw1u9HyIyTM2GSHBwMACo7lUoFArk5uaqntI0dXZyt6+//hpVVVV47bXX1Jb/8MMPsLOza1XRRGQ4RPUTqampwVtvvYX+/ftjwoQJyMzMxJgxY5CdnX3PbZcsWYIrV64gIyND7Q8DhOj+ICpEoqOjUVZWhqSkJEilUnTv3h2DBg1qtpcqEbUPop7OHD9+HMnJyejUqRNMTEwglUqxbNkyDBo0SNv1EZGBE3UmYmZmBrlcDgCqsUWqqqpgZcXenUTtnagQCQwMxKJFi3Dp0iUAwJ9//omVK1di5MiRWi2OiAyfqBB57bXX4O7ujtDQUJSXl2P8+PGwtbVt8MSFiNofUfdEzM3NMWTIEPTs2RM1NTWQSCSQSCT4+uuvjXLKCCJqO6JCJDIyEgkJCejbty8kEolqubHOO0NEbUdUiBw9ehSff/45+vbtq+16iMjIiLonYmpqit69e2u7FiIyQqJC5KWXXkJERASysrI47wwRqRF1OfPhhx+iuroa3377rWoZ550xPmlpafjqq68wdepU+Pn56bscuk+ICpGkpCRt10E6sHv3bly7dg1VVVUMEWozokLkzvwzZNyqqqrUfhK1BVH3RIiImsIQISKNMESISCMMESLSCEOEiDTCELkP1Sjk+i6B2hHR886Q8bCUmqPLyhcaLO9bnA9LAH8U5zf6PQCUrNmtzdLoPsQzESLSCEOEiDTCECEijTBEiEgjDJF2pF5iovaTqC0wRNqRW26dUW5tgVtunfVdCt1H+Ii3HZHZWkFmy7mCqG3xTISINMIQISKN6DREdu7ciVWrVumySSLSMp2EiEKhwObNm7FhwwZdNEdEOqSTG6tr165FTk4Opk2bhvr6el00SUQ6opMQefXVV+Hg4IAtW7agsLCw0XVkMhlkMpnasry8PF2UR0Qa0EmIODg43HOd+Ph4xMbG6qAaImpLBtNPJCwsDJMmTVJblpeXh5CQED1VRERiGEyIWFtbw9raWt9lEFELsZ8IEWmEIUJEGtHp5cyCBQt02RwR6QDPRIhIIwwRItIIQ4SINMIQISKNMESISCMMESLSCEOEiDTCECEijTBEiEgjDBEi0ghDhIg0whAhIo0wRIhIIwwRItIIQ4SINMIQISKNMESISCMMESLSCEOEiDTCECEijTBEiEgjDBEi0ghDhIg0whAhIo0wRIhIIwwRItIIQ4SINMIQISKNMESISCM6CZH09HSMGzcOnp6emDt3LioqKnTRLBHpgNZDpLq6GgsXLsSbb76JtLQ0WFpaYuvWrdpuloh0ROshkpaWBldXVwwZMgSWlpaYP38+EhIStN0sEemImbYbyMrKQs+ePVWfu3fvjuLiYpSWlsLGxka1XCaTQSaTqW178+ZNAEBeXl6D/ZaVlra6ppycHJhW1rR625qyklZv2xKtPUZDOr47vzulUtmqfZLhMxEEQdBmA1u3bkV+fj6ioqJUyzw8PPDDDz/A2dlZtWzLli2IjY3VZimkR3v37oWPj4++yyAt0PqZiJWVFWpra1Wf6+rqoFAoYGVlpbZeWFgYJk2apLZMLpcjOzsbPXv2hEQi0XapyMvLQ0hICPbu3asWcPcLfRyfUqlEYWEhPDw8dNIe6Z7WQ+TBBx/E999/r/qclZUFW1tbdO7cWW09a2trWFtbN9i+V69e2i6xAWdnZ7i5uem8XV3R9fH16NFDZ22R7mn9xqqfnx+ysrJw7Ngx1NTUYNu2bRg7dqy2myUiHdF6iFhZWWHr1q3YtGkTAgICUFtbi9dee03bzRKRjmj9cgYAvL29kZiYqIumiEjH2O39LtbW1pg/f36j92buB/f78ZF+aP0RLxHd33gmQkQaYYj8T2VlJcrKyvRdht7l5ubquwQyMgyR/5kxYwZu3LjR5HeHDh3ScUW6V1RU1KDDH9G9MET+p1SDd3HuFzU1NaiqqtJ3GWRk2mWIpKSkICgoCN7e3pg1axZmz56N3NxchIWF4fTp05gxYwaWLVsGX19fbN++Xd/linb27Fk8//zzCA8Ph5eXFyZPnoyffvoJ06ZNg5eXF958803k5OSgX79+qm3u/hwWFga5XA4vLy/I5XJ9HQYZmXYXIvn5+ViyZAnCw8Nx/vx5dO/eHTY2NnBxcUF8fDwCAgIA3H7PJCUlBc8//7yeK26Z8+fPw9/fH+fPn4e1tTXmzp2LtWvXIjk5GSdPnkRhYWGT28bHx8Pc3BwZGRkwNzfXYdVkzHTS2cyQpKamwtvbG4MHDwYALF68GAUFBUhPT1dbb/jw4Q1eEjQGtra2GD9+PADA09MTdnZ2qvePHnrooQbHSaSpdhciJSUlcHR0VH1u6sU/Ozs7XZbVZu5+sVEikaBjx46qz6ampmC3IGpr7e5yxtHREQUFBarPBQUFjQ7XaGJiosuy2sy96nZ0dFQbIOifA0ERtVS7C5HBgwfj0qVLOHPmDOrq6hAXF4esrCxIpVJUVlbquzytk0qlsLCwwA8//AC5XI4dO3aovjM3N4dSqVQb/4XoXtpdiNjb22PTpk2Ijo6Gr68vcnNzsWLFCjz11FOYN28ekpOT9V2i1kVERGD16tUYNmwYvL29VcsdHBzg6+sLPz8/tbM1oubw3Rki0ki7OxMhorbFECEijTBEiEgjDBEi0ghDhIg0whAhIo0wRAxIS6fZJDIEDBEDsW7dOsTHx+u7DKIWY4gYiJKS1k2iTaRv7e4tXrG+//57vP/++ygtLcWIESOQmZmJqVOnYsSIEYiJiUFqairMzMwwYcIELFy4EGZmZtiyZQv+/PNPVFVVIS0tDV27dsXSpUsxcuRIAMDp06fx/vvv48aNG3B1dcW8efMwZswYfPLJJ0hMTISJiQmys7Ph4OCAmpoarF+/XlXPqFGjEB4ejl9//RXXr19HZWUl0tPT8eCDDyIiIgKenp4AgOvXr2P16tX49ddfYWdnh3nz5iE4OFgff4XUXgjUQGZmptCvXz8hJSVFkMvlwkcffSQ88sgjwv79+4U5c+YICxYsEMrLy4X8/Hxh2rRpQmxsrCAIgrB582ahb9++wokTJwS5XC58+OGHwhNPPCHU19cLv//+u+Dh4SEcOnRIUCgUwunTpwVPT08hPT1dEARBWLZsmbB69WpBEATh/PnzgpeXl1BTUyMIgiD8/PPPgq+vryCXy4XNmzcL7u7uQkJCgiCXy4WdO3cKfn5+Qnl5uVBRUSEMGjRI2LFjhyCXy4WLFy8K/v7+wvnz5/XzF0ntAi9nGnHo0CEEBARgyJAhkEqlmDNnDpycnFBUVIQTJ05g1apV6NSpExwdHTFv3jzs27dPtW2/fv0wdOhQSKVSjB8/HoWFhaisrMShQ4fg6+uLoKAgmJmZwd/fH8HBwThw4ECD9gcMGAAbGxukpKQAAJKSkhAUFASpVAoAePzxxzF+/HhIpVK8+OKLMDc3x5kzZ3Dy5El06tQJL730EqRSKR577DFMmTIFX3zxhU7+3qh94uVMIwoKCuDk5KT6bGJiAmdnZ9VYHWPGjFF9JwgCFAqF6vX5uwczMjO7/ddbX1+PkpISuLi4qLXj5uaGs2fPNmjfxMQEwcHBOHToEEaNGoUjR46ojXnSo0cPtXWdnJxUYfXXX3/Bx8dH9b1SqcSjjz7aqr8HIjEYIo1wdnbGxYsXVZ8FQUB+fj4UCgVMTU1x6tQp1dCJFRUVKC4uhoWFRbP77Nq1a4OhCbOzs2Fvb9/o+hMmTMDkyZNx+vRpdOjQAf3791d9l5+fr1bbrVu30LVrV8hkMnh4eODLL79UW9dYB1gi48DLmUYEBwcjLS0Np06dQl1dHeLj45GXlwdnZ2cMHDgQMTExqKysREVFBcLDw/HWW2/dc59BQUFIT0/H4cOHoVQqcebMGSQmJqpuepqbm6OiokK1fq9evdC7d2+sX79eNWbqHT/++CNOnjwJhUKBjz/+GKampvD398eQIUPw119/4ZtvvkFdXR2ys7MRGhqqFipEbY0h0ohu3bph7dq1iIiIQEBAAP744w+4uLhAKpXi/fffR0VFBUaNGoXhw4fDxMQEmzZtuuc+e/TogW3btmHHjh3w8fHBO++8g3feeQeDBg0CAIwdOxbHjx/H9OnTVdtMmDABV65caRAijz32GPbs2QNfX1+cOHECO3bsgKWlJWxsbLBjxw4cOHAA/v7+mD59OkaOHIm5c+e26d8P0d04KFEjcnNzUVVVhYceeki1LCAgAOvWrVP9o9eFo0ePYteuXWo3Rrds2YLLly9j27ZtOquDqDk8E2lEQUEBQkNDkZ2dDUEQsG/fPsjlclVfDG0rLy/HlStXEBcXh2eeeUYnbRK1Fm+sNsLT0xOzZ8/GjBkzUFZWhl69emH79u3o1KmTTtq/ceMGQkND8eSTT2LChAk6aZOotXg5Q0Qa4eUMEWmEIUJEGmGIEJFGGCJEpBGGCBFp5P8AbbQ0xD6uhJ4AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 288x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "from statannotations.Annotator import Annotator\n",
    "\n",
    "with sns.plotting_context(\"notebook\", font_scale=1.1):\n",
    "    fig, ax = plt.subplots(figsize=(4, 4))\n",
    "\n",
    "    sns.set_style(\"ticks\")\n",
    "    ax = sns.barplot(\n",
    "        x=\"genotype\",\n",
    "        y=\"neighbour_count\",\n",
    "        hue=\"neigh_type\",\n",
    "        ax=ax,\n",
    "        data=tidy_gfp_df,\n",
    "        palette=[\"#95a5a6\", \"#008b68\", \"#539DC2\"],\n",
    "    )\n",
    "\n",
    "    # adjust positon of legend\n",
    "    handles, labels = ax.get_legend_handles_labels()\n",
    "    ax.legend(handles, labels, bbox_to_anchor=(1, 1), loc=\"upper left\")\n",
    "    plt.tight_layout()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "aba0a32f-f73c-4b8d-9bf9-3943f0aebab1",
   "metadata": {},
   "source": [
    "Indeed, compared to control clones, mutant clones have, on average, fewer GFP positive neighbours and more GFP negative neighbours."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "0fa8fc2c-c89d-42db-8538-520333bfe01e",
   "metadata": {},
   "outputs": [],
   "source": [
    "# hide\n",
    "nc_df = df.query('int_img_ch == \"C0\"')\n",
    "nc_df.set_index(\n",
    "    [\"int_img\", \"label\", \"GFP_clone\", \"genotype\", \"clone_status\"], inplace=True\n",
    ")\n",
    "nc_df = nc_df.filter(regex=r\"pos|tot_nc|area_um2\")\n",
    "nc_df = nc_df.drop(columns=nc_df.filter(regex=r\"GFP\"))\n",
    "nc_df[\"cell_type\"] = nc_df.filter(regex=r\"pos$\").idxmax(axis=1)\n",
    "nc_df[\"cell_type\"] = nc_df[\"cell_type\"].str.replace(\"_pos\", \"\")\n",
    "nc_df = nc_df.drop(columns=nc_df.filter(regex=r\"pos$\"))\n",
    "nc_df = nc_df.reset_index()\n",
    "# simplify column names\n",
    "nc_df.columns = [re.sub(\"pos_\", \"\", col) for col in nc_df.columns]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "71640073-196e-4cca-b709-35f62b6cf0de",
   "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>int_img</th>\n",
       "      <th>label</th>\n",
       "      <th>GFP_clone</th>\n",
       "      <th>genotype</th>\n",
       "      <th>clone_status</th>\n",
       "      <th>tot_nc</th>\n",
       "      <th>ECnc</th>\n",
       "      <th>EEnc</th>\n",
       "      <th>ECEEnc</th>\n",
       "      <th>ISCorEBnc</th>\n",
       "      <th>preECnc</th>\n",
       "      <th>area_um2</th>\n",
       "      <th>cell_type</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>a1g01p1</td>\n",
       "      <td>11</td>\n",
       "      <td>0</td>\n",
       "      <td>ctrl</td>\n",
       "      <td>OC</td>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>73.280625</td>\n",
       "      <td>EC</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>a1g01p1</td>\n",
       "      <td>12</td>\n",
       "      <td>0</td>\n",
       "      <td>ctrl</td>\n",
       "      <td>OC</td>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>98.690625</td>\n",
       "      <td>EC</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>a1g01p1</td>\n",
       "      <td>15</td>\n",
       "      <td>0</td>\n",
       "      <td>ctrl</td>\n",
       "      <td>OC</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>29.040000</td>\n",
       "      <td>ISCorEB</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>a1g01p1</td>\n",
       "      <td>17</td>\n",
       "      <td>2</td>\n",
       "      <td>ctrl</td>\n",
       "      <td>IC</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>12.175625</td>\n",
       "      <td>ISCorEB</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>a1g01p1</td>\n",
       "      <td>18</td>\n",
       "      <td>0</td>\n",
       "      <td>ctrl</td>\n",
       "      <td>OC</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>28.661875</td>\n",
       "      <td>ISCorEB</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   int_img  label  GFP_clone genotype clone_status  tot_nc  ECnc  EEnc  \\\n",
       "0  a1g01p1     11          0     ctrl           OC       4     3     0   \n",
       "1  a1g01p1     12          0     ctrl           OC       4     3     0   \n",
       "2  a1g01p1     15          0     ctrl           OC       2     0     0   \n",
       "3  a1g01p1     17          2     ctrl           IC       2     2     0   \n",
       "4  a1g01p1     18          0     ctrl           OC       2     1     0   \n",
       "\n",
       "   ECEEnc  ISCorEBnc  preECnc   area_um2 cell_type  \n",
       "0       0          1        0  73.280625        EC  \n",
       "1       0          1        0  98.690625        EC  \n",
       "2       0          2        0  29.040000   ISCorEB  \n",
       "3       0          0        0  12.175625   ISCorEB  \n",
       "4       0          1        0  28.661875   ISCorEB  "
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# hide\n",
    "nc_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b1d9e790-9e23-4137-a95d-9c0e713ba7f4",
   "metadata": {},
   "outputs": [],
   "source": [
    "# hide\n",
    "tidy_nc_df = (\n",
    "    pd.melt(\n",
    "        nc_df,\n",
    "        id_vars=[\n",
    "            \"int_img\",\n",
    "            \"label\",\n",
    "            \"cell_type\",\n",
    "            \"genotype\",\n",
    "            \"GFP_clone\",\n",
    "            \"tot_nc\",\n",
    "            \"clone_status\",\n",
    "            \"area_um2\",\n",
    "        ],\n",
    "        value_vars=[\"ECnc\", \"EEnc\", \"ECEEnc\", \"ISCorEBnc\", \"preECnc\"],\n",
    "        var_name=\"neigh_cell_type\",\n",
    "        value_name=\"neigh_count\",\n",
    "    )\n",
    "    .sort_values([\"int_img\", \"label\"])\n",
    "    .reset_index(drop=True)\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "54f82bd9-0acd-4774-91c0-87a63ccaa7b8",
   "metadata": {},
   "outputs": [],
   "source": [
    "# hide\n",
    "major_cell_types = tidy_nc_df.query(\n",
    "    'cell_type in [\"ISCorEB\", \"EC\", \"EE\"] and neigh_cell_type in [\"ISCorEBnc\", \"ECnc\", \"EEnc\"]'\n",
    ").copy()\n",
    "\n",
    "major_cell_types[\"neigh_cell_type\"] = pd.Categorical(\n",
    "    major_cell_types[\"neigh_cell_type\"], categories=[\"ISCorEBnc\", \"ECnc\", \"EEnc\"]\n",
    ")\n",
    "\n",
    "major_cell_types[\"cell_type\"] = pd.Categorical(\n",
    "    major_cell_types[\"cell_type\"], categories=[\"ISCorEB\", \"EC\", \"EE\"]\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "461272a1-8e18-4d11-a36f-b85802ec6788",
   "metadata": {},
   "outputs": [],
   "source": [
    "# hide\n",
    "import matplotlib as mpl\n",
    "mpl.rcParams['pdf.fonttype'] = 42\n",
    "mpl.rcParams['ps.fonttype'] = 42"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c46b8318-20d1-47e1-a0e0-361150c70b9f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# hide\n",
    "import seaborn as sns\n",
    "from matplotlib import pyplot as plt\n",
    "\n",
    "with sns.plotting_context(\"notebook\", font_scale=1.1):\n",
    "    fig, ax = plt.subplots(figsize=(5, 4))\n",
    "\n",
    "    sns.boxplot(\n",
    "        data=major_cell_types.query('genotype ==\"ctrl\"'),\n",
    "        x=\"cell_type\",\n",
    "        y=\"neigh_count\",\n",
    "        hue=\"neigh_cell_type\",\n",
    "        ax=ax,\n",
    "        palette=[\"#008b68\", \"#F57171\", \"#95a5a6\"],\n",
    "        fliersize=2,\n",
    "        zorder=0,\n",
    "        saturation=0.9,\n",
    "        linewidth=1.5,\n",
    "        notch=False,\n",
    "    )\n",
    "\n",
    "    for patch in ax.artists:\n",
    "        r, g, b, a = patch.get_facecolor()\n",
    "        patch.set_facecolor((r, g, b, 0.5))\n",
    "\n",
    "    ax.tick_params(axis=\"both\", which=\"major\", pad=1)\n",
    "\n",
    "    handles, labels = ax.get_legend_handles_labels()\n",
    "    ax.legend(handles, labels, bbox_to_anchor=(1, 1), loc=\"upper left\")\n",
    "    ax.set_ylim(top=8.5)\n",
    "    plt.tight_layout()\n",
    "    plt.setp(ax.spines.values(), linewidth=1)\n",
    "    sns.despine()\n",
    "    fig.savefig('/Users/ottomorris/Desktop/num_of_neighbours.pdf', dpi=300)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8b0dbe9d-dfda-4022-87e1-e4dcf1a257a4",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}