{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "09774124-8ae3-4102-934a-d38780b53eac",
   "metadata": {},
   "source": [
    "# Download expression data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "e951c648-a373-470a-a5aa-ed678de40b05",
   "metadata": {},
   "outputs": [],
   "source": [
    "import GEOparse\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import re\n",
    "from pathlib import Path\n",
    "import pylab as pl\n",
    "import seaborn as sns\n",
    "pl.rcParams['figure.figsize'] = (14, 10)\n",
    "pl.rcParams['ytick.labelsize'] = 12\n",
    "pl.rcParams['xtick.labelsize'] = 11\n",
    "pl.rcParams['axes.labelsize'] = 23\n",
    "pl.rcParams['legend.fontsize'] = 20\n",
    "sns.set_style('ticks')\n",
    "c1, c2, c3, c4 = sns.color_palette(\"Set1\", 4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "6cb55a19-bbb6-4bab-978d-0fcedc753268",
   "metadata": {},
   "outputs": [],
   "source": [
    "# !pip install GEOparse"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "cecc6a63-5b5f-436f-8178-0ae082aef0d7",
   "metadata": {},
   "outputs": [],
   "source": [
    "Dir_Expression = \"1_Expression_data/\"\n",
    "Path(Dir_Expression).mkdir(parents=True, exist_ok=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "1ece80d2-9ed3-4843-b4c5-af811b39ed17",
   "metadata": {},
   "outputs": [],
   "source": [
    "Dir_WGCNA = \"2_WGCNA_data/\"\n",
    "Path(Dir_WGCNA).mkdir(parents=True, exist_ok=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "5f0f662d-5430-4d99-8c3b-e87d097512f9",
   "metadata": {},
   "outputs": [],
   "source": [
    "Dir_GRN = \"3_GRN_data/\"\n",
    "Path(Dir_GRN).mkdir(parents=True, exist_ok=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7bdcb850-eea9-408c-96ba-7659c2b6f58f",
   "metadata": {},
   "source": [
    "## Download data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "452e20a4-9428-48b3-8f47-5513aa4d231a",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "  % Total    % Received % Xferd  Average Speed   Time    Time     Time  Current\n",
      "                                 Dload  Upload   Total   Spent    Left  Speed\n",
      "\n",
      "  0     0    0     0    0     0      0      0 --:--:-- --:--:-- --:--:--     0\n",
      "  3 6608k    3  208k    0     0   360k      0  0:00:18 --:--:--  0:00:18  360k\n",
      " 77 6608k   77 5152k    0     0  3232k      0  0:00:02  0:00:01  0:00:01 3232k\n",
      "100 6608k  100 6608k    0     0  3570k      0  0:00:01  0:00:01 --:--:-- 3570k\n",
      "  % Total    % Received % Xferd  Average Speed   Time    Time     Time  Current\n",
      "                                 Dload  Upload   Total   Spent    Left  Speed\n",
      "\n",
      "  0     0    0     0    0     0      0      0 --:--:-- --:--:-- --:--:--     0curl: (6) Could not resolve host: GSE74488_sc_expression.csv.gz\n"
     ]
    }
   ],
   "source": [
    "URL = 'https://ftp.ncbi.nlm.nih.gov/geo/series/GSE74nnn/GSE74488/suppl/GSE74488_sc_expression.csv.gz'\n",
    "!curl {URL} -O GSE74488_sc_expression.csv.gz"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "692b975e-9a42-4cb0-b331-3a3d06f97baf",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.read_csv(\"GSE74488_sc_expression.csv.gz\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "0c6e9fa6-0177-4220-add0-07ecaec531cb",
   "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>Locus</th>\n",
       "      <th>wolsc_kb2_4_1</th>\n",
       "      <th>wolsc_kb2_4_10</th>\n",
       "      <th>wolsc_kb2_4_11</th>\n",
       "      <th>wolsc_kb2_4_13</th>\n",
       "      <th>wolsc_kb2_4_14</th>\n",
       "      <th>wolsc_kb2_4_15</th>\n",
       "      <th>wolsc_kb2_4_18</th>\n",
       "      <th>wolsc_kb2_4_19</th>\n",
       "      <th>wolsc_kb2_4_22</th>\n",
       "      <th>...</th>\n",
       "      <th>wolsc_kb3_2_1</th>\n",
       "      <th>sc_0113_pa_19</th>\n",
       "      <th>sc_0113_pa_3</th>\n",
       "      <th>sc_0113_pa_44</th>\n",
       "      <th>sc_0113_pa_52</th>\n",
       "      <th>sc_0113_pa_58</th>\n",
       "      <th>sc_0113_pa_59</th>\n",
       "      <th>sc_0113_pa_60</th>\n",
       "      <th>sc_0113_pa_68</th>\n",
       "      <th>sc_0113_pa_83</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>AT1G01010</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>7.702431</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>AT1G01020</td>\n",
       "      <td>8.378906</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>4.298833</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.275709</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>3.614329</td>\n",
       "      <td>0.0</td>\n",
       "      <td>4.642478</td>\n",
       "      <td>3.406784</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>AT1G01030</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>AT1G01040</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>AT1G01046</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 239 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       Locus  wolsc_kb2_4_1  wolsc_kb2_4_10  wolsc_kb2_4_11  wolsc_kb2_4_13  \\\n",
       "0  AT1G01010       0.000000        7.702431             0.0        0.000000   \n",
       "1  AT1G01020       8.378906        0.000000             0.0        4.298833   \n",
       "2  AT1G01030       0.000000        0.000000             0.0        0.000000   \n",
       "3  AT1G01040       0.000000        0.000000             0.0        0.000000   \n",
       "4  AT1G01046       0.000000        0.000000             0.0        0.000000   \n",
       "\n",
       "   wolsc_kb2_4_14  wolsc_kb2_4_15  wolsc_kb2_4_18  wolsc_kb2_4_19  \\\n",
       "0             0.0             0.0             0.0             0.0   \n",
       "1             0.0             0.0             0.0             0.0   \n",
       "2             0.0             0.0             0.0             0.0   \n",
       "3             0.0             0.0             0.0             0.0   \n",
       "4             0.0             0.0             0.0             0.0   \n",
       "\n",
       "   wolsc_kb2_4_22  ...  wolsc_kb3_2_1  sc_0113_pa_19  sc_0113_pa_3  \\\n",
       "0             0.0  ...            0.0       0.000000           0.0   \n",
       "1             0.0  ...            0.0       2.275709           0.0   \n",
       "2             0.0  ...            0.0       0.000000           0.0   \n",
       "3             0.0  ...            0.0       0.000000           0.0   \n",
       "4             0.0  ...            0.0       0.000000           0.0   \n",
       "\n",
       "   sc_0113_pa_44  sc_0113_pa_52  sc_0113_pa_58  sc_0113_pa_59  sc_0113_pa_60  \\\n",
       "0            0.0            0.0            0.0       0.000000            0.0   \n",
       "1            0.0            0.0            0.0       3.614329            0.0   \n",
       "2            0.0            0.0            0.0       0.000000            0.0   \n",
       "3            0.0            0.0            0.0       0.000000            0.0   \n",
       "4            0.0            0.0            0.0       0.000000            0.0   \n",
       "\n",
       "   sc_0113_pa_68  sc_0113_pa_83  \n",
       "0       0.000000       0.000000  \n",
       "1       4.642478       3.406784  \n",
       "2       0.000000       0.000000  \n",
       "3       0.000000       0.000000  \n",
       "4       0.000000       0.000000  \n",
       "\n",
       "[5 rows x 239 columns]"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "79f69437-84cc-42c3-b6b3-8676753bba89",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "  % Total    % Received % Xferd  Average Speed   Time    Time     Time  Current\n",
      "                                 Dload  Upload   Total   Spent    Left  Speed\n",
      "\n",
      "  0     0    0     0    0     0      0      0 --:--:-- --:--:-- --:--:--     0\n",
      "100 26662  100 26662    0     0  32796      0 --:--:-- --:--:-- --:--:-- 32834\n",
      "  % Total    % Received % Xferd  Average Speed   Time    Time     Time  Current\n",
      "                                 Dload  Upload   Total   Spent    Left  Speed\n",
      "\n",
      "  0     0    0     0    0     0      0      0 --:--:-- --:--:-- --:--:--     0curl: (6) Could not resolve host: mmc4.xlsx\n"
     ]
    }
   ],
   "source": [
    "URL = \"https://www.cell.com/cms/10.1016/j.cell.2016.04.046/attachment/ccb8f6e8-4822-4e06-9400-2eccfd98dd56/mmc4.xlsx\"\n",
    "!curl {URL} -O mmc4.xlsx"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "8d8cf849-9d92-4d5d-bc81-9f1302aba4a5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "46hpc    86\n",
       "3hpc     67\n",
       "16hpc    55\n",
       "Uncut    30\n",
       "Name: Timepoint, dtype: int64"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Metadata = pd.read_excel('mmc4.xlsx', header=1)\n",
    "Metadata = Metadata[[\"Cell\", \"Timepoint\", \"Identity\"]]\n",
    "Metadata.Timepoint.value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1c90bb7e-5e1e-44d7-9e00-f4780f1c0ed1",
   "metadata": {},
   "source": [
    "## Uncut"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "8ef1763a-1451-425a-96c3-ed3d2b5483ed",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(30, 3)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Cell</th>\n",
       "      <th>Timepoint</th>\n",
       "      <th>Identity</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>wolsc_kb2_4_10</td>\n",
       "      <td>Uncut</td>\n",
       "      <td>Pericycle</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>wolsc_kb2_4_1</td>\n",
       "      <td>Uncut</td>\n",
       "      <td>Vasculature</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>wolsc_kb2_4_18</td>\n",
       "      <td>Uncut</td>\n",
       "      <td>Vasculature</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>wolsc_kb2_4_22</td>\n",
       "      <td>Uncut</td>\n",
       "      <td>Vasculature</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>wolsc_kb2_4_26</td>\n",
       "      <td>Uncut</td>\n",
       "      <td>Vasculature</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             Cell Timepoint     Identity\n",
       "0  wolsc_kb2_4_10     Uncut    Pericycle\n",
       "1   wolsc_kb2_4_1     Uncut  Vasculature\n",
       "2  wolsc_kb2_4_18     Uncut  Vasculature\n",
       "3  wolsc_kb2_4_22     Uncut  Vasculature\n",
       "4  wolsc_kb2_4_26     Uncut  Vasculature"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Metadata_Uncut = Metadata[Metadata[\"Timepoint\"] == \"Uncut\"]\n",
    "print(Metadata_Uncut.shape)\n",
    "Dic_uncut = {a:a + \"_\" + c for a, b, c in Metadata_Uncut.values.tolist()}\n",
    "Metadata_Uncut.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "6c612bec-a390-4beb-ade7-e2d69e1c57e4",
   "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>Gene_ID</th>\n",
       "      <th>Gene_name</th>\n",
       "      <th>wolsc_kb2_4_10_Pericycle</th>\n",
       "      <th>wolsc_kb2_4_1_Vasculature</th>\n",
       "      <th>wolsc_kb2_4_18_Vasculature</th>\n",
       "      <th>wolsc_kb2_4_22_Vasculature</th>\n",
       "      <th>wolsc_kb2_4_26_Vasculature</th>\n",
       "      <th>wolsc_kb2_4_27_Vasculature</th>\n",
       "      <th>wolsc_kb2_4_30_Vasculature</th>\n",
       "      <th>wolsc_kb2_4_41_Vasculature</th>\n",
       "      <th>...</th>\n",
       "      <th>wolsc_kb2_4_11_Unknown</th>\n",
       "      <th>wolsc_kb2_4_13_Unknown</th>\n",
       "      <th>wolsc_kb2_4_14_Unknown</th>\n",
       "      <th>wolsc_kb2_4_15_Unknown</th>\n",
       "      <th>wolsc_kb2_4_19_Unknown</th>\n",
       "      <th>wolsc_kb2_4_24_Unknown</th>\n",
       "      <th>wolsc_kb2_4_66_Unknown</th>\n",
       "      <th>wolsc_kb2_4_76_Unknown</th>\n",
       "      <th>wolsc_kb2_4_78_Unknown</th>\n",
       "      <th>wolsc_kb2_4_80_Unknown</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>AT1G01010</td>\n",
       "      <td>AT1G01010</td>\n",
       "      <td>7.702431</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>AT1G01020</td>\n",
       "      <td>AT1G01020</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>8.378906</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>4.298833</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>11.596565</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>AT1G01030</td>\n",
       "      <td>AT1G01030</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>AT1G01040</td>\n",
       "      <td>AT1G01040</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>AT1G01046</td>\n",
       "      <td>AT1G01046</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 32 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     Gene_ID  Gene_name  wolsc_kb2_4_10_Pericycle  wolsc_kb2_4_1_Vasculature  \\\n",
       "0  AT1G01010  AT1G01010                  7.702431                   0.000000   \n",
       "1  AT1G01020  AT1G01020                  0.000000                   8.378906   \n",
       "2  AT1G01030  AT1G01030                  0.000000                   0.000000   \n",
       "3  AT1G01040  AT1G01040                  0.000000                   0.000000   \n",
       "4  AT1G01046  AT1G01046                  0.000000                   0.000000   \n",
       "\n",
       "   wolsc_kb2_4_18_Vasculature  wolsc_kb2_4_22_Vasculature  \\\n",
       "0                         0.0                         0.0   \n",
       "1                         0.0                         0.0   \n",
       "2                         0.0                         0.0   \n",
       "3                         0.0                         0.0   \n",
       "4                         0.0                         0.0   \n",
       "\n",
       "   wolsc_kb2_4_26_Vasculature  wolsc_kb2_4_27_Vasculature  \\\n",
       "0                         0.0                         0.0   \n",
       "1                         0.0                         0.0   \n",
       "2                         0.0                         0.0   \n",
       "3                         0.0                         0.0   \n",
       "4                         0.0                         0.0   \n",
       "\n",
       "   wolsc_kb2_4_30_Vasculature  wolsc_kb2_4_41_Vasculature  ...  \\\n",
       "0                         0.0                         0.0  ...   \n",
       "1                         0.0                         0.0  ...   \n",
       "2                         0.0                         0.0  ...   \n",
       "3                         0.0                         0.0  ...   \n",
       "4                         0.0                         0.0  ...   \n",
       "\n",
       "   wolsc_kb2_4_11_Unknown  wolsc_kb2_4_13_Unknown  wolsc_kb2_4_14_Unknown  \\\n",
       "0                     0.0                0.000000                     0.0   \n",
       "1                     0.0                4.298833                     0.0   \n",
       "2                     0.0                0.000000                     0.0   \n",
       "3                     0.0                0.000000                     0.0   \n",
       "4                     0.0                0.000000                     0.0   \n",
       "\n",
       "   wolsc_kb2_4_15_Unknown  wolsc_kb2_4_19_Unknown  wolsc_kb2_4_24_Unknown  \\\n",
       "0                     0.0                     0.0                     0.0   \n",
       "1                     0.0                     0.0                     0.0   \n",
       "2                     0.0                     0.0                     0.0   \n",
       "3                     0.0                     0.0                     0.0   \n",
       "4                     0.0                     0.0                     0.0   \n",
       "\n",
       "   wolsc_kb2_4_66_Unknown  wolsc_kb2_4_76_Unknown  wolsc_kb2_4_78_Unknown  \\\n",
       "0                     0.0                     0.0                0.000000   \n",
       "1                     0.0                     0.0               11.596565   \n",
       "2                     0.0                     0.0                0.000000   \n",
       "3                     0.0                     0.0                0.000000   \n",
       "4                     0.0                     0.0                0.000000   \n",
       "\n",
       "   wolsc_kb2_4_80_Unknown  \n",
       "0                     0.0  \n",
       "1                     0.0  \n",
       "2                     0.0  \n",
       "3                     0.0  \n",
       "4                     0.0  \n",
       "\n",
       "[5 rows x 32 columns]"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_Uncut = df[['Locus'] + list(Metadata_Uncut.Cell.tolist())]\n",
    "# Gene_ID, GeneName\n",
    "df_Uncut.to_csv(Dir_Expression+\"Expr_Uncut.csv\", index=False)\n",
    "df_Uncut = df_Uncut.rename(columns={\"Locus\":\"Gene_ID\"})\n",
    "df_Uncut = df_Uncut.copy()\n",
    "df_Uncut['Gene_name'] = df_Uncut['Gene_ID']\n",
    "# Rearrange\n",
    "df_Uncut_cols = ['Gene_ID', 'Gene_name']+[i for i in list(df_Uncut) if i not in ['Gene_ID', 'Gene_name']]\n",
    "df_Uncut = df_Uncut[df_Uncut_cols]\n",
    "df_Uncut = df_Uncut.rename(columns=Dic_uncut)\n",
    "df_Uncut.to_csv(Dir_WGCNA+\"WGCNA_input_Uncut.csv\", index=False)\n",
    "df_Uncut.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a407b5f4-81f6-4645-9bad-3b6325b045bf",
   "metadata": {},
   "source": [
    "## 3hpc"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "8b2bdb8e-596d-4665-bb61-246af634e4d3",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(67, 3)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Cell</th>\n",
       "      <th>Timepoint</th>\n",
       "      <th>Identity</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>30</th>\n",
       "      <td>sc_1228_pa_30</td>\n",
       "      <td>3hpc</td>\n",
       "      <td>Epidermis\\LRC</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31</th>\n",
       "      <td>wolsc_kb2_3_13</td>\n",
       "      <td>3hpc</td>\n",
       "      <td>Mixed distal</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32</th>\n",
       "      <td>wolsc_kb2_3_14</td>\n",
       "      <td>3hpc</td>\n",
       "      <td>Mixed distal</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33</th>\n",
       "      <td>wolsc_kb2_3_2</td>\n",
       "      <td>3hpc</td>\n",
       "      <td>Mixed distal</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>34</th>\n",
       "      <td>wolsc_kb2_3_27</td>\n",
       "      <td>3hpc</td>\n",
       "      <td>Mixed distal</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              Cell Timepoint       Identity\n",
       "30   sc_1228_pa_30      3hpc  Epidermis\\LRC\n",
       "31  wolsc_kb2_3_13      3hpc   Mixed distal\n",
       "32  wolsc_kb2_3_14      3hpc   Mixed distal\n",
       "33   wolsc_kb2_3_2      3hpc   Mixed distal\n",
       "34  wolsc_kb2_3_27      3hpc   Mixed distal"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Metadata_3hpc = Metadata[Metadata[\"Timepoint\"] == \"3hpc\"]\n",
    "print(Metadata_3hpc.shape)\n",
    "Dic_3hpc = {a:a + \"_\" + c for a, b, c in Metadata_3hpc.values.tolist()}\n",
    "Metadata_3hpc.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "a9f7c0eb-dce6-4cda-90e4-ed3af6568ed6",
   "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>Gene_ID</th>\n",
       "      <th>Gene_name</th>\n",
       "      <th>sc_1228_pa_30_Epidermis\\LRC</th>\n",
       "      <th>wolsc_kb2_3_13_Mixed distal</th>\n",
       "      <th>wolsc_kb2_3_14_Mixed distal</th>\n",
       "      <th>wolsc_kb2_3_2_Mixed distal</th>\n",
       "      <th>wolsc_kb2_3_27_Mixed distal</th>\n",
       "      <th>wolsc_kb2_3_51_Mixed distal</th>\n",
       "      <th>sc_1228_pa_14_Mixed distal</th>\n",
       "      <th>sc_1228_pa_86_Mixed distal</th>\n",
       "      <th>...</th>\n",
       "      <th>sc_1228_pb_5_Unknown</th>\n",
       "      <th>sc_1228_pb_70_Unknown</th>\n",
       "      <th>sc_1228_pb_78_Unknown</th>\n",
       "      <th>sc_1228_pb_86_Unknown</th>\n",
       "      <th>sc_1228_pb_93_Unknown</th>\n",
       "      <th>sc_1228_pa_36_Unknown</th>\n",
       "      <th>sc_1228_pa_57_Unknown</th>\n",
       "      <th>sc_1228_pa_77_Unknown</th>\n",
       "      <th>sc_1228_pa_78_Unknown</th>\n",
       "      <th>sc_1228_pa_85_Unknown</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>AT1G01010</td>\n",
       "      <td>AT1G01010</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>3.829904</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>AT1G01020</td>\n",
       "      <td>AT1G01020</td>\n",
       "      <td>0.0</td>\n",
       "      <td>7.092747</td>\n",
       "      <td>5.949744</td>\n",
       "      <td>7.912041</td>\n",
       "      <td>0.0</td>\n",
       "      <td>6.881387</td>\n",
       "      <td>3.328156</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.212596</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>AT1G01030</td>\n",
       "      <td>AT1G01030</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>AT1G01040</td>\n",
       "      <td>AT1G01040</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>3.963564</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>AT1G01046</td>\n",
       "      <td>AT1G01046</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 69 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     Gene_ID  Gene_name  sc_1228_pa_30_Epidermis\\LRC  \\\n",
       "0  AT1G01010  AT1G01010                          0.0   \n",
       "1  AT1G01020  AT1G01020                          0.0   \n",
       "2  AT1G01030  AT1G01030                          0.0   \n",
       "3  AT1G01040  AT1G01040                          0.0   \n",
       "4  AT1G01046  AT1G01046                          0.0   \n",
       "\n",
       "   wolsc_kb2_3_13_Mixed distal  wolsc_kb2_3_14_Mixed distal  \\\n",
       "0                     0.000000                     0.000000   \n",
       "1                     7.092747                     5.949744   \n",
       "2                     0.000000                     0.000000   \n",
       "3                     0.000000                     0.000000   \n",
       "4                     0.000000                     0.000000   \n",
       "\n",
       "   wolsc_kb2_3_2_Mixed distal  wolsc_kb2_3_27_Mixed distal  \\\n",
       "0                    3.829904                          0.0   \n",
       "1                    7.912041                          0.0   \n",
       "2                    0.000000                          0.0   \n",
       "3                    0.000000                          0.0   \n",
       "4                    0.000000                          0.0   \n",
       "\n",
       "   wolsc_kb2_3_51_Mixed distal  sc_1228_pa_14_Mixed distal  \\\n",
       "0                     0.000000                    0.000000   \n",
       "1                     6.881387                    3.328156   \n",
       "2                     0.000000                    0.000000   \n",
       "3                     0.000000                    0.000000   \n",
       "4                     0.000000                    0.000000   \n",
       "\n",
       "   sc_1228_pa_86_Mixed distal  ...  sc_1228_pb_5_Unknown  \\\n",
       "0                         0.0  ...                   0.0   \n",
       "1                         0.0  ...                   0.0   \n",
       "2                         0.0  ...                   0.0   \n",
       "3                         0.0  ...                   0.0   \n",
       "4                         0.0  ...                   0.0   \n",
       "\n",
       "   sc_1228_pb_70_Unknown  sc_1228_pb_78_Unknown  sc_1228_pb_86_Unknown  \\\n",
       "0                    0.0                    0.0                    0.0   \n",
       "1                    0.0                    0.0                    0.0   \n",
       "2                    0.0                    0.0                    0.0   \n",
       "3                    0.0                    0.0                    0.0   \n",
       "4                    0.0                    0.0                    0.0   \n",
       "\n",
       "   sc_1228_pb_93_Unknown  sc_1228_pa_36_Unknown  sc_1228_pa_57_Unknown  \\\n",
       "0                    0.0                    0.0               0.000000   \n",
       "1                    0.0                    0.0               0.000000   \n",
       "2                    0.0                    0.0               0.000000   \n",
       "3                    0.0                    0.0               3.963564   \n",
       "4                    0.0                    0.0               0.000000   \n",
       "\n",
       "   sc_1228_pa_77_Unknown  sc_1228_pa_78_Unknown  sc_1228_pa_85_Unknown  \n",
       "0                    0.0                    0.0               0.000000  \n",
       "1                    0.0                    0.0               2.212596  \n",
       "2                    0.0                    0.0               0.000000  \n",
       "3                    0.0                    0.0               0.000000  \n",
       "4                    0.0                    0.0               0.000000  \n",
       "\n",
       "[5 rows x 69 columns]"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_3hpc = df[['Locus'] + list(Metadata_3hpc.Cell.tolist())]\n",
    "df_3hpc.to_csv(Dir_Expression+\"Expr_3hpc.csv\", index=False)\n",
    "# Gene_ID, GeneName\n",
    "df_3hpc = df_3hpc.rename(columns={\"Locus\":\"Gene_ID\"})\n",
    "df_3hpc = df_3hpc.copy()\n",
    "df_3hpc['Gene_name'] = df_3hpc['Gene_ID']\n",
    "# Rearrange\n",
    "df_3hpc_cols = ['Gene_ID', 'Gene_name']+[i for i in list(df_3hpc) if i not in ['Gene_ID', 'Gene_name']]\n",
    "df_3hpc = df_3hpc[df_3hpc_cols]\n",
    "df_3hpc = df_3hpc.rename(columns=Dic_3hpc)\n",
    "df_3hpc.to_csv(Dir_WGCNA+\"WGCNA_input_3hpc.csv\", index=False)\n",
    "df_3hpc.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7a77a4bd-9e1c-46f1-a1e1-4134bccdd13c",
   "metadata": {},
   "source": [
    "END"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}