figure1d_weights_cshl.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# GENERAL THINGS FOR COMPUTING AND PLOTTING\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import os, sys, time\n",
    "from datetime import datetime\n",
    "from datetime import timedelta\n",
    "import scipy as sp\n",
    "\n",
    "# visualisation\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "sns.set(style=\"ticks\", context=\"talk\")\n",
    "\n",
    "# ibl specific things\n",
    "import datajoint as dj\n",
    "from ibl_pipeline import reference, subject, action, acquisition, data, behavior\n",
    "from ibl_pipeline.analyses import behavior as behavioral_analyses\n",
    "\n",
    "# set path for saving figures\n",
    "figpath = 'figures' # put in the same folder, but add to .gitignore for now"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['CSHL022',\n",
       " 'CSHL024',\n",
       " 'CSHL025',\n",
       " 'CSHL026',\n",
       " 'CSHL027',\n",
       " 'CSHL029',\n",
       " 'CSHL031',\n",
       " 'CSHL045',\n",
       " 'CSHL046',\n",
       " 'CSHL058',\n",
       " 'CSHL059',\n",
       " 'CSHL060',\n",
       " 'CSHL061',\n",
       " 'CSHL062',\n",
       " 'CSHL063',\n",
       " 'CSHL064',\n",
       " 'CSHL065',\n",
       " 'CSHL066',\n",
       " 'CSHL067',\n",
       " 'CSHL068',\n",
       " 'CSHL069',\n",
       " 'CSHL070',\n",
       " 'CSHL071',\n",
       " 'CSHL072',\n",
       " 'CSHL073',\n",
       " 'CSHL074',\n",
       " 'CSHL075',\n",
       " 'CSHL_018',\n",
       " 'CSHL_019']"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "########### SUBJECTS ########### \n",
    "\n",
    "subj = subject.Subject * subject.SubjectLab  & 'lab_name=\"churchlandlab\"'\n",
    "\n",
    "########### WEIGHINGS ########### \n",
    "# grab the data from the start of baseline weights until lab shutdown\n",
    "weight_with_date = subj * action.Weighing.proj('weight', session_date='DATE(weighing_time)')\n",
    "weight_date = (dj.U('subject_uuid', 'session_date') & weight_with_date) \\\n",
    "& 'session_date BETWEEN \"2020-02-22\" AND \"2020-03-31\"'\n",
    "#baseline_date = (dj.U('subject_uuid', 'session_date') & weight_with_date) & 'session_date < \"2020-02-25\"'\n",
    "\n",
    "avg_weight_date = weight_date.aggr(weight_with_date, avg_weight='AVG(weight)') \n",
    "#avg_baseline_weight_date = baseline_date.aggr(weight_with_date, avg_weight = 'AVG(weight)')\n",
    "\n",
    "b = subj * avg_weight_date \n",
    "#bb = subj * avg_baseline_weight_date\n",
    "weight_df = pd.DataFrame(b.fetch(as_dict=True))\n",
    "#baseline_df = pd.DataFrame(bb.fetch(as_dict=True))\n",
    "sorted(weight_df.subject_nickname.unique())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "## save to file for archiving\n",
    "weight_df[['subject_uuid', 'subject_nickname', 'subject_birth_date', 'sex', 'subject_strain','subject_line', \n",
    "           'session_date', 'avg_weight']].to_csv('data/figure1D_cshl_weights.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "########### WEIGHINGS NORMALIZATION and ########### \n",
    "########### CONVERSION OF SESSION DATES INTO SESSION NUMBERS ########### \n",
    "\n",
    "normalized_df = weight_df.copy()\n",
    "\n",
    "#  how old are these mice on the first day of the intervention?\n",
    "# 3 March was the first day with CA water\n",
    "normalized_df['session_age'] = (datetime(2020, 3, 2).date() - normalized_df.subject_birth_date)\n",
    "normalized_df['session_age'].describe()\n",
    "\n",
    "########### ADD JAX DATA ########### \n",
    "jax_weights = pd.read_csv('data/jax_mouse_weights.csv')\n",
    "jax_weights.head(n=3)\n",
    "\n",
    "for weeknr, group in jax_weights.groupby(['Weeks']):\n",
    "    \n",
    "    if 44 <= group.Weeks.mean().item() <= 47:\n",
    "        normalized_df = normalized_df.append({'subject_nickname':'Jax_male',\n",
    "                              'sex':'M',\n",
    "                              'session_date':datetime(2020, 3, 2).date() + \n",
    "                                              timedelta(days=(group.Weeks.mean().item() - 44) * 7),\n",
    "                              'avg_weight':group.Males.mean().item()}, ignore_index=True)\n",
    "        \n",
    "        normalized_df = normalized_df.append({'subject_nickname':'Jax_female',\n",
    "                              'sex':'F',\n",
    "                              'session_date':datetime(2020, 3, 2).date() + \n",
    "                                              timedelta(days=(group.Weeks.mean().item() - 44) * 7),\n",
    "                              'avg_weight':group.Females.mean().item()}, ignore_index=True)\n",
    "normalized_df\n",
    "# day 0 is the start of the experiment\n",
    "normalized_df['session_day'] = (normalized_df['session_date'] - (datetime(2020, 3, 2)).date()).dt.days\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/anaconda3/envs/iblenv/lib/python3.6/site-packages/ipykernel_launcher.py:3: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  This is separate from the ipykernel package so we can avoid doing imports until\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array(['CSHL063', 'CSHL065', 'CSHL066', 'CSHL067', 'CSHL068', 'CSHL069',\n",
       "       'CSHL070', 'CSHL072', 'CSHL073', 'CSHL074', 'CSHL075',\n",
       "       'Jax_female', 'Jax_male'], dtype=object)"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "normalized2 = pd.DataFrame([])\n",
    "for index, sjdat in normalized_df.groupby(['subject_nickname']):\n",
    "    sjdat['avg_weight'] = sjdat['avg_weight'] / sjdat.loc[sjdat.session_day < 1, 'avg_weight'].mean()\n",
    "    normalized2 = normalized2.append(sjdat)\n",
    "\n",
    "########### CONDITIONS ########### \n",
    "CA = 'Citric Acid'\n",
    "Jax = 'Jax'\n",
    "\n",
    "# condition_map = {'CSHL061':CA, 'CSHL062':CA, 'CSHL063':CA, 'CSHL064':CA, \n",
    "# 'CSHL065': CA, 'CSHL066':CA, 'CSHL067':CA, 'CSHL068':CA, 'CSHL069':CA,\n",
    "# 'CSHL070':CA, 'CSHL071':CA, 'CSHL072':CA, 'CSHL073':CA, \n",
    "# 'CSHL074':CA, 'CSHL075':CA}\n",
    "# 071 was overweight at the start\n",
    "# CSHL061, 062, 064 had surgery during the CA experiment - exclude\n",
    "\n",
    "# remove mice that have dropped out\n",
    "condition_map = {'CSHL063':CA, 'CSHL065': CA, 'CSHL066':CA, 'CSHL067':CA, 'CSHL068':CA, 'CSHL069':CA,\n",
    "                 'CSHL070':CA, 'CSHL072':CA, 'CSHL073':CA, 'CSHL074':CA, 'CSHL075':CA, \n",
    "                 'Jax_male':Jax, 'Jax_female':Jax}\n",
    "\n",
    "normalized2['condition'] = normalized2.subject_nickname.map(condition_map)\n",
    "\n",
    "# choose only these animals\n",
    "normalized2 = normalized2.loc[normalized2.subject_nickname.isin(condition_map.keys()), :]\n",
    "normalized2.groupby(['subject_nickname'])['sex'].unique()\n",
    "normalized2['subject_nickname'].unique()\n",
    "#normalized2.groupby(['sex'])['subject_nickname'].unique()\n",
    "#normalized2.subject_nickname.nunique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "# # compute age range\n",
    "# normalized_df['session_age'] = (normalized_df.session_date - normalized_df.subject_birth_date)\n",
    "# normalized_df.session_age.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    244.000000\n",
       "mean       0.894621\n",
       "std        0.080425\n",
       "min        0.716134\n",
       "25%        0.825746\n",
       "50%        0.886780\n",
       "75%        0.986811\n",
       "max        1.043427\n",
       "Name: avg_weight, dtype: float64"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# what's the maximum weight loss?\n",
    "normalized2.avg_weight.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "subject_nickname\n",
       "CSHL063       0.716134\n",
       "CSHL065       0.788118\n",
       "CSHL066       0.801421\n",
       "CSHL067       0.767575\n",
       "CSHL068       0.787897\n",
       "CSHL069       0.771154\n",
       "CSHL070       0.810255\n",
       "CSHL072       0.770467\n",
       "CSHL073       0.797905\n",
       "CSHL074       0.867459\n",
       "CSHL075       0.830707\n",
       "Jax_female    1.000000\n",
       "Jax_male      1.000000\n",
       "Name: avg_weight, dtype: float64"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "normalized2.groupby(['subject_nickname'])['avg_weight'].min()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 259.2x324 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# REMOVE ONE OVERWEIGHT MOUSE\n",
    "#normalized3 = normalized2[~normalized2['subject_nickname'].isin(['CSHL071'])]\n",
    "f, ax = plt.subplots(ncols=1, nrows=1, figsize=(3.6,4.5))\n",
    "    \n",
    "palette = sns.xkcd_palette(['black', 'raspberry'])\n",
    "sns.lineplot(x=\"session_day\", y=\"avg_weight\", style='condition', hue='condition',\n",
    "             markers={'Jax':'.', 'Citric Acid':'v'}, \n",
    "             err_style='bars', marker='v', hue_order=['Jax', 'Citric Acid'], dashes=False,\n",
    "             ci=68, data=normalized2, ax=ax, legend=False, palette=palette)\n",
    "ax.set_xlabel(\"Days from protocol start\")    \n",
    "ax.set_ylabel(\"Weight (normalized)\")     \n",
    "ax.set(ylim=[0.79, 1.05], yticks=[0.8, 0.9, 1], xticks=[-7, 0, 7, 14], \n",
    "       xticklabels=['', '0', '7', '14'], xlim=[-10, 17])\n",
    "#ax.text(-5, 0.85, 'n = %d'%normalized3.subject_nickname.nunique(), {'color':'k', 'fontsize':16})\n",
    "ax.set_title('Cohort 2, n = %d\\n(6-17 months old)'%(normalized2.subject_nickname.nunique()-2), \n",
    "            fontsize=16, fontstyle='italic')\n",
    "\n",
    "ax.text(3, 0.94,'Expected growth\\nC57BL/6J', {'color':palette[0], 'fontsize':14})\n",
    "ax.text(-7, 0.82,'Free\\nCA water', {'color':palette[1], 'fontsize':14})\n",
    "\n",
    "f.tight_layout()\n",
    "sns.despine(trim=True)\n",
    "f.savefig(os.path.join(figpath, \"citricacid_study1d.pdf\"))\n",
    "f.savefig(os.path.join(figpath, \"citricacid_study1d.png\"), dpi=600)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 259.2x324 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# REMOVE ONE OVERWEIGHT MOUSE\n",
    "#normalized3 = normalized2[~normalized2['subject_nickname'].isin(['CSHL071'])]\n",
    "f, ax = plt.subplots(ncols=1, nrows=1, figsize=(3.6,4.5))\n",
    "    \n",
    "palette = sns.xkcd_palette(['raspberry'])\n",
    "sns.lineplot(x=\"session_day\", y=\"avg_weight\", hue='condition', units='subject_nickname',estimator=None,\n",
    "             err_style='bars', marker='v', alpha=0.4, palette=palette,\n",
    "            markers={'Jax':'o', 'Citric Acid':'v'}, \n",
    "             style='condition', dashes=False, markersize=5,\n",
    "             hue_order=['Citric Acid'], \n",
    "             ci=68, data=normalized2, ax=ax, legend=False)\n",
    "ax.set_xlabel(\"Days from protocol start\")    \n",
    "ax.set_ylabel(\"Weight (normalized)\")     \n",
    "ax.set(ylim=[0.71, 1.15], yticks=[0.7, 0.8, 0.9, 1, 1.1],xticks=[-7, 0, 7, 14], \n",
    "       xticklabels=['', '0', '7', '14'], xlim=[-10, 17])\n",
    "#ax.text(-5, 0.85, 'n = %d'%normalized3.subject_nickname.nunique(), {'color':'k', 'fontsize':16})\n",
    "ax.set_title('Cohort 2, n = %d\\n(6-17 months old)'%(normalized2.subject_nickname.nunique()-2), \n",
    "            fontsize=16, fontstyle='italic')\n",
    "f.tight_layout()\n",
    "sns.despine(trim=True)\n",
    "f.savefig(os.path.join(figpath, \"citricacid_study1d_individualmice.pdf\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    11.00\n",
       "mean      0.83\n",
       "std       0.04\n",
       "min       0.78\n",
       "25%       0.80\n",
       "50%       0.82\n",
       "75%       0.86\n",
       "max       0.91\n",
       "Name: avg_weight, dtype: float64"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# what were the weights in the last week?\n",
    "normalized3 = normalized2.loc[(normalized2.session_day > 7) & ~(normalized2['condition'].str.contains('Jax')), \n",
    "                              :].groupby(['subject_nickname'])['avg_weight'].mean()\n",
    "normalized3.describe().round(2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "thirst = pd.read_csv('data/figure1D_cshl_thirst.csv', parse_dates=['Date'])\n",
    "thirst['date'] = thirst['Date'].dt.date\n",
    "thirst['session_day'] = (thirst['date'] - (datetime(2020, 3, 2)).date()).dt.days\n",
    "\n",
    "# unstack\n",
    "thirst = pd.melt(thirst, id_vars=['session_day'], value_vars=['Avg_permouse_1', 'Av_permouse_2', \n",
    "                                                       'Avg_permouse_3'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "# f, ax = plt.subplots(ncols=1, nrows=1, figsize=(4, 4))    \n",
    "# palette = sns.xkcd_palette(['black', 'raspberry'])\n",
    "# sns.lineplot(x=\"session_day\", y=\"value\", \n",
    "#              err_style='bars', marker='o', \n",
    "#              ci=68, data=thirst, ax=ax, legend=False, color=palette[1])\n",
    "# ax.set_xlabel(\"Days from protocol start\")    \n",
    "# ax.set_ylabel(\"2% CA water consumed\")     \n",
    "# ax.set(ylim=[0,6], yticks=[0, 3, 6], xticks=[-7, 0, 7, 14], xlim=[-1, 17])\n",
    "# #ax.text(-5, 0.85, 'n = %d'%normalized3.subject_nickname.nunique(), {'color':'k', 'fontsize':16})\n",
    "# ax.set_title('Cohort 3, n = %d (6-17 months)'%(normalized2.subject_nickname.nunique()-2), \n",
    "#             fontsize=16, fontstyle='italic')\n",
    "# f.tight_layout()\n",
    "# sns.despine(trim=True)\n",
    "# #f.savefig(os.path.join(figpath, \"citricacid_weights_churchlandlab.pdf\"))\n",
    "# f.savefig(os.path.join(figpath, \"suppfig2b.png\"), dpi=600)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "# #normalized3 = normalized2[~normalized2['subject_nickname'].isin(['CSHL071'])]\n",
    "# f, ax = plt.subplots(ncols=2, nrows=1, figsize=(8, 4))\n",
    "    \n",
    "# palette = sns.xkcd_palette(['black', 'raspberry'])\n",
    "# sns.lineplot(x=\"session_day\", y=\"avg_weight\", hue='condition', \n",
    "#              err_style='bars', marker='o', hue_order=['Jax', 'Citric Acid'],\n",
    "#              ci=68, data=normalized2, ax=ax[0], legend=False, palette=palette)\n",
    "# ax[0].set_xlabel(\"Days from protocol start\")    \n",
    "# ax[0].set_ylabel(\"Weight (norm.)\")     \n",
    "# ax[0].set(ylim=[0.79, 1.08], yticks=[0.8, 0.9, 1], xticks=[-7, 0, 7, 14], xlim=[-10, 17])\n",
    "# #ax.text(-5, 0.85, 'n = %d'%normalized3.subject_nickname.nunique(), {'color':'k', 'fontsize':16})\n",
    "# #ax[0].set_title()\n",
    "\n",
    "# sns.lineplot(x=\"session_day\", y=\"value\", \n",
    "#              err_style='bars', marker='o', \n",
    "#              ci=68, data=thirst, ax=ax[1], legend=False, color=palette[1])\n",
    "# ax[1].set_xlabel(\"Days from protocol start\")    \n",
    "# ax[1].set_ylabel(\"2% CA water consumed\")     \n",
    "# ax[1].set(ylim=[0,6], yticks=[0, 2, 4], xticks=[-7, 0, 7, 14], xlim=[-1, 17])\n",
    "# #ax.text(-5, 0.85, 'n = %d'%normalized3.subject_nickname.nunique(), {'color':'k', 'fontsize':16})\n",
    "# # ax=ax[1].set_title('Cohort 3, n = %d (6-17 months)'%(normalized3.subject_nickname.nunique()-2), \n",
    "# #             fontsize=16, fontstyle='italic')\n",
    "\n",
    "# f.suptitle('Cohort 3, n = %d (6-17 months)'%(normalized2.subject_nickname.nunique()-2), \n",
    "#             fontsize=16, fontstyle='italic')\n",
    "\n",
    "# f.tight_layout()\n",
    "# sns.despine(trim=True)\n",
    "# #f.savefig(os.path.join(figpath, \"citricacid_weights_churchlandlab.pdf\"))\n",
    "# f.savefig(os.path.join(figpath, \"suppfig2_2panels.png\"), dpi=600)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python [conda env:iblenv] *",
   "language": "python",
   "name": "conda-env-iblenv-py"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}