explore_bias.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### just trying to take a glance at Hi-C biases associated with allele1/2 to try to understand 2x difference in coverage"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "require([\"codemirror/keymap/sublime\", \"notebook/js/cell\", \"base/js/namespace\"],\n",
       "    function(sublime_keymap, cell, IPython) {\n",
       "        cell.Cell.options_default.cm_config.keyMap = 'sublime';    \n",
       "        var cells = IPython.notebook.get_cells();    \n",
       "        for(var cl=0; cl< cells.length ; cl++){\n",
       "            cells[cl].code_mirror.setOption('keyMap', 'sublime');\n",
       "        }\n",
       "    }\n",
       ");\n"
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%%javascript\n",
    "require([\"codemirror/keymap/sublime\", \"notebook/js/cell\", \"base/js/namespace\"],\n",
    "    function(sublime_keymap, cell, IPython) {\n",
    "        cell.Cell.options_default.cm_config.keyMap = 'sublime';    \n",
    "        var cells = IPython.notebook.get_cells();    \n",
    "        for(var cl=0; cl< cells.length ; cl++){\n",
    "            cells[cl].code_mirror.setOption('keyMap', 'sublime');\n",
    "        }\n",
    "    }\n",
    ");"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "%load_ext autoreload\n",
    "%autoreload 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "import matplotlib\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib.gridspec as gridspec\n",
    "from matplotlib import gridspec\n",
    "from matplotlib import colors\n",
    "\n",
    "import os\n",
    "import re\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import cooler\n",
    "\n",
    "import itertools\n",
    "\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "# cooltools used dekkerlab-fork, develop-branch, commit-af508bba9c6577b8271b716ba27c330bead15981\n",
    "from cooler.tools import split, partition\n",
    "from functools import partial\n",
    "from cooltools import expected\n",
    "\n",
    "import bioframe\n",
    "from bioframe.io import resources\n",
    "from bioframe import io\n",
    "\n",
    "from cooltools import numutils\n",
    "from cooltools.eigdecomp import cooler_cis_eig"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "from helper_module import fillcolor_compartment_style, to_uscs, bar_signal, random_test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "cooler_names = [\n",
    "    'all-allele-spe-2percent-highcis.galGal5.mapq_30.1000.mcool',\n",
    "#     'all-allele-spe-1percent-highcis.galGal5.mapq_30.1000.mcool',\n",
    "    'MNP-DT40-1-3-16-2p-R1-T1__galGal5.galGal5.mapq_30.1000.mcool',\n",
    "#     'MNP-DT40-1-3-16-R2-T1__galGal5.galGal5.mapq_30.1000.mcool',\n",
    "    'MNP-DT40-1-3-17-2p-R1-T1__galGal5.galGal5.mapq_30.1000.mcool',\n",
    "#     'MNP-DT40-1-3-17-R1-T1__galGal5.galGal5.mapq_30.1000.mcool',\n",
    "    'MNP-DT40-WT1-R1-T1__galGal5.galGal5.mapq_30.1000.mcool',\n",
    "    'MNP-DT40-WT2-R1-T1__galGal5.galGal5.mapq_30.1000.mcool']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### generate some nicer sample names from coolers ..."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "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>sample</th>\n",
       "      <th>mcool</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>pool-2percent-highcis</td>\n",
       "      <td>all-allele-spe-2percent-highcis.galGal5.mapq_3...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>clone-16-2p</td>\n",
       "      <td>MNP-DT40-1-3-16-2p-R1-T1__galGal5.galGal5.mapq...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>clone-17-2p</td>\n",
       "      <td>MNP-DT40-1-3-17-2p-R1-T1__galGal5.galGal5.mapq...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>WT1</td>\n",
       "      <td>MNP-DT40-WT1-R1-T1__galGal5.galGal5.mapq_30.10...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>WT2</td>\n",
       "      <td>MNP-DT40-WT2-R1-T1__galGal5.galGal5.mapq_30.10...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                  sample                                              mcool\n",
       "0  pool-2percent-highcis  all-allele-spe-2percent-highcis.galGal5.mapq_3...\n",
       "1            clone-16-2p  MNP-DT40-1-3-16-2p-R1-T1__galGal5.galGal5.mapq...\n",
       "2            clone-17-2p  MNP-DT40-1-3-17-2p-R1-T1__galGal5.galGal5.mapq...\n",
       "3                    WT1  MNP-DT40-WT1-R1-T1__galGal5.galGal5.mapq_30.10...\n",
       "4                    WT2  MNP-DT40-WT2-R1-T1__galGal5.galGal5.mapq_30.10..."
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# multiqc learnings here, btw!...\n",
    "renamings = {\"all-allele-spe\":\"pool\",\n",
    "            \".galGal5.mapq_30.1000.mcool\":\"\",\n",
    "            \"MNP-DT40-\":\"\",\n",
    "            \"-T1__galGal5\":\"\",\n",
    "            \"-R1\":\"\",\n",
    "            \"1-3-\":\"clone-\"}\n",
    "\n",
    "sample_names = []\n",
    "for c in cooler_names:\n",
    "    for rf,rt in renamings.items():\n",
    "        c = re.sub(rf,rt,c)\n",
    "    sample_names.append(c)\n",
    "\n",
    "# sample_names\n",
    "\n",
    "# form a df of samples here:\n",
    "samples = pd.DataFrame({\"sample\":sample_names,\"mcool\":cooler_names})\n",
    "\n",
    "samples"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### read some coolers and extract raw number of interactions between allele1/2 and other chromosomes ..."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "# READ SOME COOLERS ...\n",
    "# oldref-20200617\n",
    "cool_path = \"/home/venevs/DOTS_TESTING/alv-chicken/oldref-20200617\"\n",
    "get_cpath = lambda c,b: os.path.join(cool_path,c+f\"::/resolutions/{b}\")\n",
    "\n",
    "insertion_site = 72_500_000\n",
    "\n",
    "dfs = {}\n",
    "for s in samples[\"sample\"]:\n",
    "    _bsize = 1_000_000\n",
    "    s_cname, = samples[\"mcool\"][samples[\"sample\"]==s]\n",
    "    _c = cooler.Cooler(get_cpath(s_cname,_bsize))\n",
    "    _df = pd.DataFrame(_c.chromsizes).reset_index().rename({\"name\":\"chrom\"},axis=1)\n",
    "    a1_idx = _c.offset(\"allele1\")\n",
    "    a2_idx = _c.offset(\"allele2\")\n",
    "    chr1_idx = slice(*_c.extent(\"chr1\"))\n",
    "    # chr1_idx = _c.offset((\"chr1\",insertion_site,insertion_site+_bsize))\n",
    "    _df[\"allele1\"] = [np.sum(_c.matrix(balance=False)[a1_idx,slice(*_c.extent(chrom))]) for chrom in _c.chromnames]\n",
    "    _df[\"allele2\"] = [np.sum(_c.matrix(balance=False)[a2_idx,slice(*_c.extent(chrom))]) for chrom in _c.chromnames]\n",
    "    _df[\"chr1\"] = [np.sum(_c.matrix(balance=False)[chr1_idx,slice(*_c.extent(chrom))]) for chrom in _c.chromnames]\n",
    "    dfs[s] = _df"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### test GC content of allele1 and allele2, and \"density\" of DpnII cut-sites"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "chr1 40.01674361571989 196202544\n",
      "chr1 cutsites 0.002134\n",
      "allele1 56.674565560821485 2532\n",
      "allele1 cutsites 0.004344391785150079\n",
      "allele2 46.15384615384615 2509\n",
      "allele2 cutsites 0.0031885213232363493\n"
     ]
    }
   ],
   "source": [
    "gc_dict = {}\n",
    "from Bio import SeqIO, SeqUtils\n",
    "for seq in SeqIO.parse(\"galGal5.fa\",\"fasta\"):\n",
    "    if seq.id in [\"allele1\",\"allele2\"]:\n",
    "        print(seq.id, SeqUtils.GC(seq.seq), len(seq))\n",
    "        cuts = len(SeqUtils.nt_search(str(seq.seq),\"GATC\"))-1\n",
    "        print(seq.id, \"cutsites\", cuts/len(seq) )\n",
    "        gc_dict[seq.id] = {\"GC\":SeqUtils.GC(seq.seq),\"length\":len(seq),\"cut\":cuts/len(seq)}\n",
    "    elif seq.id == \"chr1\":\n",
    "        # print(seq.id, SeqUtils.GC(seq.seq[insertion_site-1_000_000:insertion_site+1_000_000]), len(seq))\n",
    "        # only an estimate of cut sites in chr1 - only in the first 1Mb\n",
    "        # otherwise it takes forever to calculate ...\n",
    "        cuts = len(SeqUtils.nt_search(str(seq.seq)[:1_000_000],\"GATC\"))-1\n",
    "        print(seq.id, SeqUtils.GC(seq.seq), len(seq))\n",
    "        print(seq.id, \"cutsites\", cuts/1_000_000 )\n",
    "        gc_dict[seq.id] = {\"GC\":SeqUtils.GC(seq.seq),\"length\":len(seq),\"cut\":cuts/1_000_000}\n",
    "    else:\n",
    "        pass\n",
    "# allele1 is a bit more GC-rich"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "pool-2percent-highcis {'bin-size': 50000, 'bin-type': 'fixed', 'creation-date': '2020-06-11T04:43:47.331890', 'format': 'HDF5::Cooler', 'format-url': 'https://github.com/mirnylab/cooler', 'format-version': 3, 'generated-by': 'cooler-0.8.1', 'genome-assembly': 'unknown', 'metadata': {}, 'nbins': 20433, 'nchroms': 37, 'nnz': 111179105, 'storage-mode': 'symmetric-upper', 'sum': 897082537}\n"
     ]
    }
   ],
   "source": [
    "s = samples[\"sample\"][0]\n",
    "_bsize = 50_000\n",
    "s_cname, = samples[\"mcool\"][samples[\"sample\"]==s]\n",
    "clr = cooler.Cooler(get_cpath(s_cname,_bsize))\n",
    "print(s,clr.info)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "binss = []\n",
    "for seqrec in SeqIO.parse(\"galGal5.fa\",\"fasta\"):\n",
    "    chrom = seqrec.id\n",
    "    if chrom in clr.chromnames[:10]:\n",
    "        bins = clr.bins().fetch(chrom)\n",
    "        bins[\"GC\"] = [ SeqUtils.GC(seqrec.seq[r.start:r.end])  for r in bins.itertuples()]\n",
    "        bins[\"cut\"] = [ len(SeqUtils.nt_search(str(seqrec.seq[r.start:r.end]),\"GATC\"))/(r.end-r.start) for r in bins.itertuples()]\n",
    "        binss.append(bins) \n",
    "binss = pd.concat(binss)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.PairGrid at 0x7f6bf8d67e10>"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 540x540 with 12 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# sns.jointplot(data=binss,y=\"weight\",x=\"GC\",kind=\"hex\", color=\"k\");\n",
    "g = sns.PairGrid(binss[[\"GC\",\"weight\",\"cut\"]], diag_sharey=False)\n",
    "g.map_upper(sns.scatterplot,**{\"alpha\":0.1})\n",
    "g.map_lower(plt.hexbin ,gridsize=1000)#, **{\"kind\":\"hex\"})\n",
    "g.map_diag(plt.hist, bins=50)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### let's estimate fraction of `cis` interactions for allele1/2\n",
    "\n",
    "this is a rough estimate , calculated by summing raw allele1/2 interactions with chr1 and dividing by the total number of interactions detected with allele1/2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "# construct a little table to characterize each sample allele1/allele2 interactions\n",
    "df_dict = {}\n",
    "for s in samples[\"sample\"]:\n",
    "    if \"WT\" not in s:\n",
    "        df_dict[s] = {}\n",
    "        df_dict[s][\"allele1\"] = {}\n",
    "        df_dict[s][\"allele2\"] = {}\n",
    "        df_dict[s][\"chr1\"] = {}\n",
    "\n",
    "df_combo = []\n",
    "for s in samples[\"sample\"]:\n",
    "    if \"WT\" not in s:\n",
    "        dd = dfs[s]\n",
    "        for ccc in [\"allele1\",\"allele2\",\"chr1\"]:\n",
    "            df_dict[s][ccc][\"total interactions\"] = str(dd[ccc].sum())\n",
    "            df_dict[s][ccc][\"interactions w. chr1\"] = str(dd[dd[\"chrom\"]==\"chr1\"][ccc].sum())\n",
    "            df_dict[s][ccc][\"cis_percent\"] = str(f'{100*dd[dd[\"chrom\"].isin([ccc,\"chr1\"])][ccc].sum()/dd[ccc].sum():.2f}')\n",
    "            df_dict[s][ccc][\"GC\"] = gc_dict[ccc][\"GC\"]\n",
    "            df_dict[s][ccc][\"DpnII sites/bp\"] = gc_dict[ccc][\"cut\"]\n",
    "            df_dict[s][ccc][\"length\"] = gc_dict[ccc][\"length\"]\n",
    "            #\n",
    "            _df = pd.DataFrame(df_dict[s]).T\n",
    "            _df[\"sample\"] = s\n",
    "            df_combo.append(_df)\n",
    "            \n",
    "        \n",
    "# allele1 seem to have more inter-chromosomal interactions\n",
    "# allele2 have average level of interactions with other chromosomes - roughly as a whole chr1\n",
    "\n",
    "dtypes = {\"total interactions\":int,\n",
    "        \"interactions w. chr1\":int,\n",
    "        \"cis_percent\":float,\n",
    "        \"GC\":float,\n",
    "        \"length\":int,\n",
    "        }\n",
    "# nice summary table about the samples/alleles :\n",
    "final_df = pd.concat(df_combo) \\\n",
    "    .dropna() \\\n",
    "    .drop_duplicates() \\\n",
    "    .reset_index() \\\n",
    "    .rename({\"index\":\"name\"},axis=1) \\\n",
    "    .sort_values(\"name\") \\\n",
    "    .reset_index(drop=True) \\\n",
    "    .astype(dtypes)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "clone-16-2p\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>name</th>\n",
       "      <th>length</th>\n",
       "      <th>GC</th>\n",
       "      <th>DpnII sites/bp</th>\n",
       "      <th>total interactions</th>\n",
       "      <th>cis_percent</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>allele1</td>\n",
       "      <td>2532</td>\n",
       "      <td>56.674566</td>\n",
       "      <td>0.00434439</td>\n",
       "      <td>2433</td>\n",
       "      <td>67.41</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>allele2</td>\n",
       "      <td>2509</td>\n",
       "      <td>46.153846</td>\n",
       "      <td>0.00318852</td>\n",
       "      <td>1318</td>\n",
       "      <td>80.20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>chr1</td>\n",
       "      <td>196202544</td>\n",
       "      <td>40.016744</td>\n",
       "      <td>0.002134</td>\n",
       "      <td>121564119</td>\n",
       "      <td>81.08</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      name     length         GC DpnII sites/bp  total interactions  \\\n",
       "0  allele1       2532  56.674566     0.00434439                2433   \n",
       "1  allele2       2509  46.153846     0.00318852                1318   \n",
       "2     chr1  196202544  40.016744       0.002134           121564119   \n",
       "\n",
       "   cis_percent  \n",
       "0        67.41  \n",
       "1        80.20  \n",
       "2        81.08  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "clone-17-2p\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>name</th>\n",
       "      <th>length</th>\n",
       "      <th>GC</th>\n",
       "      <th>DpnII sites/bp</th>\n",
       "      <th>total interactions</th>\n",
       "      <th>cis_percent</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>allele1</td>\n",
       "      <td>2532</td>\n",
       "      <td>56.674566</td>\n",
       "      <td>0.00434439</td>\n",
       "      <td>2369</td>\n",
       "      <td>69.78</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>allele2</td>\n",
       "      <td>2509</td>\n",
       "      <td>46.153846</td>\n",
       "      <td>0.00318852</td>\n",
       "      <td>1290</td>\n",
       "      <td>78.53</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>chr1</td>\n",
       "      <td>196202544</td>\n",
       "      <td>40.016744</td>\n",
       "      <td>0.002134</td>\n",
       "      <td>116517761</td>\n",
       "      <td>80.63</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      name     length         GC DpnII sites/bp  total interactions  \\\n",
       "0  allele1       2532  56.674566     0.00434439                2369   \n",
       "1  allele2       2509  46.153846     0.00318852                1290   \n",
       "2     chr1  196202544  40.016744       0.002134           116517761   \n",
       "\n",
       "   cis_percent  \n",
       "0        69.78  \n",
       "1        78.53  \n",
       "2        80.63  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "cols = [\"name\",\n",
    "\"length\",\n",
    "\"GC\",\n",
    "\"DpnII sites/bp\",\n",
    "\"total interactions\",\n",
    "# \"interactions w. chr1\",\n",
    "\"cis_percent\"]\n",
    "\n",
    "\n",
    "print(\"clone-16-2p\")\n",
    "display( final_df[final_df[\"sample\"] == \"clone-16-2p\"].reset_index(drop=True)[cols] )\n",
    "print(\"clone-17-2p\")\n",
    "display( final_df[final_df[\"sample\"] == \"clone-17-2p\"].reset_index(drop=True)[cols] )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "clone-16-2p\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>name</th>\n",
       "      <th>length</th>\n",
       "      <th>GC</th>\n",
       "      <th>DpnII sites/bp</th>\n",
       "      <th>sample</th>\n",
       "      <th>total interactions</th>\n",
       "      <th>cis_percent</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>allele1</td>\n",
       "      <td>2532</td>\n",
       "      <td>56.674566</td>\n",
       "      <td>0.00434439</td>\n",
       "      <td>clone-16-2p</td>\n",
       "      <td>2433</td>\n",
       "      <td>67.41</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>allele2</td>\n",
       "      <td>2509</td>\n",
       "      <td>46.153846</td>\n",
       "      <td>0.00318852</td>\n",
       "      <td>clone-16-2p</td>\n",
       "      <td>1318</td>\n",
       "      <td>80.20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>chr1</td>\n",
       "      <td>196202544</td>\n",
       "      <td>40.016744</td>\n",
       "      <td>0.002134</td>\n",
       "      <td>clone-16-2p</td>\n",
       "      <td>121564119</td>\n",
       "      <td>81.08</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      name     length         GC DpnII sites/bp       sample  \\\n",
       "0  allele1       2532  56.674566     0.00434439  clone-16-2p   \n",
       "1  allele2       2509  46.153846     0.00318852  clone-16-2p   \n",
       "2     chr1  196202544  40.016744       0.002134  clone-16-2p   \n",
       "\n",
       "   total interactions  cis_percent  \n",
       "0                2433        67.41  \n",
       "1                1318        80.20  \n",
       "2           121564119        81.08  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "clone-17-2p\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>name</th>\n",
       "      <th>length</th>\n",
       "      <th>GC</th>\n",
       "      <th>DpnII sites/bp</th>\n",
       "      <th>sample</th>\n",
       "      <th>total interactions</th>\n",
       "      <th>cis_percent</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>allele1</td>\n",
       "      <td>2532</td>\n",
       "      <td>56.674566</td>\n",
       "      <td>0.00434439</td>\n",
       "      <td>clone-17-2p</td>\n",
       "      <td>2369</td>\n",
       "      <td>69.78</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>allele2</td>\n",
       "      <td>2509</td>\n",
       "      <td>46.153846</td>\n",
       "      <td>0.00318852</td>\n",
       "      <td>clone-17-2p</td>\n",
       "      <td>1290</td>\n",
       "      <td>78.53</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>chr1</td>\n",
       "      <td>196202544</td>\n",
       "      <td>40.016744</td>\n",
       "      <td>0.002134</td>\n",
       "      <td>clone-17-2p</td>\n",
       "      <td>116517761</td>\n",
       "      <td>80.63</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      name     length         GC DpnII sites/bp       sample  \\\n",
       "0  allele1       2532  56.674566     0.00434439  clone-17-2p   \n",
       "1  allele2       2509  46.153846     0.00318852  clone-17-2p   \n",
       "2     chr1  196202544  40.016744       0.002134  clone-17-2p   \n",
       "\n",
       "   total interactions  cis_percent  \n",
       "0                2369        69.78  \n",
       "1                1290        78.53  \n",
       "2           116517761        80.63  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "cols = [\"name\",\n",
    "\"length\",\n",
    "\"GC\",\n",
    "\"DpnII sites/bp\", \"sample\",\n",
    "\"total interactions\",\n",
    "# \"interactions w. chr1\",\n",
    "\"cis_percent\"]\n",
    "\n",
    "\n",
    "print(\"clone-16-2p\")\n",
    "display( final_df[final_df[\"sample\"] == \"clone-16-2p\"].reset_index(drop=True)[cols] )\n",
    "print(\"clone-17-2p\")\n",
    "display( final_df[final_df[\"sample\"] == \"clone-17-2p\"].reset_index(drop=True)[cols] )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "final_df[final_df[\"sample\"] != \"pool-2percent-highcis\"][cols].to_csv(\"xxx.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.7.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}