{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "nbsphinx": "hidden"
   },
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 1;\n",
       "                var nbb_unformatted_code = \"%load_ext nb_black\";\n",
       "                var nbb_formatted_code = \"%load_ext nb_black\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%load_ext nb_black"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 1000 Genomes analysis\n",
    "The tree sequence files for 1kg are available [here](https://zenodo.org/record/3051855). The other metadata files are part of the public 1kg release."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 2;\n",
       "                var nbb_unformatted_code = \"import os\\nimport tskit\\nimport numpy as np\\nimport xsmc\\nimport matplotlib.pyplot as plt\\nfrom concurrent.futures import ThreadPoolExecutor, as_completed, ProcessPoolExecutor\\nfrom xsmc.supporting.kde_ne import kde_ne\\nfrom xsmc.supporting.plotting import *\\nimport logging\\nimport os\\n\\n\\nlogging.getLogger(\\\"xsmc\\\").setLevel(logging.INFO)\\nPAPER_ROOT = os.path.expanduser(os.environ.get(\\\"PAPER_ROOT\\\", \\\".\\\"))\";\n",
       "                var nbb_formatted_code = \"import os\\nimport tskit\\nimport numpy as np\\nimport xsmc\\nimport matplotlib.pyplot as plt\\nfrom concurrent.futures import ThreadPoolExecutor, as_completed, ProcessPoolExecutor\\nfrom xsmc.supporting.kde_ne import kde_ne\\nfrom xsmc.supporting.plotting import *\\nimport logging\\nimport os\\n\\n\\nlogging.getLogger(\\\"xsmc\\\").setLevel(logging.INFO)\\nPAPER_ROOT = os.path.expanduser(os.environ.get(\\\"PAPER_ROOT\\\", \\\".\\\"))\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import os\n",
    "import tskit\n",
    "import numpy as np\n",
    "import xsmc\n",
    "import matplotlib.pyplot as plt\n",
    "from concurrent.futures import ThreadPoolExecutor, as_completed, ProcessPoolExecutor\n",
    "from xsmc.supporting.kde_ne import kde_ne\n",
    "from xsmc.supporting.plotting import *\n",
    "import logging\n",
    "import os\n",
    "\n",
    "\n",
    "logging.getLogger(\"xsmc\").setLevel(logging.INFO)\n",
    "PAPER_ROOT = os.path.expanduser(os.environ.get(\"PAPER_ROOT\", \".\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 3;\n",
       "                var nbb_unformatted_code = \"np.random.seed(1)\\n\\n\\ndef seed():\\n    return np.random.randint(1, np.iinfo(np.int32).max)\";\n",
       "                var nbb_formatted_code = \"np.random.seed(1)\\n\\n\\ndef seed():\\n    return np.random.randint(1, np.iinfo(np.int32).max)\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "np.random.seed(1)\n",
    "\n",
    "\n",
    "def seed():\n",
    "    return np.random.randint(1, np.iinfo(np.int32).max)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 4;\n",
       "                var nbb_unformatted_code = \"full_chroms = [tskit.load(f\\\"/scratch/1kg/trees/1kg_chr{i}.trees\\\") for i in range(1, 23)]\\n# chr1 = chr1.delete_intervals([[0, 16e6]], simplify=False).trim()\";\n",
       "                var nbb_formatted_code = \"full_chroms = [tskit.load(f\\\"/scratch/1kg/trees/1kg_chr{i}.trees\\\") for i in range(1, 23)]\\n# chr1 = chr1.delete_intervals([[0, 16e6]], simplify=False).trim()\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "full_chroms = [tskit.load(f\"/scratch/1kg/trees/1kg_chr{i}.trees\") for i in range(1, 23)]\n",
    "# chr1 = chr1.delete_intervals([[0, 16e6]], simplify=False).trim()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Accessibility masking\n",
    "The 1kg tree sequences feature some trees that span huge intervals. These represent centromeres, telomeres, or other inaccessible regions. To our method they appear as huge tracts of IBD, leading to downward bias in the size history estimates. To properly correct for this we should use the 1kg accessibility masks, but here I adopt the simpler approach of heuristically chopping out trees that span large segments (>1Mb)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{1: [Interval(left=121347335.0, right=142643153.0)],\n",
       " 2: [Interval(left=90490322.0, right=91634056.0),\n",
       "  Interval(left=92102205.0, right=95333904.0)],\n",
       " 3: [Interval(left=90291128.0, right=93527692.0)],\n",
       " 4: [Interval(left=49629442.0, right=52700457.0)],\n",
       " 5: [Interval(left=46118241.0, right=49571959.0)],\n",
       " 6: [Interval(left=58665913.0, right=61967504.0)],\n",
       " 7: [Interval(left=57669630.0, right=62465883.0)],\n",
       " 8: [Interval(left=43529404.0, right=47458633.0)],\n",
       " 9: [Interval(left=47309856.0, right=65510886.0)],\n",
       " 10: [Interval(left=39043557.0, right=42746452.0)],\n",
       " 11: [Interval(left=50381334.0, right=51436554.0),\n",
       "  Interval(left=51510699.0, right=55254669.0)],\n",
       " 12: [Interval(left=34340941.0, right=38610698.0)],\n",
       " 13: [Interval(left=0.0, right=19037838.0)],\n",
       " 14: [Interval(left=0.0, right=19083007.0)],\n",
       " 15: [Interval(left=0.0, right=20253797.0)],\n",
       " 16: [Interval(left=35146682.0, right=46554124.0)],\n",
       " 17: [Interval(left=22155426.0, right=25392232.0)],\n",
       " 18: [Interval(left=15357425.0, right=18535477.0)],\n",
       " 19: [Interval(left=24310556.0, right=28366182.0)],\n",
       " 20: [Interval(left=26193540.0, right=29862565.0)],\n",
       " 21: [Interval(left=0.0, right=9491942.0),\n",
       "  Interval(left=11010799.0, right=14383028.0)],\n",
       " 22: [Interval(left=0.0, right=16103536.0)]}"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 5;\n",
       "                var nbb_unformatted_code = \"long_spans = {}\\nfor i, chrom in enumerate(full_chroms, 1):\\n    long_spans[i] = [t.interval for t in chrom.trees() if np.diff(t.interval) > 1e6]\\nlong_spans\";\n",
       "                var nbb_formatted_code = \"long_spans = {}\\nfor i, chrom in enumerate(full_chroms, 1):\\n    long_spans[i] = [t.interval for t in chrom.trees() if np.diff(t.interval) > 1e6]\\nlong_spans\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "long_spans = {}\n",
    "for i, chrom in enumerate(full_chroms, 1):\n",
    "    long_spans[i] = [t.interval for t in chrom.trees() if np.diff(t.interval) > 1e6]\n",
    "long_spans"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now we simply chop these intervals out. This technically allows IBD tracts to span these gaps during inference, but the effect should be minimal."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 6;\n",
       "                var nbb_unformatted_code = \"chroms = [\\n    chrom.delete_intervals(long_spans[i], simplify=False).trim()\\n    for i, chrom in enumerate(full_chroms, 1)\\n]\";\n",
       "                var nbb_formatted_code = \"chroms = [\\n    chrom.delete_intervals(long_spans[i], simplify=False).trim()\\n    for i, chrom in enumerate(full_chroms, 1)\\n]\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "chroms = [\n",
    "    chrom.delete_intervals(long_spans[i], simplify=False).trim()\n",
    "    for i, chrom in enumerate(full_chroms, 1)\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 7;\n",
       "                var nbb_unformatted_code = \"# 1kg metadata\\n\\nwith open(\\\"/scratch/1kg/integrated_call_samples_v3.20130502.ALL.panel\\\", \\\"rt\\\") as f:\\n    next(f)\\n    rows = (line.strip().split(\\\"\\\\t\\\") for line in f)\\n    sample_map = {sample_id: (pop, superpop) for sample_id, pop, superpop, _ in rows}\";\n",
       "                var nbb_formatted_code = \"# 1kg metadata\\n\\nwith open(\\\"/scratch/1kg/integrated_call_samples_v3.20130502.ALL.panel\\\", \\\"rt\\\") as f:\\n    next(f)\\n    rows = (line.strip().split(\\\"\\\\t\\\") for line in f)\\n    sample_map = {sample_id: (pop, superpop) for sample_id, pop, superpop, _ in rows}\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# 1kg metadata\n",
    "\n",
    "with open(\"/scratch/1kg/integrated_call_samples_v3.20130502.ALL.panel\", \"rt\") as f:\n",
    "    next(f)\n",
    "    rows = (line.strip().split(\"\\t\") for line in f)\n",
    "    sample_map = {sample_id: (pop, superpop) for sample_id, pop, superpop, _ in rows}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 8;\n",
       "                var nbb_unformatted_code = \"# map each 1kg sample id ts nodes\\nimport json\\n\\nsamples_to_nodes = {\\n    json.loads(ind.metadata)[\\\"individual_id\\\"]: ind.nodes\\n    for ind in chroms[0].individuals()\\n}\";\n",
       "                var nbb_formatted_code = \"# map each 1kg sample id ts nodes\\nimport json\\n\\nsamples_to_nodes = {\\n    json.loads(ind.metadata)[\\\"individual_id\\\"]: ind.nodes\\n    for ind in chroms[0].individuals()\\n}\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# map each 1kg sample id ts nodes\n",
    "import json\n",
    "\n",
    "samples_to_nodes = {\n",
    "    json.loads(ind.metadata)[\"individual_id\"]: ind.nodes\n",
    "    for ind in chroms[0].individuals()\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 9;\n",
       "                var nbb_unformatted_code = \"superpops = {}\\nfor sample_id, (p, sp) in sample_map.items():\\n    superpops.setdefault(sp, [])\\n    superpops[sp].append(sample_id)\";\n",
       "                var nbb_formatted_code = \"superpops = {}\\nfor sample_id, (p, sp) in sample_map.items():\\n    superpops.setdefault(sp, [])\\n    superpops[sp].append(sample_id)\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "superpops = {}\n",
    "for sample_id, (p, sp) in sample_map.items():\n",
    "    superpops.setdefault(sp, [])\n",
    "    superpops[sp].append(sample_id)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 10;\n",
       "                var nbb_unformatted_code = \"K = 20  # number of samples to take from each superpopulation\\nmu = 1.4e-8  # assumed mutation rate for humans\\n\\n\\ndef process_samples(sample_dict, w, rho_over_theta):\\n    sampled_heights = {}\\n    lines = {}\\n    for sp in sample_dict:\\n        print(sp, flush=True)\\n        xs = []\\n        for sample_id in sample_dict[sp][:K]:\\n            print(\\\"\\\\t%s\\\" % sample_id)\\n            f, p = samples_to_nodes[sample_id]\\n            for data in chroms:\\n                xs.append(\\n                    xsmc.XSMC(\\n                        data, focal=f, panel=[p], w=w, rho_over_theta=rho_over_theta\\n                    )\\n                )\\n        with ThreadPoolExecutor(24) as p:\\n            futs = [\\n                p.submit(\\n                    x.sample_heights,\\n                    j=100,\\n                    k=int(x.ts.get_sequence_length() / 50_000),\\n                    seed=seed(),\\n                )\\n                for x in xs\\n            ]\\n\\n        sampled_heights[sp] = np.concatenate(\\n            [f.result() * 2 * x.theta / (4 * mu) for f, x in zip(futs, xs)], axis=1\\n        )  # rescale each sampled path by 2N0 so that segment heights are in generations\\n    return sampled_heights\";\n",
       "                var nbb_formatted_code = \"K = 20  # number of samples to take from each superpopulation\\nmu = 1.4e-8  # assumed mutation rate for humans\\n\\n\\ndef process_samples(sample_dict, w, rho_over_theta):\\n    sampled_heights = {}\\n    lines = {}\\n    for sp in sample_dict:\\n        print(sp, flush=True)\\n        xs = []\\n        for sample_id in sample_dict[sp][:K]:\\n            print(\\\"\\\\t%s\\\" % sample_id)\\n            f, p = samples_to_nodes[sample_id]\\n            for data in chroms:\\n                xs.append(\\n                    xsmc.XSMC(\\n                        data, focal=f, panel=[p], w=w, rho_over_theta=rho_over_theta\\n                    )\\n                )\\n        with ThreadPoolExecutor(24) as p:\\n            futs = [\\n                p.submit(\\n                    x.sample_heights,\\n                    j=100,\\n                    k=int(x.ts.get_sequence_length() / 50_000),\\n                    seed=seed(),\\n                )\\n                for x in xs\\n            ]\\n\\n        sampled_heights[sp] = np.concatenate(\\n            [f.result() * 2 * x.theta / (4 * mu) for f, x in zip(futs, xs)], axis=1\\n        )  # rescale each sampled path by 2N0 so that segment heights are in generations\\n    return sampled_heights\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "K = 20  # number of samples to take from each superpopulation\n",
    "mu = 1.4e-8  # assumed mutation rate for humans\n",
    "\n",
    "\n",
    "def process_samples(sample_dict, w, rho_over_theta):\n",
    "    sampled_heights = {}\n",
    "    lines = {}\n",
    "    for sp in sample_dict:\n",
    "        print(sp, flush=True)\n",
    "        xs = []\n",
    "        for sample_id in sample_dict[sp][:K]:\n",
    "            print(\"\\t%s\" % sample_id)\n",
    "            f, p = samples_to_nodes[sample_id]\n",
    "            for data in chroms:\n",
    "                xs.append(\n",
    "                    xsmc.XSMC(\n",
    "                        data, focal=f, panel=[p], w=w, rho_over_theta=rho_over_theta\n",
    "                    )\n",
    "                )\n",
    "        with ThreadPoolExecutor(24) as p:\n",
    "            futs = [\n",
    "                p.submit(\n",
    "                    x.sample_heights,\n",
    "                    j=100,\n",
    "                    k=int(x.ts.get_sequence_length() / 50_000),\n",
    "                    seed=seed(),\n",
    "                )\n",
    "                for x in xs\n",
    "            ]\n",
    "\n",
    "        sampled_heights[sp] = np.concatenate(\n",
    "            [f.result() * 2 * x.theta / (4 * mu) for f, x in zip(futs, xs)], axis=1\n",
    "        )  # rescale each sampled path by 2N0 so that segment heights are in generations\n",
    "    return sampled_heights"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Time to process 1 whole genome"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "test\n",
      "\tNA12878\n",
      "CPU times: user 7min 36s, sys: 743 ms, total: 7min 37s\n",
      "Wall time: 37.8 s\n"
     ]
    },
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 11;\n",
       "                var nbb_unformatted_code = \"%%time\\n_ = process_samples({'test': ['NA12878']}, w=500, rho_over_theta=1.)\";\n",
       "                var nbb_formatted_code = \"%%time\\n_ = process_samples({'test': ['NA12878']}, w=500, rho_over_theta=1.)\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%%time\n",
    "_ = process_samples({'test': ['NA12878']}, w=500, rho_over_theta=1.)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Main event\n",
    "Run pipeline for all 1kg samples split into 5 superpopulations"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 12;\n",
       "                var nbb_unformatted_code = \"def make_plot(sampled_heights, g, name):\\n    fig, ax = plt.subplots(figsize=(8, 5))\\n    x0 = np.geomspace(1e2, 1e6, 1000)\\n    for sp, col in zip(sampled_heights, TABLEAU):\\n        lines = []\\n        A = np.array(sampled_heights[sp])\\n        with ProcessPoolExecutor() as p:\\n            futs = [p.submit(kde_ne, a.reshape(-1)) for a in A]\\n            for f in as_completed(futs):\\n                x, y = f.result()\\n                lines.append((x * g, y / 2))  # rescale to years, plot diploid Ne\\n        plot_summary(ax, lines, x0, color=col, label=sp)\\n    ax.set_xscale(\\\"log\\\")\\n    ax.set_yscale(\\\"log\\\")\\n    ax.legend(loc=\\\"lower right\\\")\\n    ax.set_xlabel(r\\\"Years ($\\\\mu=1.4\\\\times 10^{-8}, g=%d$)\\\" % g)\\n    ax.set_ylabel(\\\"Effective Population Size ($N_e$)\\\")\\n    fig.tight_layout()\\n    fig.savefig(os.path.join(PAPER_ROOT, \\\"figures\\\", \\\"xsmc_1kg_%s.pdf\\\" % name))\\n    return fig\";\n",
       "                var nbb_formatted_code = \"def make_plot(sampled_heights, g, name):\\n    fig, ax = plt.subplots(figsize=(8, 5))\\n    x0 = np.geomspace(1e2, 1e6, 1000)\\n    for sp, col in zip(sampled_heights, TABLEAU):\\n        lines = []\\n        A = np.array(sampled_heights[sp])\\n        with ProcessPoolExecutor() as p:\\n            futs = [p.submit(kde_ne, a.reshape(-1)) for a in A]\\n            for f in as_completed(futs):\\n                x, y = f.result()\\n                lines.append((x * g, y / 2))  # rescale to years, plot diploid Ne\\n        plot_summary(ax, lines, x0, color=col, label=sp)\\n    ax.set_xscale(\\\"log\\\")\\n    ax.set_yscale(\\\"log\\\")\\n    ax.legend(loc=\\\"lower right\\\")\\n    ax.set_xlabel(r\\\"Years ($\\\\mu=1.4\\\\times 10^{-8}, g=%d$)\\\" % g)\\n    ax.set_ylabel(\\\"Effective Population Size ($N_e$)\\\")\\n    fig.tight_layout()\\n    fig.savefig(os.path.join(PAPER_ROOT, \\\"figures\\\", \\\"xsmc_1kg_%s.pdf\\\" % name))\\n    return fig\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def make_plot(sampled_heights, g, name):\n",
    "    fig, ax = plt.subplots(figsize=(8, 5))\n",
    "    x0 = np.geomspace(1e2, 1e6, 1000)\n",
    "    for sp, col in zip(sampled_heights, TABLEAU):\n",
    "        lines = []\n",
    "        A = np.array(sampled_heights[sp])\n",
    "        with ProcessPoolExecutor() as p:\n",
    "            futs = [p.submit(kde_ne, a.reshape(-1)) for a in A]\n",
    "            for f in as_completed(futs):\n",
    "                x, y = f.result()\n",
    "                lines.append((x * g, y / 2))  # rescale to years, plot diploid Ne\n",
    "        plot_summary(ax, lines, x0, color=col, label=sp)\n",
    "    ax.set_xscale(\"log\")\n",
    "    ax.set_yscale(\"log\")\n",
    "    ax.legend(loc=\"lower right\")\n",
    "    ax.set_xlabel(r\"Years ($\\mu=1.4\\times 10^{-8}, g=%d$)\" % g)\n",
    "    ax.set_ylabel(\"Effective Population Size ($N_e$)\")\n",
    "    fig.tight_layout()\n",
    "    fig.savefig(os.path.join(PAPER_ROOT, \"figures\", \"xsmc_1kg_%s.pdf\" % name))\n",
    "    return fig"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "EUR\n",
      "\tHG00096\n",
      "\tHG00097\n",
      "\tHG00099\n",
      "\tHG00100\n",
      "\tHG00101\n",
      "\tHG00102\n",
      "\tHG00103\n",
      "\tHG00105\n",
      "\tHG00106\n",
      "\tHG00107\n",
      "\tHG00108\n",
      "\tHG00109\n",
      "\tHG00110\n",
      "\tHG00111\n",
      "\tHG00112\n",
      "\tHG00113\n",
      "\tHG00114\n",
      "\tHG00115\n",
      "\tHG00116\n",
      "\tHG00117\n",
      "EAS\n",
      "\tHG00403\n",
      "\tHG00404\n",
      "\tHG00406\n",
      "\tHG00407\n",
      "\tHG00409\n",
      "\tHG00410\n",
      "\tHG00419\n",
      "\tHG00421\n",
      "\tHG00422\n",
      "\tHG00428\n",
      "\tHG00436\n",
      "\tHG00437\n",
      "\tHG00442\n",
      "\tHG00443\n",
      "\tHG00445\n",
      "\tHG00446\n",
      "\tHG00448\n",
      "\tHG00449\n",
      "\tHG00451\n",
      "\tHG00452\n",
      "AMR\n",
      "\tHG00551\n",
      "\tHG00553\n",
      "\tHG00554\n",
      "\tHG00637\n",
      "\tHG00638\n",
      "\tHG00640\n",
      "\tHG00641\n",
      "\tHG00731\n",
      "\tHG00732\n",
      "\tHG00734\n",
      "\tHG00736\n",
      "\tHG00737\n",
      "\tHG00739\n",
      "\tHG00740\n",
      "\tHG00742\n",
      "\tHG00743\n",
      "\tHG01047\n",
      "\tHG01048\n",
      "\tHG01049\n",
      "\tHG01051\n",
      "SAS\n",
      "\tHG01583\n",
      "\tHG01586\n",
      "\tHG01589\n",
      "\tHG01593\n",
      "\tHG02490\n",
      "\tHG02491\n",
      "\tHG02493\n",
      "\tHG02494\n",
      "\tHG02597\n",
      "\tHG02600\n",
      "\tHG02601\n",
      "\tHG02603\n",
      "\tHG02604\n",
      "\tHG02648\n",
      "\tHG02649\n",
      "\tHG02651\n",
      "\tHG02652\n",
      "\tHG02654\n",
      "\tHG02655\n",
      "\tHG02657\n",
      "AFR\n",
      "\tHG01879\n",
      "\tHG01880\n",
      "\tHG01882\n",
      "\tHG01883\n",
      "\tHG01885\n",
      "\tHG01886\n",
      "\tHG01889\n",
      "\tHG01890\n",
      "\tHG01894\n",
      "\tHG01896\n",
      "\tHG01912\n",
      "\tHG01914\n",
      "\tHG01915\n",
      "\tHG01956\n",
      "\tHG01958\n",
      "\tHG01985\n",
      "\tHG01986\n",
      "\tHG01988\n",
      "\tHG01989\n",
      "\tHG01990\n",
      "CPU times: user 15h 50min 50s, sys: 32.1 s, total: 15h 51min 22s\n",
      "Wall time: 41min 10s\n"
     ]
    },
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 13;\n",
       "                var nbb_unformatted_code = \"%%time\\nsampled_heights = process_samples(superpops, w=500, rho_over_theta=1.)\";\n",
       "                var nbb_formatted_code = \"%%time\\nsampled_heights = process_samples(superpops, w=500, rho_over_theta=1.)\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%%time\n",
    "sampled_heights = process_samples(superpops, w=500, rho_over_theta=1.)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/terhorst/opt/py37/lib/python3.7/site-packages/numpy/lib/nanfunctions.py:1392: RuntimeWarning: All-NaN slice encountered\n",
      "  overwrite_input, interpolation)\n",
      "2020-09-21 16:03:00,023 WARNING matplotlib.font_manager MainThread findfont: Font family ['sans-serif'] not found. Falling back to DejaVu Sans.\n",
      "2020-09-21 16:03:00,225 WARNING matplotlib.font_manager MainThread findfont: Font family ['sans-serif'] not found. Falling back to DejaVu Sans.\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x360 with 1 Axes>"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 14;\n",
       "                var nbb_unformatted_code = \"make_plot(sampled_heights, 29, \\\"final\\\")\";\n",
       "                var nbb_formatted_code = \"make_plot(sampled_heights, 29, \\\"final\\\")\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "make_plot(sampled_heights, 29, \"final\")"
   ]
  }
 ],
 "metadata": {
  "celltoolbar": "Edit 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": "3.7.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}