jscatter demo.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "ff3b9947",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import pyarrow.parquet as pq\n",
    "import bioframe as bf\n",
    "\n",
    "from pathlib import Path\n",
    "\n",
    "data = Path('data')\n",
    "\n",
    "df = pq.read_table(data / 'HCT116_Unsynchronized.hg38.250000.E0-E128.trans.eigvecs.pq').to_pandas()\n",
    "df.dropna(axis=0, subset=['E0', 'E1'], inplace=True)\n",
    "df.chrom = df.chrom.astype(\"category\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "99faa0c6",
   "metadata": {},
   "outputs": [],
   "source": [
    "import slugid\n",
    "from dataclasses import dataclass, field\n",
    "import numpy as np\n",
    "from rich import print\n",
    "import bioframe as bf\n",
    "import pandas as pd\n",
    "\n",
    "\n",
    "def get_lens(chromsizes: pd.Series):\n",
    "    abslen = chromsizes.cumsum()\n",
    "    starts = pd.Series([0] + abslen[:-1].tolist(), index=abslen.index)\n",
    "    return dict(starts)\n",
    "\n",
    "def prepare_tiles(df, starts):\n",
    "    tiles = bf.merge(df)\n",
    "    return [\n",
    "        {\n",
    "            \"chrOffset\": int(starts[l[0]]),\n",
    "            \"xStart\": int(starts[l[0]] + l[1]),\n",
    "            \"xEnd\": int(starts[l[0]] + l[2]),\n",
    "            \"importance\": 0,\n",
    "            \"uid\": slugid.nice(),\n",
    "            \"fields\": [s if isinstance(s, str) else int(s) for s in l]\n",
    "        } for l in tiles.to_records(index=False)\n",
    "    ]\n",
    "\n",
    "@dataclass\n",
    "class Tileset:\n",
    "    df: pd.DataFrame\n",
    "    chromsizes: pd.Series\n",
    "    uid: str = field(default_factory=slugid.nice)\n",
    "    datatype = \"bedlike\"\n",
    "        \n",
    "    def __post_init__(self):\n",
    "        self._starts = get_lens(self.chromsizes)\n",
    "        self._tiles = prepare_tiles(self.df, self._starts)\n",
    "\n",
    "        \n",
    "    def subset(self, locs):\n",
    "        if locs is None:\n",
    "            df = self.df\n",
    "        else:\n",
    "            df = self.df.iloc[locs]\n",
    "        self._tiles = prepare_tiles(df, self._starts)\n",
    "\n",
    "        \n",
    "    def info(self):\n",
    "        genome_length = int(np.sum(self.chromsizes.values))\n",
    "        return {\n",
    "            \"uuid\": self.uid,\n",
    "            \"max_width\": genome_length,\n",
    "            \"min_pos\": [1],\n",
    "            \"max_pos\": [genome_length],\n",
    "            \"max_zoom\": 0,\n",
    "        }\n",
    "     \n",
    "    def tiles(self, _tileids):\n",
    "        return [(f\"{self.uid}.0.0\", self._tiles)]\n",
    "        "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "0de8b820",
   "metadata": {},
   "outputs": [],
   "source": [
    "import hg\n",
    "\n",
    "chromsizes = bf.fetch_chromsizes('hg38')[:'chrY']\n",
    "t = Tileset(df, chromsizes)\n",
    "ts = hg.server.add(t)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "a2dfffbd",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b6227b4bcf1b4ca18fd97c1af58636f1",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HiGlassWidget()"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "remote_tileset = hg.remote(\n",
    "  uid='CQMd6V_cRw6iCI_-Unl3PQ',\n",
    "  server='https://higlass.io/api/v1/',\n",
    ")\n",
    "\n",
    "dynamic_track = ts.track()\n",
    "\n",
    "v = hg.view(\n",
    "    dynamic_track,\n",
    "    remote_tileset.track('heatmap', height=300),\n",
    ")\n",
    "\n",
    "conf = hg.viewconf(v)\n",
    "w = conf.widget()\n",
    "w"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "be626760",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "674c3909b4214a96b66a9699bd234592",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(VBox(children=(Button(button_style='primary', icon='arrows', layout=Layout(width='36px'), style…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import ipywidgets as widgets\n",
    "import jscatter\n",
    "\n",
    "scatter = jscatter.Scatter(x='E1', y='E2', color_by=\"GC\", data=df)\n",
    "scatter.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "db5e5251",
   "metadata": {},
   "outputs": [],
   "source": [
    "def on_selection_change(change):\n",
    "    t.subset(None if len(change[\"new\"]) == 0 else change[\"new\"])\n",
    "    reload_selector = { \"trackId\": dynamic_track.uid, \"viewId\": conf.views[0].uid }\n",
    "    w.reload(reload_selector)\n",
    "    \n",
    "scatter.widget.observe(on_selection_change, names=\"selection\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "7b0d90cc",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.9.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}