add placement comments to db

ktmeaton · Apr 15, 2021 · 4713040 · 4713040
1 parent c7566fc
commit 4713040
Show file tree

Hide file tree

Showing 2 changed files with 130 additions and 0 deletions.
diff --git a/results/sqlite_db/yersinia_pestis_db.sqlite b/results/sqlite_db/yersinia_pestis_db.sqlite
diff --git a/workflow/notebooks/dataset.py.ipynb b/workflow/notebooks/dataset.py.ipynb
@@ -0,0 +1,130 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 46,
+   "id": "collectible-letter",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import os\n",
+    "import sqlite3\n",
+    "import random\n",
+    "\n",
+    "random.seed(1234)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 59,
+   "id": "preliminary-tobago",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "2.MED1 SAMN03941711\n",
+      "0.ANT1 SAMN07722859\n",
+      "4.ANT1 SAMN05219223\n",
+      "0.PE2 SAMN05225434\n",
+      "2.MED0 SAMN07176224\n",
+      "0.PE5 SAMN12138635\n",
+      "1.ORI1 SAMN01831179\n",
+      "0.ANT3 SAMN02403038\n",
+      "2.ANT3 SAMN02403007\n",
+      "1.ANT1 SAMN10228581\n",
+      "1.ORI3 SAMN02470617\n",
+      "2.MED2 SAMN02403012\n",
+      "0.ANT2 SAMN02404401\n",
+      "1.ORI2 SAMN14943533\n",
+      "1.IN3 SAMN02403021\n",
+      "3.ANT2 SAMN02403094\n",
+      "0.PE4 SAMN03861481\n",
+      "0.ANT5 SAMN08866640\n",
+      "2.ANT1 SAMN13021974\n",
+      "1.IN2 SAMN02403011\n",
+      "2.MED3 SAMN13907443\n",
+      "1.IN1 SAMN02403014\n",
+      "2.ANT2 SAMN02403001\n",
+      "3.ANT1 SAMN02403051\n",
+      "0.PE7 SAMN02403027\n",
+      "0.PE10 SAMN02403015\n"
+     ]
+    }
+   ],
+   "source": [
+    "project_dir = os.path.dirname(os.path.dirname(os.getcwd()))\n",
+    "database = os.path.join(project_dir, \"results\", \"sqlite_db\", \"yersinia_pestis_db.sqlite\")\n",
+    "\n",
+    "query = \"\"\"\n",
+    "    SELECT\n",
+    "      BioSampleAccession,\n",
+    "      BioSampleBranch,\n",
+    "      BioSampleComment\n",
+    "    FROM\n",
+    "      BioSample\n",
+    "    LEFT Join\n",
+    "      Assembly ON BioSampleAccession==AssemblyBioSampleAccession\n",
+    "    WHERE\n",
+    "      BioSampleComment LIKE '%KEEP%Assembly%Modern%'\n",
+    "    \"\"\"\n",
+    "    \n",
+    "conn = sqlite3.connect(database)\n",
+    "cur = conn.cursor()\n",
+    "result = cur.execute(query).fetchall()\n",
+    "\n",
+    "# Create a mapping of branches to samples\n",
+    "branch_dict = {}\n",
+    "for rec in result:\n",
+    "    accession = rec[0]\n",
+    "    branch = rec[1]\n",
+    "    # Strip sub branches that are letter designations\n",
+    "    while branch[-1].isalpha():\n",
+    "        branch = branch[:-1]\n",
+    "    if branch not in branch_dict:\n",
+    "        branch_dict[branch] = []        \n",
+    "    branch_dict[branch].append(accession)        \n",
+    "\n",
+    "random_samples = {}\n",
+    "# Randomly sample\n",
+    "for branch in branch_dict:\n",
+    "    sample = random.choice(branch_dict[branch])\n",
+    "    random_samples[branch] = sample\n",
+    "\n",
+    "# print out\n",
+    "for branch in random_samples:\n",
+    "    print(branch, random_samples[branch])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "abandoned-jackson",
+   "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": "3.7.3"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}