Generators vs. lists for sequence filtering

lists_generators_and_laziness.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import os\n",
    "\n",
    "from Bio import SeqIO\n",
    "\n",
    "wd = '/Users/Bede/Research/Notebooks/res/2016-08-16'\n",
    "contigs_path = '/Users/Bede/Research/Notebooks/res/2016-08-16/31_c100.fa'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## List"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "10 loops, best of 3: 339 ms per loop\n"
     ]
    }
   ],
   "source": [
    "%%timeit -n 10\n",
    "\n",
    "def filter_length(records, min_len=0): # List comprehension\n",
    "    return [r for r in records if len(r.seq) >= min_len]\n",
    "\n",
    "def filter_longest(records, n=1): # List slicing; assumes seqs sorted by len\n",
    "    return records[:n]\n",
    "\n",
    "records = list(SeqIO.parse(contigs_path, 'fasta'))\n",
    "filtered_records = filter_length(records, 200)\n",
    "filter_longest(filtered_records, 100) # List"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Generator"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Good case; optimised by lazy eval"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "10 loops, best of 3: 2.34 ms per loop\n"
     ]
    }
   ],
   "source": [
    "%%timeit -n 10\n",
    "\n",
    "def filter_length(records, min_len=0): # Generator\n",
    "    return (r for r in records if len(r.seq) >= min_len)\n",
    "\n",
    "def filter_longest(records, n=1): # Generator friendly slice; assumes seqs sorted by len\n",
    "    return (x for _, x in zip(range(n), records))\n",
    "            \n",
    "records = SeqIO.parse(contigs_path, 'fasta')\n",
    "filtered_records = filter_length(records, 200)\n",
    "list(filter_longest(filtered_records, 100)) # List"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Bad case"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "10 loops, best of 3: 313 ms per loop\n"
     ]
    }
   ],
   "source": [
    "%%timeit -n 10\n",
    "\n",
    "def filter_length(records, min_len=0): # Generator\n",
    "    return (r for r in records if len(r.seq) >= min_len)\n",
    "\n",
    "def filter_longest(records, n=1): # Generator friendly slice; assumes seqs sorted by len\n",
    "    return (x for _, x in zip(range(n), records))\n",
    "            \n",
    "records = SeqIO.parse(contigs_path, 'fasta')\n",
    "filtered_records = filter_length(records, 500)\n",
    "list(filter_longest(filtered_records, 100)) # List"
   ]
  }
 ],
 "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": "3.5.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}