{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Colab initialization\n",
    "- install the pipeline in the colab runtime\n",
    "- download files neccessary for this example"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "!pip3 install -U pip > /dev/null\n",
    "!pip3 install -U bio_embeddings[all] > /dev/null"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "!wget http://data.bioembeddings.com/public/embeddings/notebooks/custom_data/tiny_sampled.fasta --output-document tiny_sampled.fasta"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Embed sequences in a FASTA file"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from bio_embeddings.embed import ProtTransBertBFDEmbedder\n",
    "from Bio import SeqIO"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "sequences = []\n",
    "for record in SeqIO.parse(\"tiny_sampled.fasta\", \"fasta\"):\n",
    "    sequences.append(record)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "embedder = ProtTransBertBFDEmbedder()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "embeddings = embedder.embed_many([str(s.seq) for s in sequences])\n",
    "\n",
    "# `embed_many` returns a generator.\n",
    "# We want to keep both RAW embeddings and reduced embeddings in memory.\n",
    "# To do so, we simply turn the generator into a list!\n",
    "# (this will start embedding the sequences!)\n",
    "\n",
    "embeddings = list(embeddings)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "reduced_embeddings = [ProtTransBertBFDEmbedder.reduce_per_protein(e) for e in embeddings]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "for (per_amino_acid, per_protein) in zip(embeddings, reduced_embeddings):\n",
    "    print(per_amino_acid.shape, per_protein.shape)"
   ]
  }
 ],
 "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": "3.7.6"
  },
  "pycharm": {
   "stem_cell": {
    "cell_type": "raw",
    "source": [],
    "metadata": {
     "collapsed": false
    }
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}