{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Using Metadensity with skipper outputs\n",
    "[Skipper](https://github.com/YeoLab/skipper) is a new eCLIP pipeline that is faster and more sensitive. \n",
    "\n",
    "Technically, Metadensity does not care how the bigwig files/bed/bams files are generated. Therefore, whether you process the data with skipper, or ENCODE pipeline, or whatever pipeline you wrote, as long as the file formats (.bigwig, .bam, .bed) are right, Metadensity runs.\n",
    "\n",
    "This notebook offers a shortcut for users that finds python annoying to copy paste from."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "please set the right config according to genome coordinate\n",
      "Using /home/hsher/gencode_coords/GRCh38.p13.genome.fa\n",
      "Using HG38 by default\n",
      "Using /tscc/nfs/home/hsher/gencode_coords/GRCh38.p13.genome.fa\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Matplotlib created a temporary config/cache directory at /tmp/matplotlib-z0irv727 because the default path (/home/jovyan/.cache/matplotlib) is not a writable directory; it is highly recommended to set the MPLCONFIGDIR environment variable to a writable directory, in particular to speed up the import of Matplotlib and to better support multiprocessing.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Using:  /tscc/nfs/home/hsher/gencode_coords/gencode.v33.transcript.gff3\n"
     ]
    }
   ],
   "source": [
    "# set up files associated with each genome coordinates\n",
    "import metadensity as md\n",
    "md.settings.from_config_file('/tscc/nfs/home/hsher/Metadensity/config/hg38-tscc2.ini')\n",
    "\n",
    "\n",
    "# then import the modules\n",
    "from metadensity.metadensity import *\n",
    "from metadensity.plotd import *\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "\n",
    "# I have a precompiles list of ENCODE datas as a csv that loads in this dataloader\n",
    "plt.style.use('seaborn-white')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## load RBPs into eCLIP object from Skipper outputs\n",
    "\n",
    "Notice the bigwig files skipper produce are 'read truncation counts', which is different from the coverage track from encode pipeline. Therefore, it is sparser than read coverage. the plot looks more noisy than coverage bigwig tracks. To calculate coverage tracks instead, you can make use of rules in [here](https://github.com/algaebrown/make_track/blob/master/Snake_CLIP_track.py)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "from pathlib import Path\n",
    "skipper_output_dir = Path('/tscc/projects/ps-yeolab5/encore/processing/skipper_output/')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'/tscc/nfs/home/hsher/gencode_coords/gencode.v33.transcript.gff3'"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "transcript.fn"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "def get_file_path(prefix, skipper_output_dir, ip_suffix = '_IP_',\n",
    "                 in_suffix = '_IN_', rep = 1):\n",
    "    return pd.Series({'uid':prefix+f'_rep{rep}',\n",
    "               'RBP': prefix+f'_rep{rep}',\n",
    "               'bam_0': str(skipper_output_dir/'bams'/'dedup'/'genome'/f'{prefix}{ip_suffix}{rep}.genome.Aligned.sort.dedup.bam'),\n",
    "               'bam_control_0':str(skipper_output_dir/'bams'/'dedup'/'genome'/f'{prefix}{in_suffix}{rep}.genome.Aligned.sort.dedup.bam'),\n",
    "               'minus_0': str(skipper_output_dir / 'bigwigs/unscaled' / 'minus' / f'{prefix}{ip_suffix}{rep}.unscaled.minus.bw'),\n",
    "               'minus_control_0': str(skipper_output_dir / 'bigwigs/unscaled' / 'minus' / f'{prefix}{in_suffix}{rep}.unscaled.minus.bw'),\n",
    "               'plus_0': str(skipper_output_dir / 'bigwigs/unscaled' / 'plus' / f'{prefix}{ip_suffix}{rep}.unscaled.plus.bw'),\n",
    "               'plus_control_0': str(skipper_output_dir / 'bigwigs/unscaled' / 'plus' / f'{prefix}{in_suffix}{rep}.unscaled.plus.bw'),\n",
    "                'bed_0': transcript.fn # This is not being used as the later function select transcripts. \n",
    "                      #OBSOLETE. newer version will not require this\n",
    "                      # function `find_gene_ids_with_reproducible_windows` dictates what genes to be used\n",
    "              }\n",
    "             )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "uid                False\n",
       "RBP                False\n",
       "bam_0               True\n",
       "bam_control_0       True\n",
       "minus_0             True\n",
       "minus_control_0     True\n",
       "plus_0              True\n",
       "plus_control_0      True\n",
       "bed_0               True\n",
       "dtype: bool"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# check if files exist.\n",
    "# if your bigwigs don't exist, need to uncomment the rule all output bigwig/ line\n",
    "get_file_path('4121', skipper_output_dir).apply(os.path.isfile)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "e = eCLIP.from_series(get_file_path('4212', skipper_output_dir))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# The reason I did not put enriched windows into `bed_*` and `idr_` is because their format is not really a bed file\n",
    "and that will crash Metadensity, as it is using BedTool.\n",
    "alternatively, I would use `transcripts=` kwargs to specify the transcripts containing reproducible windows"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "from itertools import chain\n",
    "def find_gene_ids_with_reproducible_windows(prefix, skipper_output_dir):\n",
    "    window_path = skipper_output_dir / 'reproducible_enriched_windows' / f'{prefix}.reproducible_enriched_windows.tsv.gz'\n",
    "    window = pd.read_csv(window_path, sep = '\\t')\n",
    "    \n",
    "    # find genes with reproducible windows. Some windows contain >1 gene id. `chain` flattens the list of list\n",
    "    # chain https://www.geeksforgeeks.org/python-itertools-chain/\n",
    "    gene_with_windows = list(set(chain(*window['gene_id'].apply(lambda idlist: idlist.split(':')).tolist())))\n",
    "    \n",
    "    # skipper is run with gencode v38 which is different from metadensity's version. need some conversion\n",
    "    # remove version\n",
    "    geneid_no_version = [g.split('.')[0] for g in gene_with_windows] # this is list comprehension. lazy and short for loops\n",
    "    # map to Metadensity's version\n",
    "    metadensity_gene_ids = [t.attrs['gene_id'] for t in transcript if t.attrs['gene_id'].split('.')[0] in geneid_no_version]\n",
    "    return metadensity_gene_ids"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "geneids = find_gene_ids_with_reproducible_windows('SNRNP200_HepG2', skipper_output_dir)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Using: /tscc/nfs/home/hsher/projects/Metadensity/metadensity/data/hg38/gencode\n",
      "Done building metagene\n"
     ]
    }
   ],
   "source": [
    "# here for the set of transcript, we use the IDR peak containing transcript assuming they have good signal\n",
    "m = Metadensity(e, e.name,background_method = 'relative information', normalize = False, transcript_ids = geneids)\n",
    "m.get_density_array()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'mean relative information')"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x216 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "f=plot_mean_density([m],\n",
    "                    features_to_show = ['intron', 'exon'], ymax = 0.00003)\n",
    "f=beautify(f, offset = 0) # sns.despine \n",
    "f.get_axes()[0].set_ylabel('mean relative information')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "metadensity",
   "language": "python",
   "name": "metadensity"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}