examples/borzoi_example_eqtl_chr10_116952944_T_C.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "7dbf2734",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2024-10-06 17:16:44.159991: E external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:9261] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered\n",
      "2024-10-06 17:16:44.160215: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:607] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered\n",
      "2024-10-06 17:16:44.484774: E external/local_xla/xla/stream_executor/cuda/cuda_blas.cc:1515] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered\n",
      "2024-10-06 17:16:45.302876: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.\n",
      "To enable the following instructions: AVX2 FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.\n",
      "2024-10-06 17:17:00.194555: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT\n"
     ]
    }
   ],
   "source": [
    "import json\n",
    "import os\n",
    "import time\n",
    "import warnings\n",
    "\n",
    "import h5py\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib.patches as patches\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import pysam\n",
    "import pyfaidx\n",
    "import tensorflow as tf\n",
    "\n",
    "from baskerville import seqnn\n",
    "from baskerville import gene as bgene\n",
    "from baskerville import dna\n",
    "\n",
    "from borzoi_helpers import *\n",
    "\n",
    "tf.compat.v1.logging.set_verbosity(tf.compat.v1.logging.ERROR)\n",
    "#os.environ['CUDA_VISIBLE_DEVICES'] = '-1'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "a6315e46-79ce-4653-ba71-242e74516b47",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "f3c0 model already exists.\n",
      "f3c1 model already exists.\n",
      "f3c2 model already exists.\n",
      "f3c3 model already exists.\n",
      "Gene annotation already exists.\n",
      "Gene annotation (no read-through, protein-coding) already exists.\n",
      "Gene annotation (protein-coding) already exists.\n",
      "TSS annotation already exists.\n",
      "Splice site annotation already exist.\n",
      "Splice site annotation already exist.\n",
      "PolyA site annotation already exist.\n",
      "Human genome FASTA already exists.\n"
     ]
    }
   ],
   "source": [
    "%%bash\n",
    "\n",
    "#Download model weights (data fold 3, 4 replicates)\n",
    "for rep in f3c0,f0 f3c1,f1 f3c2,f2 f3c3,f3; do IFS=\",\"; set -- $rep; \n",
    "  mkdir -p \"saved_models/$1/train\"\n",
    "  local_model=\"saved_models/$1/train/model0_best.h5\"\n",
    "  if [ -f \"$local_model\" ]; then\n",
    "    echo \"$1 model already exists.\"\n",
    "  else\n",
    "    wget --progress=bar:force \"https://storage.googleapis.com/seqnn-share/borzoi/$2/model0_best.h5\" -O \"$local_model\"\n",
    "  fi\n",
    "done\n",
    "\n",
    "#Download and uncompress annotation files\n",
    "mkdir -p hg38/genes/gencode41\n",
    "mkdir -p hg38/genes/polyadb\n",
    "\n",
    "if [ -f hg38/genes/gencode41/gencode41_basic_nort.gtf ]; then\n",
    "  echo \"Gene annotation already exists.\"\n",
    "else\n",
    "  wget -O - https://storage.googleapis.com/seqnn-share/helper/gencode41_basic_nort.gtf.gz | gunzip -c > hg38/genes/gencode41/gencode41_basic_nort.gtf\n",
    "fi\n",
    "\n",
    "if [ -f hg38/genes/gencode41/gencode41_basic_nort_protein.gtf ]; then\n",
    "  echo \"Gene annotation (no read-through, protein-coding) already exists.\"\n",
    "else\n",
    "  wget -O - https://storage.googleapis.com/seqnn-share/helper/gencode41_basic_nort_protein.gtf.gz | gunzip -c > hg38/genes/gencode41/gencode41_basic_nort_protein.gtf\n",
    "fi\n",
    "\n",
    "if [ -f hg38/genes/gencode41/gencode41_basic_protein.gtf ]; then\n",
    "  echo \"Gene annotation (protein-coding) already exists.\"\n",
    "else\n",
    "  wget -O - https://storage.googleapis.com/seqnn-share/helper/gencode41_basic_protein.gtf.gz | gunzip -c > hg38/genes/gencode41/gencode41_basic_protein.gtf\n",
    "fi\n",
    "\n",
    "if [ -f hg38/genes/gencode41/gencode41_basic_tss2.bed ]; then\n",
    "  echo \"TSS annotation already exists.\"\n",
    "else\n",
    "  wget -O - https://storage.googleapis.com/seqnn-share/helper/gencode41_basic_tss2.bed.gz | gunzip -c > hg38/genes/gencode41/gencode41_basic_tss2.bed\n",
    "fi\n",
    "\n",
    "if [ -f hg38/genes/gencode41/gencode41_basic_protein_splice.csv.gz ]; then\n",
    "  echo \"Splice site annotation already exist.\"\n",
    "else\n",
    "  wget https://storage.googleapis.com/seqnn-share/helper/gencode41_basic_protein_splice.csv.gz -O hg38/genes/gencode41/gencode41_basic_protein_splice.csv.gz\n",
    "fi\n",
    "\n",
    "if [ -f hg38/genes/gencode41/gencode41_basic_protein_splice.gff ]; then\n",
    "  echo \"Splice site annotation already exist.\"\n",
    "else\n",
    "  wget -O - https://storage.googleapis.com/seqnn-share/helper/gencode41_basic_protein_splice.gff.gz | gunzip -c > hg38/genes/gencode41/gencode41_basic_protein_splice.gff\n",
    "fi\n",
    "\n",
    "if [ -f hg38/genes/polyadb/polyadb_human_v3.csv.gz ]; then\n",
    "  echo \"PolyA site annotation already exist.\"\n",
    "else\n",
    "  wget https://storage.googleapis.com/seqnn-share/helper/polyadb_human_v3.csv.gz -O hg38/genes/polyadb/polyadb_human_v3.csv.gz\n",
    "fi\n",
    "\n",
    "#Download and index hg38 genome\n",
    "mkdir -p hg38/assembly/ucsc\n",
    "\n",
    "if [ -f hg38/assembly/ucsc/hg38.fa ]; then\n",
    "  echo \"Human genome FASTA already exists.\"\n",
    "else\n",
    "  wget -O - http://hgdownload.cse.ucsc.edu/goldenPath/hg38/bigZips/hg38.fa.gz | gunzip -c > hg38/assembly/ucsc/hg38.fa\n",
    "fi\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "f3dfe8ad-5c40-44b1-aab6-58491694da5d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Faidx(\"hg38/assembly/ucsc/hg38.fa\")"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pyfaidx.Faidx('hg38/assembly/ucsc/hg38.fa')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "e5fbf3da",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2024-10-06 09:35:29.222863: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1929] Created device /job:localhost/replica:0/task:0/device:GPU:0 with 10520 MB memory:  -> device: 0, name: NVIDIA GeForce GTX 1080 Ti, pci bus id: 0000:02:00.0, compute capability: 6.1\n"
     ]
    }
   ],
   "source": [
    "#Model configuration\n",
    "\n",
    "params_file = 'params_pred.json'\n",
    "targets_file = 'targets_gtex.txt' #Subset of targets_human.txt\n",
    "\n",
    "seq_len = 524288\n",
    "n_reps = 1       #To use only one model replicate, set to 'n_reps = 1'. To use all four replicates, set 'n_reps = 4'.\n",
    "rc = True         #Average across reverse-complement prediction\n",
    "\n",
    "#Read model parameters\n",
    "\n",
    "with open(params_file) as params_open :\n",
    "    \n",
    "    params = json.load(params_open)\n",
    "    \n",
    "    params_model = params['model']\n",
    "    params_train = params['train']\n",
    "\n",
    "#Read targets\n",
    "\n",
    "targets_df = pd.read_csv(targets_file, index_col=0, sep='\\t')\n",
    "target_index = targets_df.index\n",
    "\n",
    "#Create local index of strand_pair (relative to sliced targets)\n",
    "if rc :\n",
    "    strand_pair = targets_df.strand_pair\n",
    "    \n",
    "    target_slice_dict = {ix : i for i, ix in enumerate(target_index.values.tolist())}\n",
    "    slice_pair = np.array([\n",
    "        target_slice_dict[ix] if ix in target_slice_dict else ix for ix in strand_pair.values.tolist()\n",
    "    ], dtype='int32')\n",
    "\n",
    "#Initialize model ensemble\n",
    "\n",
    "models = []\n",
    "for rep_ix in range(n_reps) :\n",
    "    \n",
    "    model_file = \"saved_models/f3c\" + str(rep_ix) + \"/train/model0_best.h5\"\n",
    "\n",
    "    seqnn_model = seqnn.SeqNN(params_model)\n",
    "    seqnn_model.restore(model_file, 0)\n",
    "    seqnn_model.build_slice(target_index)\n",
    "    if rc :\n",
    "        seqnn_model.strand_pair.append(slice_pair)\n",
    "    seqnn_model.build_ensemble(rc, [0])\n",
    "    \n",
    "    models.append(seqnn_model)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "6f010781",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "len(splice_df) = 404837\n"
     ]
    }
   ],
   "source": [
    "#Initialize fasta sequence extractor\n",
    "\n",
    "fasta_open = pysam.Fastafile('hg38/assembly/ucsc/hg38.fa')\n",
    "\n",
    "#Load splice site annotation\n",
    "\n",
    "splice_df = pd.read_csv('hg38/genes/gencode41/gencode41_basic_protein_splice.csv.gz', sep='\\t', compression='gzip')\n",
    "\n",
    "print(\"len(splice_df) = \" + str(len(splice_df)))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "12df90e4",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Load GTF (optional; needed to compute exon coverage attributions for example gene)\n",
    "\n",
    "transcriptome = bgene.Transcriptome('hg38/genes/gencode41/gencode41_basic_nort.gtf')\n",
    "\n",
    "search_gene = 'ENSG00000187164'\n",
    "center_pos = 116952944\n",
    "chrom = 'chr10'\n",
    "poses = [116952944]\n",
    "alts = ['C']\n",
    "\n",
    "start = center_pos - seq_len // 2\n",
    "end = center_pos + seq_len // 2\n",
    "\n",
    "#Get exon bin range\n",
    "gene_keys = [gene_key for gene_key in transcriptome.genes.keys() if search_gene in gene_key]\n",
    "\n",
    "gene = transcriptome.genes[gene_keys[0]]\n",
    "\n",
    "#Determine output sequence start\n",
    "seq_out_start = start + seqnn_model.model_strides[0]*seqnn_model.target_crops[0]\n",
    "seq_out_len = seqnn_model.model_strides[0]*seqnn_model.target_lengths[0]\n",
    "\n",
    "#Determine output positions of gene exons\n",
    "gene_slice = gene.output_slice(seq_out_start, seq_out_len, seqnn_model.model_strides[0], False)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "073e4711",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "blood tracks = [9, 10, 11]\n",
      "muscle tracks = [47, 48, 49]\n"
     ]
    }
   ],
   "source": [
    "#Print index of GTEx blood and muscle tracks in targets file\n",
    "\n",
    "targets_df['local_index'] = np.arange(len(targets_df))\n",
    "\n",
    "print(\"blood tracks = \" + str(targets_df.loc[targets_df['description'] == 'RNA:blood']['local_index'].tolist()))\n",
    "print(\"muscle tracks = \" + str(targets_df.loc[targets_df['description'] == 'RNA:muscle']['local_index'].tolist()))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "9d715d82",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2024-10-05 15:08:40.291698: I external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:454] Loaded cuDNN version 8907\n",
      "2024-10-05 15:08:40.364169: I external/local_tsl/tsl/platform/default/subprocess.cc:304] Start cannot spawn child process: No such file or directory\n",
      "2024-10-05 15:08:40.675239: I external/local_tsl/tsl/platform/default/subprocess.cc:304] Start cannot spawn child process: No such file or directory\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-- Counts --\n",
      " - sum_wt = 138084.66\n",
      " - sum_mut = 140380.31\n",
      " - max_y_wt = 1051.197\n",
      " - max_y_mut = 1060.9895\n",
      " -- (max_y = 1060.9895)\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x200 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " - sum_wt = 77193.69\n",
      " - sum_mut = 93064.84\n",
      " - max_y_wt = 696.2884\n",
      " - max_y_mut = 837.6076\n",
      " -- (max_y = 837.6076)\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x200 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " - sum_wt = 45951.043\n",
      " - sum_mut = 46931.766\n",
      " - max_y_wt = 425.3462\n",
      " - max_y_mut = 430.3084\n",
      " -- (max_y = 430.3084)\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x200 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 57.7 s, sys: 937 ms, total: 58.7 s\n",
      "Wall time: 1min 1s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "#Predict for chr10_116952944_T_C\n",
    "# (~6 minutes on CPU w 1 replicate; ~2 minutes on GPU)\n",
    "\n",
    "save_figs = False\n",
    "save_suffix = '_chr10_116952944_T_C'\n",
    "\n",
    "sequence_one_hot_wt = process_sequence(fasta_open, chrom, start, end)\n",
    "\n",
    "#Induce mutation(s)\n",
    "sequence_one_hot_mut = np.copy(sequence_one_hot_wt)\n",
    "\n",
    "for pos, alt in zip(poses, alts) :\n",
    "    alt_ix = -1\n",
    "    if alt == 'A' :\n",
    "        alt_ix = 0\n",
    "    elif alt == 'C' :\n",
    "        alt_ix = 1\n",
    "    elif alt == 'G' :\n",
    "        alt_ix = 2\n",
    "    elif alt == 'T' :\n",
    "        alt_ix = 3\n",
    "\n",
    "    sequence_one_hot_mut[pos-start-1] = 0.\n",
    "    sequence_one_hot_mut[pos-start-1, alt_ix] = 1.\n",
    "\n",
    "#Make predictions\n",
    "y_wt = predict_tracks(models, sequence_one_hot_wt)\n",
    "y_mut = predict_tracks(models, sequence_one_hot_mut)\n",
    "\n",
    "#Visualize quantized tracks over SNP\n",
    "\n",
    "plot_window = 131072\n",
    "bin_size = 32\n",
    "pad = 16\n",
    "\n",
    "untransform_old = True\n",
    "normalize_counts = False\n",
    "\n",
    "anno_df = None #splice_df\n",
    "\n",
    "#Tracks\n",
    "track_indices = [\n",
    "    np.arange(0, 89).tolist(),\n",
    "    [9, 10, 11],\n",
    "    [47, 48, 49],\n",
    "]\n",
    "\n",
    "track_names = [\n",
    "    'GTEx Coverage (All tissues)',\n",
    "    'GTEx Coverage (Blood)',\n",
    "    'GTEx Coverage (Muscle)',\n",
    "]\n",
    "\n",
    "track_scales = [0.01]*3\n",
    "track_transforms = [3./4.]*3\n",
    "soft_clips = [384.]*3\n",
    "\n",
    "print(\"-- Counts --\")\n",
    "plot_coverage_track_pair_bins(\n",
    "    y_wt,\n",
    "    y_mut,\n",
    "    chrom,\n",
    "    start,\n",
    "    center_pos,\n",
    "    poses,\n",
    "    track_indices,\n",
    "    track_names,\n",
    "    track_scales,\n",
    "    track_transforms,\n",
    "    soft_clips,\n",
    "    plot_window=plot_window,\n",
    "    normalize_window=1 * plot_window,\n",
    "    bin_size=bin_size,\n",
    "    pad=pad,\n",
    "    normalize_counts=normalize_counts,\n",
    "    save_figs=save_figs,\n",
    "    save_suffix=save_suffix,\n",
    "    gene_slice=gene_slice,\n",
    "    anno_df=anno_df,\n",
    "    untransform_old=untransform_old,\n",
    ")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "046d763d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 394s 394s/step\n",
      "1/1 [==============================] - 206s 206s/step\n",
      "1/1 [==============================] - 387s 387s/step\n",
      "1/1 [==============================] - 206s 206s/step\n",
      "CPU times: user 17min 35s, sys: 2min 38s, total: 20min 13s\n",
      "Wall time: 22min 9s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "#Get contribution scores (gradient) for blood GTEX tracks\n",
    "# (~20 minutes on CPU; otherwise runnable only on 40GB GPU cards, e.g. A100)\n",
    "\n",
    "_, _, [pred_grad_wt, pred_grad_mut] = get_prediction_gradient_w_rc(\n",
    "    models,\n",
    "    [sequence_one_hot_wt, sequence_one_hot_mut],\n",
    "    prox_bin_start=0,\n",
    "    prox_bin_end=1,\n",
    "    dist_bin_start=0,\n",
    "    dist_bin_end=1,\n",
    "    track_index=target_index[[9, 10, 11]].tolist(),\n",
    "    track_scale=0.01,\n",
    "    track_transform=3./4.,\n",
    "    clip_soft=384.,\n",
    "    dist_bin_index=gene_slice.tolist(),\n",
    "    use_mean=False,\n",
    "    use_ratio=False,\n",
    "    use_logodds=False,\n",
    "    subtract_avg=True,\n",
    "    fold_index=np.arange(n_reps).tolist(),\n",
    "    untransform_old=True,\n",
    ")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "199003f5",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "--- WT ---\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x100 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "--- Mut ---\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x100 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Visualize gradient contribution scores (128 bp window centered on variant)\n",
    "\n",
    "visualize_input_gradient_pair(\n",
    "    pred_grad_wt,\n",
    "    pred_grad_mut,\n",
    "    plot_start=(poses[0] - start) - 64,\n",
    "    plot_end=(poses[0] - start) + 64,\n",
    "    save_figs=False,\n",
    "    fig_name=chrom + '_' + str(poses[0]) + '_prediction_grad_gtex_blood'\n",
    ")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "1920b078",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "example_ix = 0\n",
      "example_ix = 1\n",
      "CPU times: user 12min 6s, sys: 2min 19s, total: 14min 26s\n",
      "Wall time: 33min 48s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "#Get contribution scores (ISM) for blood GTEX tracks\n",
    "# (not feasible to run on CPU; ~33 minutes on GPU)\n",
    "\n",
    "[pred_ism_wt, pred_ism_mut] = get_ism(\n",
    "    models,\n",
    "    [sequence_one_hot_wt, sequence_one_hot_mut],\n",
    "    ism_start=(poses[0] - start) - 64,\n",
    "    ism_end=(poses[0] - start) + 64,\n",
    "    prox_bin_start=0,\n",
    "    prox_bin_end=1,\n",
    "    dist_bin_start=0,\n",
    "    dist_bin_end=1,\n",
    "    track_index=[9, 10, 11],\n",
    "    track_scale=0.01,\n",
    "    track_transform=3./4.,\n",
    "    clip_soft=384.,\n",
    "    dist_bin_index=gene_slice.tolist(),\n",
    "    use_mean=True,\n",
    "    use_ratio=False,\n",
    "    use_logodds=False,\n",
    "    untransform_old=True,\n",
    ")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "0b2ecd61",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "--- WT ---\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x100 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "--- Mut ---\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAxYAAABZCAYAAACjWLKDAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8pXeV/AAAACXBIWXMAAA9hAAAPYQGoP6dpAAA6rElEQVR4nO2dd3xUZb7/35PeQxqQhB46IoqAXUERUFz8KdaLa111dXevK7oXV1fA6/W6V1dsu3ZEWHtD7EhHKSK9hpqE9J6ZSaZk2u+PZ85MJkxCyqR/369XXmfOM2eeeXJm5pzn83ybzuVyuRAEQRAEQRAEQWgFQR09AEEQBEEQBEEQuj4iLARBEARBEARBaDUiLARBEARBEARBaDUiLARBEARBEARBaDUiLARBEARBEARBaDUiLARBEARBEARBaDUiLARBEARBEARBaDUhTTnI6XRSUFBAbGwsOp2urcckCIIgCIIgCEInwOVyYTQaSUtLIyiocZtEk4RFQUEB/fv3D8jgBEEQBEEQBEHoWuTm5tKvX79Gj2mSsIiNjfV0GBcX1/qRCYIgCIIgCILQ6TEYDPTv39+jBxqjScJCc3+Ki4sTYSEIQrvw0taXMNlM/PXiv3b0UHosLheI96sgCIIANCkcQoK3BUHolDy54UkeW/sYdqe9o4fSddDpTv1rrL0RfvkFkpLg1VfbeMyCIAhCt0GEhSAInQ6Xy4XBagDwbIX2ZeFCqKyEefOgqqrl/RwtP8qG7A1UmisDNTRBEAShkyLCQhCETkeNrQaHywGA3qLv4NH0PE6ehJUr1ePqali+vOV9PfPzM0xeOpnNuZsDMjZBEASh8yLCQhCETkddMaG3irBob1avVvEVGqWlLe+rylLlsxUEQRC6LyIsBEFoH44vhpXnQfmvpz20rpjo7K5Q+YZ8jlccx+F0dPRQAsa2bYHrS4SFIAhCz0GEhSAIbY+lFHY8BOW/wE/Xgd3U6OE+FotO7gr1mw9/w9BXhlJUXdTRQwkYe/cGrq+GhMWLW19k8ruT2Zq3NXBvJgiCIHQoIiwEQWh7Dj0LdqN6bMqDk580enhdi0Vnd4WqtFT6bLsDx4+rbXJy69PNNiQsfi34lQ05GzhReaJ1byAIgiB0GppUx0IQBKFV5K3w3a/1TsIfW/MYFruFRdMXedq6ksWiwlzhs+3qVFdDSYl6vGEDfPll6/prSFhoWaIkW5QgCEL3QSwWgtCDcLlc/O6r3/HX1e1YdM5cBMajDT79wtYXeGHrC9gcNk+bFlcRGhTaqS0WdqfdM9ZOMUGuG3Htcnn3G2r3gyYqJk2C0aPh/vshPLxlw3G6nJ7Pr8pa5fNcdxNkgiAIgggLQehR6K16Fu9azJs732zHN93f4FMmmwmL3QL4TjD1Vj2hQaEkRyV3aotF3VX47jJB1mpWnH++2iYkwEUXtayv6tpqnC6n6reexUKEhSAIQvdDhIUg9CDKTGWAWl1vt4rW+ky1DU+Cyd9B7DDPU+Wmcu9js/ex3qInLjyO+Ij4trNYOG1qbI2s3p+OulaK7hJjoQmLjAxv29lnt7Avt5iICYs5RSB2x9gUQRCEno4IC0HoQWjCwoWr/Vx3TDlqO3IupF0JE1/3PFVXTNQVGXqrEhZx4XFtIyxcLth0M3w7Cn6aDXVSxdbU1mC0GnE1QXDUXW3vFK5QAUATFn37etuCWnin0ITFwPiBPhYLp8spFgtBELoUpaXw9ttQ2T0u9W2GCAtB6EE0ZCFwupxc+f6VzPliTuDf1OqeOKb/Rm37TIH4MY2Ox0dYtIUrVM7HkPuFepy3HE4s8Tw14a0JxP09zmc8DaGttocEhXTdCbLxGGx/EAp/BLzCok+f1netiYn+8f19hIXRavS4SHXZ8yYIQrfEbIaNG8HmDfvD5YJbb4V77oGrrgKns+PG19kRYSEIPQjNYlH/caW5kh+O/cAXh75o0kp9s6itAHQQO0Lt63TQ53KgEYuFpY0tFkde8d2vE1xeUlPis20MzUoxIH5A13TpcVjh5xvgyMuw/kooXBVQYaGJwv5xSlho3y3tXKXGpHbN8yYIQudHp/P/1wguF9x+O1x6Kdx0k9dT9sMP4Ue19sLWrbBuXRuPvQsjwkIQehANTeS1SbTFbsFYawzsm1rLIaIPBId524KCfcYQGhTafhaL2ioo2+L3KZvD5llBL64u9nlu+aHlvLXjLax2q6dNO3Zg/MCuufJ+9FWo3K0eu5xw+GX07tMdE9P67jUrxYD4AdicNsx2M+A9b4MTBnfN8yYIQtelEcHx+efw6afqsOXLYYv7VvHBB75d1NS043i7GCIsBKEHUWYqIzYs1vNYo7jGO4muP6FuNbUVENXf71Pl5nJiwmJIjko+xWIRGx5LbFhs4C0WlbsA9zJU2kwIjvI8VWoq9Tyub7F4ZNUj3PvNveQb871dWSqJCo2id3TvrrnynvNxvQYnVrduCgs75ehm43GFiuvvs+8RFr0GU2muDLyVTBC6OvKbaDa5ufDOO153zpbwzju++1VV4HDA+vWtGFgPQ4SFIPQgykxl9IvrR3hwuI+FoK6YaIoLULOorVAWCz+Um8pJjEwkMTLxVItFWBtZLKrc6W/7Xw+Tv4GLP/c8Vfd/r/vY5XJRaCwE8GxBTZATIhJIiEjoesHbdhNUbFePg7wqwu5OFhYoYREREkFKdIpnH7wuZIN7DcbqsHosGYLQ47EZ4ccL4LMEOPr6aQ83WA2kPp/K+YvPb4fBdV4MBpg2De6+G6ZOhdra5vdhs/l3cTpyRFkoQkLg/fdVjR+hYURYCEIPotxcTlJUEomRiT4Wi7qT6LrWi4BgM0BwRIPjSYhIICEywW+62bjwOIy1Rhx1sja1GrNbGAz7vdqmzVB/+AqsuufBYDV4Jr+F1V5hUWmppFdEL3pF9Gp3l54DJQf4KecnTDZTyzqo2gsuB4Qnw//LhYs+A3Q43Ke6ucLC4YDNm30zplRZqogPjyc+PN6zD16LxaBeg3z2BaHH88vdylXTpodf74fyXxs9vMBYQFF1EXuL9/YYy9/8dfN58PsHPQkgABYuhEx3ZvMdO+CXXzi1GOhpioMeOAAWC0RGwvffqzgLgGPH1PbOO+E//gPeffe0oRo9GhEWgtCDKDOVeS0EdVyPimuKPZO/gLtCOe0Q7L90c7m5/JTxuFwunxgLwCfu4+OP4dprW2Gatrj/v4Q6xRl6Twa8Ais9Nt1HbNUVE/UtFpqwqLRU+tzo2pr7vrmPS969hD1Fe1rWgSlXbYf/J0T0hgGzYdAcj7AIDW16Vy4X/OlPcOGFcMYZcPKkaq+yVPl8jnWFRbAumH5x/YDuk6pXEFqF8Tic/NS3zVzU6EvyDco102QztV3Nn06Ey+Xi/zb9Hy9ve9mzOOZ0nhoD0ajGakBwHD+udv/4R5gxA5YsgeBgKHbfMmao9SdGjYILLgjM/9MdEWEhCD2Iuq5HZeY6MRbVxaTHpRMfHh94VyiXHYLcwmLTHFieDtt+f8p4NIuF2W7G7rT7TEg1d6gffoA5c+DLL+HKKyEnpwXjsRRDaC8IT/S2uZefimuKCdYFMzRxqK+wqCMmfCwWZq/FwulyYrQGOPC9EU7q1ew915Dbsg40y01KnTvkgBtb1NXy5fDaa+pxQQEsXaoeV1mriI+IJz5CiVbtc6y0VJIQqSxVIBYLQQCg4Dvv47BEn6fsTjufHPiElcdW+rTXjfnSREZ3ptRUSq1D+Tnl6tW1LzNTTf51Orj3XkhIaFnfRW4NN3Wq2g4eDBdf7BUWw4d7j01Katl79AREWAhCD6LMVEZSZBJJUUm+rlCmElKiUkiJTgmsK5TLpdxtNItFbTmYC6BWrVB7XKEiEjyTS23yWVdYGKwGAB57DM+KusUC+S25j1qKIWaQ36dKakpIjko+5TxoYqJfXD8fYVFhrvCZILdXALfD6aDAWAB4BUaz0VZCY+vcLYOCCQlRD5vjo1w/4FFbDGzMFUoTZNq+IPR4tGx1Y/8bZpfCsAc8T53Un+Smz27ijhV3+LykrpioKzK6K5qYAMgz5AGwe7fa//3v4Y03lFW7JRRqXrLDvG1RUV5hMXhwy/rtaYiwEIRuisPp8IlNcLlcXtejiHquUNXFpESnkBKVcqrFojobilarYN+W0oBd2sdiYSr3uEGBr7DQW/VkZ8OuXep1EycqE3WLqK2EkFi/T5XUlPg9D4XGQoJ1wYxOGe1jvai0VNIrvP0nyIXVhThc6rOte6NtFpZCCAqFqHSfZu28NlVY1A14XLRI5YDXqLIoi0VseCw6dD7CQhOU0H6CTBA6FdUnoGwraNfpmmyVpW7UI6ALgvGLIGYIANlV2QAUVRdhtnmTHeQb8+kTrZJjaIsNDdKCug6dDU1MgNdam+duuvlmtb3iCjjzzOb3XVSkArT710tiWFIC0dHqTzg9IiwEoRtSYa4g5KkQ+r3Qz9NmsBqwO+1eV6h66Wb9WizyvoZvR8PaK+CbkWA43LyB6HSgCwan9ZSnHE4HVZYqz3hsThvVtdV+LRZ6i57Nm9XrbrsNtm2DZcuaNxQPzlpvMHnVfihc5QmQ9JyH+sKiupDe0b3pG9P3FItF3ZX39ooV0MRETFhMy12hrBUQEqMmMHVorrDIyQGTCa6/Hh56CBYvhvHj1XOaxSJIF0RseKw3K1SdoHcQi4XQA8l6T11TfzwfVk4ASwmY8qDXWAiJVMcEh0OvMYBXWICvlbLAWMDwpOFEhUb1CFeoXEMuOnSkRKV4roOa5XrMGO9xvXo1v+/SUoiLw2O11TCZVEC30DRCTn+IIAhdjaPlqpJ0UXURRquR2PBYj5BIjEykMrKSSkslDqeD4KBgtVIflUKto5ZDpYdUJ9Zy2DwHHO7VMVMulGyEuBHNG4wuRE3m61FpqcSFi4TIBE9tjXJzucdiMW/1PEKDVQSx3qonK0u97v771faWW1SKwWbjtEGw+y5x6B+QtRR6XwJTN1BSU8LQxKGkRKdQXVuNyWYiKjSKAmMBfWP60je6r8diYbFbsNgtvsKinVbeNTExMW1iy4WFs9Zbw6Nyj1o5DQ4nKuoOAE89i9OhxblofsnBwTBzpnqsBW+DEop1LRYjk0cSGhxKdGi0BG8LPYvqE7D1ThV/BqpIZfk25Sba9wq/L6krLLKrshmRrK7D+cZ8BvUaRKoxtce4QvWO7k3/+P7kGZWpIi9PxVW0Nu7BavUvIGw2ERbNQSwWgtANOVpx1PP4WIXKlacFR2sWAqfLSZWlyjOBPsUFKOdjsBshPEVlDopMa9lggkLB4Z6lJp0LocrfXnPF0sajtWkWi11Fu9iWvw1QFoucHGUA0UzcOh3Ex7dkQC5lRfGDJrBSolI8+6AsFn1i+tAnpg+lplJsDptnMpwQ6XXpaa+V91x9LsG6YM7ue3bLXaHqCqyiVfDr72Hnw56VPq0C9+nQhMUZZ3jbdDp3di+Lnjd2vEHKcykUGAuoslYBbktPuHqjjkjV2xUoM5Wxs3AnRdWNZwUSuiDHlyhREdUfMu6B0DiVltvlUBna/JBVlcXolNGAr8jIN+STGpNKamwThEUz0692RvKMeaTHpZMem+659pWXtzxguy4NCQiHA8L9JzYU/CDCQhC6IUfKjxAWHOZ5DN5K21rwNiixoaWX1Vyh9FY9FrsFSjepzqZ8DxNeghk7IHpg8wcT1gsc7viMM5/0BAvXFzpam7+UiXqrEhb9+qlgulahCwGH5ZRml8vlFRbR9YSFsZA+0X08vszFNcWeyfAzPz/D1R9eDfi6Qm3cCH/7m8pkFWhyDbn0jenLgPgBFNcUY7U30bxQF50OTwXyOmjCoriJMfwFbrfutHq6s8ZWg8PlwGK3UGYq8whZUOdp48mN3Lb8NgxWAxUWERb1efXXVznnzXN4euPTHT0UIdCU/qQWXKZugHPfhGlbIMjtnhniduQ/sRQOvwylygc0uyqbEUkj6B3dm6wqZb51OB0UVRcpYRGTevoYi25Arj6XtNg00mLTPPEWtbUBuC+gNFaQn1lxcHDLCu71VMQVShC6Oi6nsgiEeJdajlYc5ay+Z3Gg5IDHeqEJi/f2vueZyJWZytChgvdSolM8afxKakoYYMiEmAxIPEd1GtlX/TWXsES/udg1i8XdX91NkNvPv67Foi56ix6DAWL9x1w3j6BQr3tX3few6ql11JIclUxyVDLgrelRWO0WFjFKWBQaC7G6rTCaRQi8FouXX4YHH/T2/d13Kj1uoMg15KpVuzgVeJ1vzGdIwpDmdaLzfx40YVHUxIVys7uLmBjfdk1E1G+z2q3U2GrILMsks0xVtGorVyi9Xk04mlOTo7Owt3iv2pbs7eCR9EwOlx3m7q/uZkzKGN74zRuB7dyQCUnnQYw7zVD8aAhyV6TUrIj7n4Lq4zDmMUi5gOyqbM5JPYdcQ67HYlFcU4zD5SA1NpVUYyo/n/w5sOPshOQZ8hidMpr02HTyDHk4XU7s9qCAWBRCQ1W2QX/t5lMvlUIDiMVCELoyR1+Dz5PhkyhYeR4YlIg4Wn6UjIQMMhIzPMJCm8gv2rqId3e/CyhhoQVrR4dGExOmZocl1cVgPOIVFa0hLBHMp5roNYtFdlU2JypPeNoasljU1gYoK0dIjHdCPXYhpFwMeEWE2W72nKuSmhJMNhMGq0HFWMQoYVVYXejXfafSUkl5OTz6qFrluuUWlbrQGODyFrn6XNJjlTuAtt9sgsO85yGir/rDv7Bo7KZqs6ltfReChoSFvziUQLtClZXBddep/yUlBT75JKDdtwt7i/cSrAvuURWVOxNb8rawKXcTH+z/wCe7XlF1EQ+vfJiXtr7Uso5r9SrldeIE33bNPVOLu6iD1W4l35DPgPgBDIgf4BEWmoVCc4UqrinG7jz19d0Fp8tJniFPXfvi0rE5bZTWlBIS0vSYsMYIC1OB2vWJiPAvOLpDlq22QISFIHRVcpfDrw+AzagCqit+harduFwujpQfYUjCEIYkDPEEctfNAqVRbvK6Qp23+Dyu++Q6AEqMBWCvhnB3NFzxBvV+FTubP87wRJXxxOFrS66b7rZum1+LhVWPTqcqrLaaiD7e4nAxg5SrFl63p7+s+guXLbvM06YFa9fYajz+7oXGQr+r7BXmCtasURPx559X1WB37YKxYwMw7jrkGnI9N1dtv9mEJanvjtMBg2+FIXcCXmGRmek99OdGFkK1DCr1b+z+hIXeom/wvAWS++5TRfvS05XgaXGV9kBSsQP2zoeD/3fa7Go1tTUcqzjGZYMvo8pS5ZNi83DZYd7d/S47C1vwWxSazPaC7ejQUV1b7XEnBVifvZ5FWxfxt3V/8xEcTabW/V3Xrq1Fq1X2vRr3b9hec8pLcg25uHApYRHnFRZaFqjUWOUK5XQ5Pdeo1avh/POV++jvfheYiXdHU1JTgs1p87hCgTo3DQmC5pKQoBaB6uv43r1Ve3OsFiaTugfceCM89VQLE400Qm4u/Pa3KjXuFVf4Xq87GhEW3YiCArVS1+GIgm8fjr2h0qZeuROuzoSZByFmCCU1JRhrjcpikVDHYmE+dSJfZirzW2m7rMZtYdD8ffc+Dj9dB0f+1fxxhiUCLjAc9Gn2N5ls0GJh0RMeDtXVzX/7U4joDea8U4SOv/NQXFPsSS/7xLonuOLfKmNLYxaLHTvUYy2nenQ0jBoVgHG7sdqtFFUX0Tu6NwkRCejQtcxiEZmqVkfrWZM0YbFunRJyx497C1D5I8LtGl7/s2nIYtHQeQsU5eXwxRcwZQpkZ6vg8htbVlC8Uex22LMHT7ayRjnwv/DDBOXesvtRWHlu44eXHsCFixtG3wDAvpJ9nude/uVl7lxxJwvXL2zF6IXTsb1gO5cPudzzWGNz7maCdcFU11azv2R/8zvWElmEuH0Hf/kdbJwFBd+rfZvbvHnZjyqoG2+wtmaxKK4pxmQzeYK1n1j3BO/sVlUqC4wF7NgBM2bA3r1q4vnee01PxtCZ0QS2w+XwxJXl6nNJTAzM/5eaqgRYfTfQPsoD1pOowkMDwfB2O1x+OTzyCHz1FcyfH9hYu9pamD5dfa7Jyepz3rgxcP23FhEW3YBDh+Dcc9XqXEqKWq3rUOr+2Lpg1okugdOhgvrSZqq856CsFonneIRERqISFmWmMirNlf4tFuZyv5W2i0zuyV9riuJpaCtzJz91N7g87+1vPHqrnkG9BrHprk1sumsT41PHo7fqSUmBikAsbEf0UXEp1cd9mv2dh7oWi7oUGgv9ToYrzZVkZalYEO1mFGi0ycT89fOJ+3scLlwts1i4XZ8wHvVp7tNHrQUcPw5ffqluio39hLX/s6SeLtOExTUjruHWM2/l7L5nY3VY/WY5qrJUtWz11w9bt6rtH/6grClhYTB5ckC69rB2LWRkwFlnwZAhMHduIwdbSmHvAkg4S4n/WVkw8qFG+9fiK2aNmEVoUKhnH2BN1hoiQiLYkLOhW7u9dCQ2h43dRbuZOngq/eP6s6Nwh+e5TbmbmDViFsG6YDblbmp+5+6kGp6EFnXbQ+NVKlpQhfHc7lFZlUq97ijYQY5ezW5zqnI8FosvM79kY46aWeYb8nnjDbUosHkzbNkC+/YFJri5o9EWUO775j5mfTQLUGKjXz+1oNDa+0Nf9yXxeJ1bg8PhvcYdrXOpbMy99fvv1XXoz39WCy6HDqnrRaBcp9asUX0++aSyiGdlwcUXN7ubNkOERRfH5YKbboITJ+DTT9UNb/jwjh5V1+HYMXj7bWWy/Omnjh5NM6gtV6lgtZoSx5fAweegeJ3H9WlF5grPje9oxVHKTGVEhkQyPGk4w5OGExIU4hNjUZfC6jJV46DWPXk+61mI6nfKcfWxO+08u+lZnt30rHei6K4cy5F/wqFFUKVWX8vN5cSGxXL96Ou5fvT1nurbeouePtF9uKD/BVzQ/wLSY9PRW/QMHKgKGNWfwIKqKTHu9XEMf2U4OVX1l5XqEeG+SxR869Psz2JRUlPiUxBPQ7NYxITF8Os9v/LrPb9y1bCrqDBXYLWeGsgcSPxZJ1okLLRA/FLfyVF4uLfy7OzZyp2rMQa6E4UdOODbrgmL12a+xr+v/Td3n3034BVGS65ZwspbV3L/BFWYxJ+lqiVoFoRAXQeX7FrCTZ/dxKcHlDCurVXX3LAwdc3YuvU0VX6LVivL0Lj/hfhRyv1u7IJG33Nf8T5PsoARySM8wiLfkM/h8sPcO/5eDFYDOwp2NNpPYxwqPcTr21/3TEhPR5mpjE0nN/kkK+gyOCxw6HnYcA2suRy2P6jq9DTAgdIDWB1Wzup7Fmf1PctjsaiurWZP0R6mDJrCGb3PYHPu5uaPJcydF1W7tk5601uwM6o/6E+1gmgWiwe+e4CXfnnJ0+YvvWy+MZ+NG5Ub1Lhxqm3YsLa9Jnlo45iDui6BGrmGXNKVRygH6xjFWxLXpgkLbXFCr4dNm7zCYu1atc3ObtxCsGaN+rfnz1eLGyNHwjkBCFfU2LBBbbWaTlFRgbWKt5YeJyysditGqxGbw9Ym/b+39z1uW34bH+zzvRs7HKo6ZGFhYBfwMzPVasQjj6jKt1OmwMMPB67/gBKoC06A+nn7bfVj/OQTdaGYN69lw+kQXO5Ju86d7ibzedj9X5D/jccfeNHWRby/731ABXOXm8uZOmQqh/94mMN/PMyIpBGUm8spqSmhX1w/HjrvIR467yESIhIoMZdC3HBVuAkg5QIIOX1Kpu+Pfs+81fOYt3oeK4+vVI2xbvFjM8Cuhz1VuMtN5YzpPYZPb/iUT2/4lAlpE6gwV6C36omP8BaoiAuPU1aMQWr/l1/U1mz2TiJXZK5gX/E+SmpK+Pfefzc+yNiharv/aTj5mSoOhxIRUaFRLLx0IQsvXcjolNHKFcptsQgNCiU0SJ3vwmplsUiKTGJC2gQmpE1gSK8hVFoqiY6Gqqq2M9T5ExEtcoXSUgcffklV/DV5+xg6tOndaMJCi8NwudSNWRMW2mepbTWhNnPYTKZlTGNS+iQgcHEWWhyOVkG8NTicDhasX8BXh7/iqY1P4XK52LhRuZwuXAgXXaSsxXfc0UgnmmUs/gwwHoctd6g/U8OpQfeW7CUiJIIXtryAy+XyCIu1WWpm89D5DxEREuHZbwlPrHuC+7+9n9u/vB2n6/TBSwvXL+SiJRcx84OZXS+YfMM1sO9JSLkIht4HQSHKktQAmpDQhMWuol3YnXa25W/D4XIwKX0Sk9IntVxYhCd5r62p07zX8ah+ymJR6ZsJLFuffUo32VXZftPL5hvyKSryLg60Kw3VyQiQ4PB37dMsFgCffaa2GzaolfzmMmCA2r74orqGP/qosjgMca+NvfGGite67z41p2uInBzlPXJKbY0A1REpKlILQMnJzX5pu9AlhMXevfDXv8KcOXD33UoNtgSXy8WUpVOI+3scd311V2AHCZhtZuaunMv67PXMXTkXi92CywX/8z/Kd+++++CBB05jNm8mJ0+qbZewUgTKRSoA/djt6js1fTr8+CO88opamegyhCWp9IQm9xdg0lvuWAbf4ngamsVCq18Bqn5EmamM4upiJqVPYtH0RSyavojhScNVQHfsCDAcAsORU/priDd3vsnkQZO5eMDFvLXzLdWouWrVo9xc7qlfoY2n3KwsFlq1ZnALC4ueYcPU/vz56sL6pz8poQ6wdM9SJg+azE1jbmLZnmWNT3wSzlZbWxX8fIPnHBbXFNMvrh8LJi9gweQFXNDvAo/FIi02jdonaql9opbbx93uCd7WKm6DKpRnsBoYOcqB2QxHmn7amoUmIh4890EeOu8hxqeOb5nFotdYVdOjtgK+7A/Z73meGj++6d0MGqTSMS5bplwAnnlG/aaqLFWEBIUQ6U6DrH2mWrIA7dwFurigVk/jFH/oFrAmaw25hlz+ccU/2Feyj52FO8l3LxJrk43T4hajOGtVQoTiNarau82/hUYTEjn6HOb+OJcDpQc4XH4Yq93K2uy1DO41mD7RfZiYNpE1WS27GRZVF7Hi8AoeOf8RsquyWXV8VaPHG61Glu1Zxu3jbudI+REfQXOs4hjL9ixjffb6Fo2lUQKxiGQ4AkU/wpi/wui/QPL5MOLPjVpgNUvQ2NfG8sLWFzDZTGSWZXqExA/HfqDAWEBWVZZfV8nTEjtCWQot9dxTtWvl5ltg31Oe70h2VTbhweGeehU6dGRVZZFvzOeSgZew7vZ1rLt9HX1j+lJQXUB0dOCDhVtFgCbUuYZckiKTeHLykzw5+UkyEjLINeR6LIavvKIWVlsaU3XWWaqORX6+quL9+uuqfdQoFUtmNqvF2x9/bLyf8HAVvN1W+js6WsWCdNbaGu0jLFphCtuyRZmQwsLgn/+E//u/livxr498zda8rbw4/UXe3/u+j99qc3C5VKD03r2wf783aGjJ7iUYrAa+m/MdlZZKlu1ZxubN8MQT8Pe/wzffqEwlixa1bPz+0G6i2dmB67MnYDSqVcf61YK7DMFhKl1h/tdgrYCU8yFYJfI+WnGUSemT+OyGz/jshs9Ii03jSPkRyk3lJEZ4J/JJUUkqK1RNsafSNKh6FsU1xeoGDLD+StjzhFfENECeIY/vjn7HbWfexm3jbuPrw1+rm254IsSP8T04KEyNp46wSIhIUK5QVj1xYb7CwmA1cOFFToKDVSBxaiosXqyeLzQWsvL4SqYOmcrkQZM5WnGUrXlbGx5ozBC/k4qSmhJ6R3ur3qZEp1BmKqPAWOApjAfQJ7qPp0BeQqR3SUqbKI8Yq6KYn39erZ4fPOg1XR87pixkb70FH37Y9FoRdck15BIfHs+LM15k0fRFzB41mwpzBSZbM+NhgiMg+Ty/T02a1PRuIiOV24XNBlddBY8/rtqrLFXEh8ejc/+w4sOVxaLMVEZ0aDShwWrCrZ23QNWyOM/9L/3bbbhyueBw40mYGuSdXe8wOmU0t591O2mxabyz6x2P28WJE03sREvZXLIBEsbBxMZrIhQYC6gwV5AWm8ao5FEMiB+A3WnnUOkh1pxYQ1ZVFlH/G8VPJ39iU+4mVcyymSzZtYTw4HAWTF7AhLQJvLnzTfVEbaUKIj6xVLlX5n8DTgfv73sfs93Ms1c8y4S0Cby6/VVAuT7e+OmNLNqyiBnvzfDJnhQQWrCI9NH+j5iydAqzP5mtfhNW9+Q9MlVt10yBFQMgu2HL5vZCZbEoNZVSXat+z9sLtntcSxduWMi3R5UrZaNWC5cTqvZDwQ+QtwJKNoKtWl1bHSZYdwXsfgwc7kxQvSerrf4g7JuvXo+Ksbhl7C0UPFxAwcMFDE0cqlyhDPmMSRnD5EGTmTxoMgPiB5BvyGfSJLVQpiVUsNm8aaG7CpXmSh758RHmrpzriSXJM+QxKmUU8y+dz/xL53Nuv3PJM+QxZoxavXc61fzKn7tsU4iOhgnuLMB1MxCGhMCFFza9n4kT1bnf7o35b/H5d7lU7EhOjnIFdrm81+dVddYDOlOdjY61WDRhJrd/v1pdvuYaZVZKTm7Z6rzD6eCxNY8xqNcg7E47seGxPL728Wb3c+iQev8nn1TBQmVlyhXJ7rTzj83/IDY8lqc2PkVsWCzPbX4Ou9te1qJJaxNedMYZajz/+IdKL7d7N7z5ZgveqzvgsKjVqYodULZNXdAd/iV9r17KL/rrr70BXx0tzk6eVP7sX3yhMkh884367jfIgBvVTXPlRNh8K1hKcbpcHKs4xvn9zmf26NnMHj2b0SmjOVx2GJvT5muxiEiktKaUKkuV74Q6KkXFGwy8SU0+q0/Agf/xmwaxLkt2LcHpcvL7b3/PH777Aw6XgyW7l3jHWpeBNymLRYSvxUJv1WOwGk6xWLhwERxZzaxZp77v+/vex+ly8re1f2POF3MAZcFoEJ0O+t/g25YwnuLqegIrKsWTN12rXwHQN6YvdqedMlOZj8VCe3zGuSWkpyvx0K+f+p4VF6tFEe0GOHu2ulH5q/J6OnINufSL8wqjVtWyGHCT737cSACmTfMN9hw5svFufvvbU9uqLFU+Lm3a4/qCTHscKItFv35wwQVKuF1xhQrcfqkFJQfKTeUsz1zOwdKDxD4TS4GxgA/2f8DE882kpChXqI0b1eLXkiWNdJRykcpEtnOu8vMv+M7zVKW5kivfv5JZH87i8mWXU1pT6lnw+mj2Rxz8w0FW3qpcClefWE2uIZdrR17LvAvnccdZd2CxWxoX0X5wupy8tfMtTDYT6YvS2VO0hxWZKyg7+R0s7wflv0Lvi9VChf4gLkspr/76Ki6Xi4lvTSSzLJMVmSvIN+Tz4tYX2Veyj79P/TvDk4Zz79f3Nsmtqq3YW7yXu1bcxb3j7+VE5Qnu//Z+XAnjITwZTixR7phX7Wm0D6vdyp6iPTx47oMUzC0gf24+8eHxbM/fzpbcLcwaMYsvbvyCL278gpiwmIaFhc0I34yCPY+pz99hgfLtYDwMQ+5Qx1TuhoPPeAQEfadCeIpPN5bgOAqrCxkQN8DTNiB+AMcrjqO36kmNSfW0p8akkm/MZ/ZsZbGYNk3NDS69FCrbpgZlm2Bz2Ljh0xs4XH6YSksl13x0DSabyVO/RyM9Nl2JDp3Tk4WvtdxU75KopdOubwUJC2u4j1mzlCvmzTerYqn33qsSYTSXL79UFuE33lBz4Q0b1ELVjBlqQefuu+Hpp+Guu1SGqIbIzVXXq0WLVCzup582bbHFalUL59XVzbO+NKvy9u7du4mpFwFkd9g5XnmcfGM+vSJ6UW2tZljSMPrG9PWsVAGws4Gc2w21uxk9Wq2CXX21ulnY7cosdcMNjb4M7BYoXKncHmIG8PXhrzmw5wALLl2Aq9DFHb3v4OX1L7M4bjFnp57dhP9ekZ+vbrjZ2crXW6dTKndj/g9kHcziuWnPkRKVwjlDzmHeqnls6fMP7rvvCubOVTef0FAVINTkOIjTnB9QriGPP65uoqC+dJrqbgxtAcjf5EZzi2w1Lfzcm92/pRy23asqRQ+eo4q7lfwEA/8DUs7zHgMQoSbXjz4KCxYoq4/mC/ntt7QbVqtaZdDSe86dq4Tqgw+qYjxGo/r3Qhr6lTrPB+tVkPUdcALC4imO6IUpx0RYehg73ecmrjyOzcc2gw1qcmrYGa3abXk2yo6r1TzLSYvneEe+g9Jjpfx6IJfgyL/C/ifVjS+iN6RP9/vZOZwOXv3qVab1ncZjFz0GwNM/Pc2/VvyLaVHTCHJMhvx4VRwq8WzMeb2wnLRgSfG+b01ODRRALbXUpNR42vVZeiiATb9s4u67+/Dzz2rlZtAgsNtdvP7961wUcxEvXalmj89vfp73f3yf23vfTnhIA+VY7ddAwb+VMItMhSGzKDzyNmMdYz3vW51TDQWQV57H0EFDPe2mkyYogJLyEhx2h6e9IqsCCmD77s0sWGBk3jzlqhUZqVaroqPVb//dd9VigNmsRIfm4tVUDu87THJUsnc8eWo8azevpaZfHfHXlN+ecxJUjAL9ISUi0/7T8/zllyvh3bevCl5s7Cd75pnq+rx5s3IZGDYMvj2cS6gl1DPOfEM+FEChuZAIXYSnvdBYCAWwZ9ceRtjc8Th2kxqPrmVrXw8/rM796tVqf+LEhsdvsVvYkruFfSX76BfXj3xDPueknUNWVRa1ubWsuGUFvaN6c6LyBHO+mMO/vnmBp56awcKFasIGyk1XC5T1S+x/q8xQRx5R+3EjKPp1N39a/TeSo5OZOWwmq/NWM/7J8Vw26DIoAHu+nZ3lO7E77YQUh/DZ2s+gAO689E7S49Ixx5n59w//5r0f3iNuUlwjbw5YK9X5DIlkS+4Wsg5m8cKMFziz95nU2Gq47pPrePb7j7nZOQq2roETtarWxsnP2DWkjH279/H0ZU8zPnU8FruFmz+/mUeWPsLyzOXMHDqTH3/+kclhk3ntl9dYELqAa0dd25yPq2loH6C9GnI+gdBYFbdiLgJzHsZeE7j1h0cZGzeWVZtWMdYxlk9WfUJ6dTrX934K9v43bEtUKbSdkZBgBaPq02hU1+PkZBXUbsuzkToylcKjys0pw5LByk0r0ZfrmTh4IgPNKrBoRO0IVm1axc4UP18uhw1KBsLJHKharmI6jr8Fw/8IQ26HsD/B4VfUsdGDwHgu7NkP0Y9D5p9Ve68zySkaBAVAIZ7fTFRZFNuyt4FVXce19pCiEE4eO8nIO3Zyyy3w0UdK+CYnq3jMvPqxz/V+FHY71NRAfDyBoQX3f5fLxTM/PcOB3ANcNewqdHod5dnlzHp+Fnk5eYSGeq8pjgIHtjwbqzetZvbsZL78Uv2PQ4eqif8pb9OE8Vx4IYwYoSbeQ4aoc7FzJ5x9tlrI3b8fBg+GxMTG+3/8cWW1fvBBNe+bNKkZ43GTlaViOU6eVO5VVquam2VkqMXtp5+Gv/1NiZizzmq4u1Wr4F//gnvugbg49TkfPKg+a39s26b6v/pqdV2rrYW4uKbnete5mhCFZTAYiA/YN00QBEEQBEEQhK6EXq8nLq7xhYxmWSw2bNhwisWiTTjnHDxVplpyfHUObLoRtBzfiefAuf79g2pqa7jqg6sA6B/Xn6zKLJKiklh+03KCg4JbNp62phXnx2aDmTNVzYvnnlPqNTNTrTBefbWyBj33nFplzc5Wq8Irj61k/rr5PD/9eZ7b9BzjU8ezYHLj6RKbTHUO/HIXpF0F/a+Hiu1gLceccCX/cU9/Bg9WKdWio5VJrkUp1QpXw+56KZ/GPAoDTmf28uXZTc/y8f6PmTJ4CnmGPI6WH+XD2R8yPPk0vnlanjn3Z7BpE/znfypLxahRapVhwwa1qtEVmL9uPuuy1jEkQUWvHqs4xm+G/4ZHL360g0fWAPXOf2fE7rSzPX87NqeN4UnD6RNTpwhGZxq/v5yJgRpXdTb8fCP0OkO5ihRvhLzlMP0XlcXndOM63ThyPlFVr89brFIOl21VMQWD5njik1rDD0d/4PG1j/PxDR8zNHEoR8uPcvNnN/PM1GeYljGt1f23iF3zVLrbyFSV7KEmB859CxKbEZ3fGup8LiabiSnvTuHcfucyJmUMNqeNJbuW8OhFj7Jo6yLOST2Hsb3HYnfaeWfXOzx+yeNcN+q6Nh3ewYO+bnxRUS1LO/5L3i888O0DzLtwHptyN5Gtz+bD2R8SFdo1ikfs3KlWs5OTVZazbduUh4XVqj6+pCQ1J3A4VGIHrShma3n4x4fJLM30fM7fHP2GPtF9eP3q1wPzBl2Ev/9duT39/HMjHgrNoZn3DLNZzQ2DgtRvwJ8XS3V1NZdqJtrT0CyLRVOUSkDQ6Zrn0FX/+P3/A3uf8D3m+koI6+X35Y+teYzFuxaz494dDHtlGM9Pe54HJj7Q/HG3F604P5s3K1PfsmW+F9SdO9V38d134fbbT+3ir6v/yldHviItNo2vb/maiJAAXVlAVSKt2guWEpVGNSwBks/DpQslM1MFyttsSgyN9Z9gqHHWXA7F62DCK9DrTFh9CUx8FYbd36xuiqqLGPLSEF6a8RKLdy2mX1w/Prvxs9O/UPMpc38GN92kLiKHDilT68GDyvQaiPSY7cHKYyuZ8f4M5p43l1pHLf/89Z/8dOdPXDTgoo4emi8N+fJ1tVSZ9b4/3ZZ9/w37FsA12RAUoeJ6jvwTbq71ZlZqiKZeEyt2Qvk25U6jC4aoATBgdkCGf6DkAGe8dgZn9T3LU5NlT/EeMv+QyYjkEQF5j2ZhM8BnCdD7UpiySsWe/XguTN2oYijag3qfy5SlU9iQvYGIkAhsTht2p528h/K4c8WdrD6xmsjQSGwOGzanjewHsxnYa2CbDu/ee1Us1KJFcO21yv2j0XiZRvjLj3/howMfUVpTyk93/sTE9ImBHWwbct99KhnG4cPKzUavVxPc6Oi2fd/Xfn2NB77znWs9O/VZ/nLhX9r2jTsZmZmqoGd4uHKZMplUHN6557awwza4ZzRHBwRCG3U+iv3l9m74BP/5vD/zwtYX+M2HvyEuPI67zg58KtrOglYARktTq5GRoZTq6tVw223qe6nXe30tn5n6DM9MfaZtBhUcDkmnXoR1qBX9VhV+cblUMGLfK2D4H1pVSbpvTF8emPgA/7X6vzBYDSyetbhF/WzapFLWaekqR49u8ZA6hMuHXE5KVArhIeHYnXb6x/Xnwv7NSJnRXnT1iXh9YdTdBUb1CQiOVPU1sj+C4++oIo00EuxV9xw15fwkjm+z1foRySMIDw5nd9FuT1tUaBTDkpoZRBMoyraq2Khh9wMusLQgBVmAmZExgy25Wyj/r3J+9/Xv2Fu8l/S4dKZnTGd99npKHinhge8eYFv+tjYXFaBW5keMgIceUvtvv93yvp6Z+gz3TbiPiJAIn+QKXYFNm9SENiND7beX5/v0odMBlaggNjyWmR/M7DjrXgcycqQqo+ByqXifqKhWWC6ae01sA7qnsNAKz8SNAmtJoxU2AXpH9+axix5j5fGVLLx0YWBX4zsZQ4Yo4fD00ypQODFRpbz817/g44+V29GgQSqYOTVV5aXv0tRkqQrVfaaofe270UIeu/gxzuh9BomRiYzpPeb0L/BDYaFyR+uqhASFcMPoG/j80OeYbCZuOeMW30QNQmDorgKiISJTwWEGczEMuln9nY5OdI5CgkIY03sMeoue3575W5btXUZSZBJBLQxAbzXV7uqRMRlQkw0br+mYcdRhxtAZPLrmUX4++TNrs9YyZ+wcT/sjqx5hY85G1pxYw/Wjr2/zsWgBrHUt9K2xGocEhTA0sRmVJTsR5eUdUwtrSMIQMhIyWJu1ltjwWPrG9OXMPo2Vse/e6HTKPb1VdIJrYucqkFe3AE5Li+GYC5XfbGgvZfI98+kmveyJS5/g57t+5v6JzXOP6YosXarqacTHKx9KLU3b1VcrS8aWLbBiBXz3XeP9dAnM7lW6SLepZs1lreouMTKRO866g1kj/OQ9bSLBwZ23sE1TuWXsLRwpP0KeIY9bzrilo4cjdAe0mKdt96jKw2XbIPPFDh1Sczmzz5kE6YI8MWjj+jSWKqq90EFoPGTco/60eg4dwLi+40iNSeWVba9QVF3E9Ay1Yj2m9xj6xfXj1e2vkm/MZ8bQGW0+FoNBudimpJz+2O5OTEzHpaOdnjGdddnrWJe9jiuGXCGLVN2AzmWxaK7S8mfyKXcHqwy8ESKST++b251pxCQ2fbr68/cSrehet8DlDuAPcluhwhJUPvGg1gdrtpSRI1Vxxa7Mhf0vZOn/W0qwLrhZ6ZoFoUESx8M5L6s4i+/dE/L+18HIP3fosJrDuD7jWLZnGRXmCrIqsxh3XgcKC60IZE0WJJ7dYAKT9mZaxjSW7llKVGgUlwy8xNM+PWM6i3ctJiIkgksHNi1INBDIPFa5QS1fDlVVKqmIw6H+GqvVECimZUzj1e2vokPH3PPmtv0bCm1O57JYNBetEEPdP0upei6hnbJedGb8nZ+eRog7i5nNvRxzXSHcUAkZHRdHM3myCpZfulQFaS1a1PWqoup0Om4bdxtzzpzT0UMRuhMj/gTXFsBV++D/5cPFn3f0iJrFuD7jcLqcfH7wc1y4OtatI+UCQAfH3vQWYCvf7n3cQcwcNpOEiASuGnaVT52Zq4dfTVJkElcNu4rI0Mg2H0dcnMrE19Iqzd2J665TmYGmT1f3oylTlIt0e3DZ4MsIdS8AX5FxRfu8qdCmdM6sUK0h6z3Y8luY/AOkTVcBgL/cDddXqNXq7kBzs0L1ZGzV8GkcDL0XJnVQCrt61qLjx1Uxm+o69Was1vZZHRIEoe2oMFeQ9GwSlw++nDVZa6icV+lTnb3dWX81FHyrsuFF9FbZ8W4yt58lv5Pfq848Uy3qHDqk9p1O/6k2ewIPPgj//Kc6BwkJcPSoSjPbHljtVly4unV8a1enOTqg+/2ErG6LRXTbZ5QQugChMSp4MW85WCtUW/l2MB7vsCFlZKiq7TfeCJdcoi7mIioEoeuTGJlIemw667PXMzB+YMeKCoBJb6p0s1V7VT2LpEkqza4AKBegzEx1Dc7PV2lXeyovvQRHjsDatZCT036iAiA8JFxERTeic8VYBAJrmdqGJ0HNSSj9uWPHI3Q86b+Bwy8ov+34M6B4DUz7pUOHNHq0ysIlCEL3YlzfceQfze8c2W2i0mDqetBnQnAExAzq6BF1Ku69F955B/70J/UXG6vqWvRUMjK8KWcFoaV0P4uFVqcgOEIVQzrRwmo3QvdhzKMqkNGUB4U/qIrsIe1QEbV+ZrOWZjoTBKHLoGWC6hwZodzEjxRR4YdJk1Sqda2S9KyWJ/sTBMFN97NYaFmAgsWsJriJ6A1X7oED/wv2Ghh8K8R1QCVcQRC6PY9f/Dh/nPTHjneDEprE/fcrt1STCfr37+jRCELXp/sJC6cddEE9O82scCrhiTD+H+37np04aFEQhLYhOiya6LDojh6G0AySkto3pkAQujPdT1jodCqdXnec1HWCUu2CIAiCIAiC4I9uKCzc/5LTqgounfu22g+WFSRBEARBEARBaCu6n7DQXKAcFogeABl3d+x4BEEQBEEQBKEH0P2yQoXEqq3D3LHjaAukkrYgCILQ2ambAU+y4QlCj6L7CYuIFLU1F3TsOARBEARBEAShB9ENhUVvta3O6thxCIIgCIIgCEIPovsJi3C3xcJ4zNtmKeuYsQiCIAiCIAhCD6H7CQvNYlG81tt29F8dMxZBEARB6GlIPKAg9Fi6n7CI6qe2RaugYCVkfwg5H3XsmARBEARBEAShm9P90s2GxkH0IKjJhvUzVFvs8I4ckSAIgiAIgiB0e7qfxQIgaVJHj0AQBEEQBEEQehTdU1ikXdnRIxAEQRAEQRCEHkX3FBYDboDogd79lAs7biyCIAiCIAiC0APonsIiJBomvga6IOg1DsY/39EjEgRBEARBEIRuTfcL3tZIuxJml0FIHAQFd/RoBEEQBEEQBKFb0yRh4XLnoDYYDG06mMATDNR09CAEQRAEQRAEoUuizf9dTahJ0yRhYTQaAejfv38rhiUIgiAIgiAIQlfEaDQSHx/f6DE6VxPkh9PppKCggNjYWHQ6XcAGKAiCIAiCIAhC58XlcmE0GklLSyMoqPHw7CYJC0EQBEEQBEEQhMbonlmhBEEQBEEQBEFoV0RYCIIgCIIgCILQakRYCIIgCIIgCILQakRYCIIgCIIgCILQakRYCIIgCIIgCILQakRYCIIgCIIgCILQakRYCIIgCIIgCILQav4/woaK9s+B0PEAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 800x100 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Visualize ISM contribution scores (128 bp window centered on variant)\n",
    "\n",
    "visualize_input_gradient_pair(\n",
    "    pred_ism_wt,\n",
    "    pred_ism_mut,\n",
    "    plot_start=(poses[0] - start) - 64,\n",
    "    plot_end=(poses[0] - start) + 64,\n",
    "    save_figs=False,\n",
    "    fig_name=chrom + '_' + str(poses[0]) + '_prediction_ism_gtex_blood'\n",
    ")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "67b3df59",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "example_ix = 0\n",
      "example_ix = 1\n",
      "CPU times: user 48min 34s, sys: 9min 1s, total: 57min 35s\n",
      "Wall time: 2h 15min\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "#Get contribution scores (ISM shuffle) for blood GTEX tracks\n",
    "# (not feasible to run on CPU; ~135 minutes on GPU)\n",
    "\n",
    "[pred_ism_wt, pred_ism_mut] = get_ism_shuffle(\n",
    "    models,\n",
    "    [sequence_one_hot_wt, sequence_one_hot_mut],\n",
    "    ism_start=(poses[0] - start) - 64,\n",
    "    ism_end=(poses[0] - start) + 64,\n",
    "    prox_bin_start=0,\n",
    "    prox_bin_end=1,\n",
    "    dist_bin_start=0,\n",
    "    dist_bin_end=1,\n",
    "    track_index=[9, 10, 11],\n",
    "    track_scale=0.01,\n",
    "    track_transform=3./4.,\n",
    "    clip_soft=384.,\n",
    "    window_size=5,\n",
    "    n_samples=12,\n",
    "    dist_bin_index=gene_slice.tolist(),\n",
    "    use_mean=True,\n",
    "    use_ratio=False,\n",
    "    use_logodds=False,\n",
    "    untransform_old=True,\n",
    ")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "a7fcc94b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "--- WT ---\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x100 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "--- Mut ---\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x100 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Visualize ISM Shuffle contribution scores (128 bp window centered on variant)\n",
    "\n",
    "visualize_input_gradient_pair(\n",
    "    pred_ism_wt,\n",
    "    pred_ism_mut,\n",
    "    plot_start=(poses[0] - start) - 64,\n",
    "    plot_end=(poses[0] - start) + 64,\n",
    "    save_figs=False,\n",
    "    fig_name=chrom + '_' + str(poses[0]) + '_prediction_ism_shuffle_gtex_blood'\n",
    ")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "17a8b494",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Load samples with reference- and alternate alleles respectively\n",
    "\n",
    "#These files are protected by dbGaP - email to request access\n",
    "\n",
    "cov_files_wt = pd.read_csv(\"gtex_ref_chr10_116952944_T_C.txt\", sep='\\t', names=['file'])['file'].values.tolist()\n",
    "cov_files_mut = pd.read_csv(\"gtex_alt_chr10_116952944_T_C.txt\", sep='\\t', names=['file'])['file'].values.tolist()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "738424ba",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "gtex_targets_wt.shape = (16352, 32)\n",
      "gtex_targets_mut.shape = (16352, 32)\n",
      "-- Counts --\n",
      " - sum_wt = 28239.895\n",
      " - sum_mut = 38557.145\n",
      " - max_y_wt = 241.0508\n",
      " - max_y_mut = 370.9016\n",
      " -- (max_y = 370.9016)\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x200 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Visualize measured coverage tracks for chr10_116952944_T_C\n",
    "\n",
    "save_figs = False\n",
    "save_suffix = '_chr10_116952944_T_C_meas_32_subjects'\n",
    "\n",
    "center_pos = 116952944\n",
    "\n",
    "chrom = 'chr10'\n",
    "poses = [116952944]\n",
    "alts = ['C']\n",
    "\n",
    "start = center_pos - seq_len // 2\n",
    "end = center_pos + seq_len // 2\n",
    "\n",
    "blacklist_bed = \"hg38/blacklist/blacklist_hg38_all.bed\"\n",
    "\n",
    "read_coverage_func_wt, close_coverage_func_wt = get_coverage_reader(cov_files_wt, 16384, 16, blacklist_bed)\n",
    "read_coverage_func_mut, close_coverage_func_mut = get_coverage_reader(cov_files_mut, 16384, 16, blacklist_bed)\n",
    "\n",
    "gtex_targets_wt = read_coverage_func_wt(chrom, start, end, clip_soft=384., clip=768., scale=0.01, transform_old=True)\n",
    "gtex_targets_mut = read_coverage_func_mut(chrom, start, end, clip_soft=384., clip=768., scale=0.01, transform_old=True)\n",
    "\n",
    "close_coverage_func_wt()\n",
    "close_coverage_func_mut()\n",
    "\n",
    "print(\"gtex_targets_wt.shape = \" + str(gtex_targets_wt.shape))\n",
    "print(\"gtex_targets_mut.shape = \" + str(gtex_targets_mut.shape))\n",
    "\n",
    "plot_window = 131072\n",
    "bin_size = 32\n",
    "pad = 16\n",
    "\n",
    "anno_df = None #splice_df\n",
    "\n",
    "untransform_old = True\n",
    "normalize_counts = True\n",
    "\n",
    "#Tracks\n",
    "track_indices = [\n",
    "    np.arange(32).tolist(),\n",
    "]\n",
    "\n",
    "track_names = [\n",
    "    'GTEx Blood (32 subjects)',\n",
    "]\n",
    "\n",
    "track_scales = [\n",
    "    0.01,\n",
    "]\n",
    "\n",
    "track_transforms = [\n",
    "    3./4.,\n",
    "]\n",
    "\n",
    "soft_clips = [\n",
    "    384.,\n",
    "]\n",
    "\n",
    "print(\"-- Counts --\")\n",
    "plot_coverage_track_pair_bins(\n",
    "    gtex_targets_wt[None, None, ...],\n",
    "    gtex_targets_mut[None, None, ...],\n",
    "    chrom,\n",
    "    start,\n",
    "    center_pos,\n",
    "    poses,\n",
    "    track_indices,\n",
    "    track_names,\n",
    "    track_scales,\n",
    "    track_transforms,\n",
    "    soft_clips,\n",
    "    plot_window=plot_window,\n",
    "    normalize_window=4 * plot_window,\n",
    "    bin_size=bin_size,\n",
    "    pad=pad,\n",
    "    normalize_counts=normalize_counts,\n",
    "    save_figs=save_figs,\n",
    "    save_suffix=save_suffix,\n",
    "    gene_slice=gene_slice,\n",
    "    anno_df=anno_df,\n",
    "    untransform_old=untransform_old,\n",
    ")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "671a9c8e",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python [conda env:borzoi_py310_new2]",
   "language": "python",
   "name": "conda-env-borzoi_py310_new2-py"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}