Thesis_Intro_AE _vs_PCA.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "6aac938d-663b-4e88-aff3-7be8d5912d83",
   "metadata": {},
   "source": [
    "# Comparative Analysis of Autoencoder-Based and PCA-Based Dimensionality Reduction Techniques for Gene Expression Data"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c170eebe-a684-4fd4-9583-57695a845bc8",
   "metadata": {},
   "source": [
    "# Introduction "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ed6c389f-5a35-48ba-be97-0f4748cf7358",
   "metadata": {},
   "source": [
    "# Project Overview\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a7cde4d8-7de1-48ae-8252-5d3bdc8776a3",
   "metadata": {},
   "source": [
    "## Loading all pakacges need in the project"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2077a6ab-4d05-40aa-a28c-36d725112089",
   "metadata": {},
   "source": [
    "## Import Libraries\n",
    "\n",
    "The following Python libraries are imported for various purposes:\n",
    "\n",
    "- **Data Manipulation and Visualization Libraries**:\n",
    "    - `os`: Used for interacting with the operating system, primarily for path operations.\n",
    "    - `numpy (np)`: Used for numerical operations.\n",
    "    - `pandas (pd)`: Used for data manipulation and reading CSV files.\n",
    "    - `matplotlib.pyplot (plt)`: Used for plotting graphs.\n",
    "    - `seaborn (sns)`: Used for creating attractive and informative statistical graphics.\n",
    "\n",
    "- **Model Building and Training Libraries**:\n",
    "    - `tensorflow.keras.models.Model`: Used for building and training the autoencoder neural network model.\n",
    "    - `tensorflow.keras.layers`: Contains `Input`, `Dense`, and `Dropout` for building the layers of the neural network model.\n",
    "    - `tensorflow (tf)`: Used for training and inference of deep neural networks.\n",
    "    - `tensorflow.keras.models.load_model`: Used to load a previously saved model.\n",
    "\n",
    "- **Data Preprocessing and Analysis Libraries**:\n",
    "    - `sklearn.model_selection.train_test_split`: Used for splitting data into training and validation sets.\n",
    "    - `scipy.stats`: In this context, it is used to compute the Pearson correlation coefficient between the original and predicted data.\n",
    "    - `sklearn.preprocessing.StandardScaler`: Used to standardize features by removing the mean and scaling to unit variance.\n",
    "    - `sklearn.decomposition.PCA`: Used for performing Principal Component Analysis (PCA).\n",
    "\n",
    "## Package Versions\n",
    "\n",
    "Knowing the versions of the packages is important for reproducibility purposes. Here are the versions of the packages used:\n",
    "\n",
    "- **numpy**: 1.26.1\n",
    "- **pandas**: 1.4.1\n",
    "- **matplotlib**: 3.5.1\n",
    "- **sklearn**: 1.3.2\n",
    "- **tensorflow**: 2.15.0\n",
    "- **scipy**: 1.11.3\n",
    "- **seaborn**: 0.12.2\n",
    "- **matplotlib_venn**: 0.11.10"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "368d4c48-f98d-42d4-a894-05438c9f1ac9",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Importing necessary libraries\n",
    "import os\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.model_selection import train_test_split\n",
    "from tensorflow.keras.models import Model\n",
    "from tensorflow.keras.layers import Input, Dense, Dropout\n",
    "import tensorflow as tf\n",
    "from tensorflow.keras.models import load_model\n",
    "import scipy.stats\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.decomposition import PCA\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "28d2a64a-cbd1-408f-85e2-eebad022444c",
   "metadata": {},
   "source": [
    "# Autoencoder Implementation "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2673f53f-41b3-4fa0-a085-3bbbf839cf8a",
   "metadata": {},
   "source": [
    "## Setup and Initialization\n",
    "\n",
    "Initially, we started setting the directory containing the gene expression data and defining a set of hyperparameters for the autoencoder model. These hyperparameters include the size of the latent space, the architecture of the neural network (such as the number of hidden layers, size of the first hidden layer, dropout rate for regularisation), and training parameters (number of epochs, batch size, and activation function).\n",
    "\n",
    "## Data Loading\n",
    "\n",
    "The load_data function was used for importing gene expression data from a given file path. As part of its operation, it reads the data and displays the initial portion, to have a quick insight into the data structure, to see if the data was loaded correctly.\n",
    "\n",
    "## Model Construction\n",
    "\n",
    "The build_ae function was used for building the model based on the previously defined parameters. This involves setting up the input layer, adding multiple hidden layers with optional dropout for regularisation, and defining the latent space where data is compressed. The model is then compiled with the adam optimiser and mse as the loss function, setting the stage for training.\n",
    "\n",
    "## Training the Model\n",
    "\n",
    "With the model constructed, train_ae function proceeds to divide the gene expression data into training and validation subsets. The model is trained on the training subset, using the validation subset to evaluate its performance throughout the training process. This step is important  for optimising the model's parameters and preventing overfitting.\n",
    "\n",
    "## Evaluation and Visualization\n",
    "\n",
    "After training, we visualise the model to assess the its performance. The plot_loss function generated plots for the training and validation loss over time, providing insights into how well the model is learning. Additionally, plot_reconstruction_error function ploted the reconstruction error, which measures how accurately the autoencoder can reproduce the input data after compression and decompression. This is particularly useful for identifying outliers or anomalies in the data.\n",
    "\n",
    "## Feature Extraction and Transformation\n",
    "\n",
    "The extract_and_multiply_weights function, first it operate in the extraction of weights from the layers of the model. These weights, which are learned during the autoencoder's training phase, are specifically extracted from three distinct layers. The weights transitioning from the input layer to the first hidden layer are captured in weight_1. The weights transitioning from the first hidden layer to the second hidden layer are captured in weight_2. Lastly, the weights transitioning from the second hidden layer to the latent layer are captured in weight_3.\n",
    "\n",
    "The second operation of the function was the computation of the combined transformation matrix. The extracted weights were then combined through matrix multiplication to compute an overall transformation matrix. This matrix multiplication process, which can be viewed as a linear approximation, involved multiplying weights along paths to derive a measure comparable to the loadings of PCA."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "7005dd8c-2f6e-4c42-a734-3b5972004453",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                 TDP93C287  TDP93TDPC014  TDP93TDPC068  TDP93TDPC134  \\\n",
      "ENSG00000000419   8.655586      9.319748      9.910092      8.151097   \n",
      "ENSG00000000457   7.536906      8.075948      8.562040      8.864825   \n",
      "ENSG00000000460   7.173539      6.843101      7.104987      7.161539   \n",
      "ENSG00000000938  10.961783      7.433909     10.440624      9.927085   \n",
      "ENSG00000000971   5.773083      7.706707      6.704198      4.321877   \n",
      "\n",
      "                 TDP94C022  TDP94C035  TDP94C079  TDP94C100  TDP94C158  \\\n",
      "ENSG00000000419   9.329317   8.670321   9.291516   8.831926   9.420048   \n",
      "ENSG00000000457   7.450732   8.457329   7.398729   7.958971   8.038661   \n",
      "ENSG00000000460   8.923964   7.818642   6.657649   7.917220   6.311222   \n",
      "ENSG00000000938   6.807541  10.697348  11.882826   7.289811   8.493930   \n",
      "ENSG00000000971   5.031730   8.032692   3.906962   8.108656   5.925756   \n",
      "\n",
      "                 TDP94C240  ...  TDPPS900181  TDPPS900196  TDPPS910112  \\\n",
      "ENSG00000000419   8.795597  ...     9.342703     9.865924     8.360825   \n",
      "ENSG00000000457   8.355285  ...     7.463325     7.786275     7.926907   \n",
      "ENSG00000000460   6.542077  ...     7.371893     7.263690     9.051097   \n",
      "ENSG00000000938   5.854342  ...    10.094467    10.702120     7.903701   \n",
      "ENSG00000000971   6.912095  ...     9.719061     7.071156     5.267116   \n",
      "\n",
      "                 TDPPS920020  TDPPS920022  TDPPS920025  TDPPS920042  \\\n",
      "ENSG00000000419     9.758897     9.151121     8.736171     9.664389   \n",
      "ENSG00000000457     8.208226     7.466056     8.312369     7.572478   \n",
      "ENSG00000000460     6.617400     7.805494     8.259546     7.389162   \n",
      "ENSG00000000938     7.505892    10.150450     9.301983     9.690991   \n",
      "ENSG00000000971     6.356763     5.823421     5.839385     8.130593   \n",
      "\n",
      "                 TDPPS920197  TDPPS920259  TDPPS930024  \n",
      "ENSG00000000419     8.882510     8.201473     9.313138  \n",
      "ENSG00000000457     8.412820     8.209651     8.196317  \n",
      "ENSG00000000460     6.706902     8.221832     8.053330  \n",
      "ENSG00000000938     9.427635     7.970690     6.204927  \n",
      "ENSG00000000971    10.023701     4.770442     6.490266  \n",
      "\n",
      "[5 rows x 506 columns]\n",
      "Model: \"model_22\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " input_12 (InputLayer)       [(None, 17259)]           0         \n",
      "                                                                 \n",
      " dense_44 (Dense)            (None, 500)               8630000   \n",
      "                                                                 \n",
      " dropout_22 (Dropout)        (None, 500)               0         \n",
      "                                                                 \n",
      " dense_45 (Dense)            (None, 500)               250500    \n",
      "                                                                 \n",
      " dropout_23 (Dropout)        (None, 500)               0         \n",
      "                                                                 \n",
      " dense_46 (Dense)            (None, 85)                42585     \n",
      "                                                                 \n",
      " dense_47 (Dense)            (None, 17259)             1484274   \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 10407359 (39.70 MB)\n",
      "Trainable params: 10407359 (39.70 MB)\n",
      "Non-trainable params: 0 (0.00 Byte)\n",
      "_________________________________________________________________\n",
      "Epoch 1/200\n",
      "5/5 [==============================] - 3s 179ms/step - loss: 73.2352 - val_loss: 38.0638\n",
      "Epoch 2/200\n",
      "5/5 [==============================] - 0s 98ms/step - loss: 32.4049 - val_loss: 27.8747\n",
      "Epoch 3/200\n",
      "5/5 [==============================] - 0s 95ms/step - loss: 26.4786 - val_loss: 24.6275\n",
      "Epoch 4/200\n",
      "5/5 [==============================] - 0s 94ms/step - loss: 23.8079 - val_loss: 22.7741\n",
      "Epoch 5/200\n",
      "5/5 [==============================] - 0s 87ms/step - loss: 22.4517 - val_loss: 21.9885\n",
      "Epoch 6/200\n",
      "5/5 [==============================] - 0s 95ms/step - loss: 21.7746 - val_loss: 21.4641\n",
      "Epoch 7/200\n",
      "5/5 [==============================] - 0s 91ms/step - loss: 21.3575 - val_loss: 21.1562\n",
      "Epoch 8/200\n",
      "5/5 [==============================] - 0s 95ms/step - loss: 21.1157 - val_loss: 20.9658\n",
      "Epoch 9/200\n",
      "5/5 [==============================] - 0s 98ms/step - loss: 20.9549 - val_loss: 20.8521\n",
      "Epoch 10/200\n",
      "5/5 [==============================] - 0s 91ms/step - loss: 20.8492 - val_loss: 20.7905\n",
      "Epoch 11/200\n",
      "5/5 [==============================] - 0s 93ms/step - loss: 20.7876 - val_loss: 20.7361\n",
      "Epoch 12/200\n",
      "5/5 [==============================] - 0s 89ms/step - loss: 20.7616 - val_loss: 20.7233\n",
      "Epoch 13/200\n",
      "5/5 [==============================] - 0s 90ms/step - loss: 20.7166 - val_loss: 20.6532\n",
      "Epoch 14/200\n",
      "5/5 [==============================] - 0s 95ms/step - loss: 20.6842 - val_loss: 20.6611\n",
      "Epoch 15/200\n",
      "5/5 [==============================] - 0s 91ms/step - loss: 20.6729 - val_loss: 20.6210\n",
      "Epoch 16/200\n",
      "5/5 [==============================] - 0s 91ms/step - loss: 20.6435 - val_loss: 20.6240\n",
      "Epoch 17/200\n",
      "5/5 [==============================] - 0s 91ms/step - loss: 20.6570 - val_loss: 20.6449\n",
      "Epoch 18/200\n",
      "5/5 [==============================] - 0s 91ms/step - loss: 20.6403 - val_loss: 20.5741\n",
      "Epoch 19/200\n",
      "5/5 [==============================] - 0s 91ms/step - loss: 20.6080 - val_loss: 20.5966\n",
      "Epoch 20/200\n",
      "5/5 [==============================] - 0s 101ms/step - loss: 20.6131 - val_loss: 20.5704\n",
      "Epoch 21/200\n",
      "5/5 [==============================] - 0s 99ms/step - loss: 20.5997 - val_loss: 20.5862\n",
      "Epoch 22/200\n",
      "5/5 [==============================] - 0s 95ms/step - loss: 20.6122 - val_loss: 20.5765\n",
      "Epoch 23/200\n",
      "5/5 [==============================] - 0s 94ms/step - loss: 20.5957 - val_loss: 20.5690\n",
      "Epoch 24/200\n",
      "5/5 [==============================] - 0s 93ms/step - loss: 20.6012 - val_loss: 20.5824\n",
      "Epoch 25/200\n",
      "5/5 [==============================] - 0s 90ms/step - loss: 20.6006 - val_loss: 20.5668\n",
      "Epoch 26/200\n",
      "5/5 [==============================] - 0s 91ms/step - loss: 20.5913 - val_loss: 20.5686\n",
      "Epoch 27/200\n",
      "5/5 [==============================] - 0s 94ms/step - loss: 20.6037 - val_loss: 20.6055\n",
      "Epoch 28/200\n",
      "5/5 [==============================] - 0s 91ms/step - loss: 20.6465 - val_loss: 20.5876\n",
      "Epoch 29/200\n",
      "5/5 [==============================] - 0s 91ms/step - loss: 20.5825 - val_loss: 20.5726\n",
      "Epoch 30/200\n",
      "5/5 [==============================] - 0s 95ms/step - loss: 20.5764 - val_loss: 20.5478\n",
      "Epoch 31/200\n",
      "5/5 [==============================] - 0s 91ms/step - loss: 20.5641 - val_loss: 20.5378\n",
      "Epoch 32/200\n",
      "5/5 [==============================] - 0s 94ms/step - loss: 20.5568 - val_loss: 20.5308\n",
      "Epoch 33/200\n",
      "5/5 [==============================] - 0s 91ms/step - loss: 20.5540 - val_loss: 20.5296\n",
      "Epoch 34/200\n",
      "5/5 [==============================] - 0s 90ms/step - loss: 20.5511 - val_loss: 20.5264\n",
      "Epoch 35/200\n",
      "5/5 [==============================] - 0s 92ms/step - loss: 20.5464 - val_loss: 20.5142\n",
      "Epoch 36/200\n",
      "5/5 [==============================] - 0s 86ms/step - loss: 20.5412 - val_loss: 20.5165\n",
      "Epoch 37/200\n",
      "5/5 [==============================] - 0s 91ms/step - loss: 20.5409 - val_loss: 20.5137\n",
      "Epoch 38/200\n",
      "5/5 [==============================] - 0s 87ms/step - loss: 20.5396 - val_loss: 20.5103\n",
      "Epoch 39/200\n",
      "5/5 [==============================] - 0s 99ms/step - loss: 20.5354 - val_loss: 20.5088\n",
      "Epoch 40/200\n",
      "5/5 [==============================] - 0s 92ms/step - loss: 20.5320 - val_loss: 20.5071\n",
      "Epoch 41/200\n",
      "5/5 [==============================] - 0s 93ms/step - loss: 20.5351 - val_loss: 20.5029\n",
      "Epoch 42/200\n",
      "5/5 [==============================] - 0s 101ms/step - loss: 20.5279 - val_loss: 20.5059\n",
      "Epoch 43/200\n",
      "5/5 [==============================] - 0s 103ms/step - loss: 20.5334 - val_loss: 20.5062\n",
      "Epoch 44/200\n",
      "5/5 [==============================] - 0s 91ms/step - loss: 20.5333 - val_loss: 20.4981\n",
      "Epoch 45/200\n",
      "5/5 [==============================] - 0s 97ms/step - loss: 20.5172 - val_loss: 20.4853\n",
      "Epoch 46/200\n",
      "5/5 [==============================] - 0s 100ms/step - loss: 20.5112 - val_loss: 20.4848\n",
      "Epoch 47/200\n",
      "5/5 [==============================] - 0s 95ms/step - loss: 20.5109 - val_loss: 20.4845\n",
      "Epoch 48/200\n",
      "5/5 [==============================] - 0s 93ms/step - loss: 20.5107 - val_loss: 20.4838\n",
      "Epoch 49/200\n",
      "5/5 [==============================] - 0s 87ms/step - loss: 20.5101 - val_loss: 20.4839\n",
      "Epoch 50/200\n",
      "5/5 [==============================] - 0s 102ms/step - loss: 20.5087 - val_loss: 20.4824\n",
      "Epoch 51/200\n",
      "5/5 [==============================] - 0s 92ms/step - loss: 20.5071 - val_loss: 20.4808\n",
      "Epoch 52/200\n",
      "5/5 [==============================] - 0s 94ms/step - loss: 20.5074 - val_loss: 20.4811\n",
      "Epoch 53/200\n",
      "5/5 [==============================] - 0s 92ms/step - loss: 20.5088 - val_loss: 20.4909\n",
      "Epoch 54/200\n",
      "5/5 [==============================] - 0s 88ms/step - loss: 20.5469 - val_loss: 20.6249\n",
      "Epoch 55/200\n",
      "5/5 [==============================] - 0s 93ms/step - loss: 20.6269 - val_loss: 20.4418\n",
      "Epoch 56/200\n",
      "5/5 [==============================] - 0s 93ms/step - loss: 20.5015 - val_loss: 20.4403\n",
      "Epoch 57/200\n",
      "5/5 [==============================] - 0s 90ms/step - loss: 20.4715 - val_loss: 20.4518\n",
      "Epoch 58/200\n",
      "5/5 [==============================] - 0s 91ms/step - loss: 20.4613 - val_loss: 20.4219\n",
      "Epoch 59/200\n",
      "5/5 [==============================] - 0s 91ms/step - loss: 20.4510 - val_loss: 20.4256\n",
      "Epoch 60/200\n",
      "5/5 [==============================] - 0s 91ms/step - loss: 20.4489 - val_loss: 20.4191\n",
      "Epoch 61/200\n",
      "5/5 [==============================] - 0s 91ms/step - loss: 20.4422 - val_loss: 20.4140\n",
      "Epoch 62/200\n",
      "5/5 [==============================] - 0s 91ms/step - loss: 20.4380 - val_loss: 20.4100\n",
      "Epoch 63/200\n",
      "5/5 [==============================] - 0s 87ms/step - loss: 20.4340 - val_loss: 20.4061\n",
      "Epoch 64/200\n",
      "5/5 [==============================] - 0s 98ms/step - loss: 20.4311 - val_loss: 20.4032\n",
      "Epoch 65/200\n",
      "5/5 [==============================] - 0s 92ms/step - loss: 20.4208 - val_loss: 20.3896\n",
      "Epoch 66/200\n",
      "5/5 [==============================] - 0s 87ms/step - loss: 20.4165 - val_loss: 20.3892\n",
      "Epoch 67/200\n",
      "5/5 [==============================] - 0s 91ms/step - loss: 20.4141 - val_loss: 20.3876\n",
      "Epoch 68/200\n",
      "5/5 [==============================] - 0s 87ms/step - loss: 20.4088 - val_loss: 20.3812\n",
      "Epoch 69/200\n",
      "5/5 [==============================] - 0s 91ms/step - loss: 20.4067 - val_loss: 20.3784\n",
      "Epoch 70/200\n",
      "5/5 [==============================] - 0s 94ms/step - loss: 20.4001 - val_loss: 20.3655\n",
      "Epoch 71/200\n",
      "5/5 [==============================] - 0s 94ms/step - loss: 20.3922 - val_loss: 20.3649\n",
      "Epoch 72/200\n",
      "5/5 [==============================] - 0s 95ms/step - loss: 20.3867 - val_loss: 20.3530\n",
      "Epoch 73/200\n",
      "5/5 [==============================] - 0s 94ms/step - loss: 20.3797 - val_loss: 20.3533\n",
      "Epoch 74/200\n",
      "5/5 [==============================] - 0s 90ms/step - loss: 20.3659 - val_loss: 20.3353\n",
      "Epoch 75/200\n",
      "5/5 [==============================] - 0s 85ms/step - loss: 20.3598 - val_loss: 20.3269\n",
      "Epoch 76/200\n",
      "5/5 [==============================] - 0s 91ms/step - loss: 20.3514 - val_loss: 20.3244\n",
      "Epoch 77/200\n",
      "5/5 [==============================] - 0s 91ms/step - loss: 20.3473 - val_loss: 20.3143\n",
      "Epoch 78/200\n",
      "5/5 [==============================] - 0s 90ms/step - loss: 20.3418 - val_loss: 20.3139\n",
      "Epoch 79/200\n",
      "5/5 [==============================] - 0s 91ms/step - loss: 20.3321 - val_loss: 20.3016\n",
      "Epoch 80/200\n",
      "5/5 [==============================] - 0s 95ms/step - loss: 20.3242 - val_loss: 20.2947\n",
      "Epoch 81/200\n",
      "5/5 [==============================] - 0s 91ms/step - loss: 20.3182 - val_loss: 20.2868\n",
      "Epoch 82/200\n",
      "5/5 [==============================] - 0s 99ms/step - loss: 20.3068 - val_loss: 20.2707\n",
      "Epoch 83/200\n",
      "5/5 [==============================] - 0s 91ms/step - loss: 20.2904 - val_loss: 20.2548\n",
      "Epoch 84/200\n",
      "5/5 [==============================] - 0s 91ms/step - loss: 20.2804 - val_loss: 20.2365\n",
      "Epoch 85/200\n",
      "5/5 [==============================] - 0s 90ms/step - loss: 20.2484 - val_loss: 20.2032\n",
      "Epoch 86/200\n",
      "5/5 [==============================] - 0s 95ms/step - loss: 20.2193 - val_loss: 20.1863\n",
      "Epoch 87/200\n",
      "5/5 [==============================] - 0s 82ms/step - loss: 20.2097 - val_loss: 20.1757\n",
      "Epoch 88/200\n",
      "5/5 [==============================] - 0s 87ms/step - loss: 20.1961 - val_loss: 20.1630\n",
      "Epoch 89/200\n",
      "5/5 [==============================] - 0s 87ms/step - loss: 20.1844 - val_loss: 20.1491\n",
      "Epoch 90/200\n",
      "5/5 [==============================] - 0s 84ms/step - loss: 20.1697 - val_loss: 20.0940\n",
      "Epoch 91/200\n",
      "5/5 [==============================] - 0s 94ms/step - loss: 20.0940 - val_loss: 20.0529\n",
      "Epoch 92/200\n",
      "5/5 [==============================] - 0s 95ms/step - loss: 20.0807 - val_loss: 20.0419\n",
      "Epoch 93/200\n",
      "5/5 [==============================] - 0s 95ms/step - loss: 20.0521 - val_loss: 20.0093\n",
      "Epoch 94/200\n",
      "5/5 [==============================] - 0s 95ms/step - loss: 20.0327 - val_loss: 19.9997\n",
      "Epoch 95/200\n",
      "5/5 [==============================] - 0s 94ms/step - loss: 20.0137 - val_loss: 19.9772\n",
      "Epoch 96/200\n",
      "5/5 [==============================] - 0s 91ms/step - loss: 19.9981 - val_loss: 19.9618\n",
      "Epoch 97/200\n",
      "5/5 [==============================] - 0s 94ms/step - loss: 19.9784 - val_loss: 19.9493\n",
      "Epoch 98/200\n",
      "5/5 [==============================] - 0s 91ms/step - loss: 19.9580 - val_loss: 19.9010\n",
      "Epoch 99/200\n",
      "5/5 [==============================] - 0s 91ms/step - loss: 19.9069 - val_loss: 19.8612\n",
      "Epoch 100/200\n",
      "5/5 [==============================] - 0s 83ms/step - loss: 19.8714 - val_loss: 19.7966\n",
      "Epoch 101/200\n",
      "5/5 [==============================] - 0s 93ms/step - loss: 19.8127 - val_loss: 19.7724\n",
      "Epoch 102/200\n",
      "5/5 [==============================] - 0s 94ms/step - loss: 19.8005 - val_loss: 19.8060\n",
      "Epoch 103/200\n",
      "5/5 [==============================] - 0s 94ms/step - loss: 19.8280 - val_loss: 19.6887\n",
      "Epoch 104/200\n",
      "5/5 [==============================] - 0s 97ms/step - loss: 19.6586 - val_loss: 19.6189\n",
      "Epoch 105/200\n",
      "5/5 [==============================] - 0s 101ms/step - loss: 19.6009 - val_loss: 19.5229\n",
      "Epoch 106/200\n",
      "5/5 [==============================] - 0s 95ms/step - loss: 19.5221 - val_loss: 19.4637\n",
      "Epoch 107/200\n",
      "5/5 [==============================] - 0s 97ms/step - loss: 19.4605 - val_loss: 19.3864\n",
      "Epoch 108/200\n",
      "5/5 [==============================] - 0s 95ms/step - loss: 19.3904 - val_loss: 19.3371\n",
      "Epoch 109/200\n",
      "5/5 [==============================] - 1s 142ms/step - loss: 19.3386 - val_loss: 19.2647\n",
      "Epoch 110/200\n",
      "5/5 [==============================] - 0s 95ms/step - loss: 19.2782 - val_loss: 19.2064\n",
      "Epoch 111/200\n",
      "5/5 [==============================] - 0s 94ms/step - loss: 19.2078 - val_loss: 19.1289\n",
      "Epoch 112/200\n",
      "5/5 [==============================] - 0s 87ms/step - loss: 19.0838 - val_loss: 19.0119\n",
      "Epoch 113/200\n",
      "5/5 [==============================] - 0s 90ms/step - loss: 19.0223 - val_loss: 18.9282\n",
      "Epoch 114/200\n",
      "5/5 [==============================] - 0s 88ms/step - loss: 18.9334 - val_loss: 18.8716\n",
      "Epoch 115/200\n",
      "5/5 [==============================] - 0s 94ms/step - loss: 18.8508 - val_loss: 18.7643\n",
      "Epoch 116/200\n",
      "5/5 [==============================] - 0s 90ms/step - loss: 18.7260 - val_loss: 18.6278\n",
      "Epoch 117/200\n",
      "5/5 [==============================] - 0s 91ms/step - loss: 18.6339 - val_loss: 18.5700\n",
      "Epoch 118/200\n",
      "5/5 [==============================] - 0s 89ms/step - loss: 18.5425 - val_loss: 18.4882\n",
      "Epoch 119/200\n",
      "5/5 [==============================] - 0s 92ms/step - loss: 18.4814 - val_loss: 18.3809\n",
      "Epoch 120/200\n",
      "5/5 [==============================] - 0s 91ms/step - loss: 18.4130 - val_loss: 18.3533\n",
      "Epoch 121/200\n",
      "5/5 [==============================] - 0s 89ms/step - loss: 18.2283 - val_loss: 18.0161\n",
      "Epoch 122/200\n",
      "5/5 [==============================] - 0s 87ms/step - loss: 18.0086 - val_loss: 17.9327\n",
      "Epoch 123/200\n",
      "5/5 [==============================] - 0s 91ms/step - loss: 17.8159 - val_loss: 17.6251\n",
      "Epoch 124/200\n",
      "5/5 [==============================] - 0s 92ms/step - loss: 17.6093 - val_loss: 17.4809\n",
      "Epoch 125/200\n",
      "5/5 [==============================] - 0s 89ms/step - loss: 17.4569 - val_loss: 17.3545\n",
      "Epoch 126/200\n",
      "5/5 [==============================] - 0s 98ms/step - loss: 17.3818 - val_loss: 17.6464\n",
      "Epoch 127/200\n",
      "5/5 [==============================] - 0s 91ms/step - loss: 17.4117 - val_loss: 16.8501\n",
      "Epoch 128/200\n",
      "5/5 [==============================] - 0s 94ms/step - loss: 16.6627 - val_loss: 16.5254\n",
      "Epoch 129/200\n",
      "5/5 [==============================] - 0s 80ms/step - loss: 16.4244 - val_loss: 16.2712\n",
      "Epoch 130/200\n",
      "5/5 [==============================] - 0s 90ms/step - loss: 16.2372 - val_loss: 16.1369\n",
      "Epoch 131/200\n",
      "5/5 [==============================] - 0s 91ms/step - loss: 16.0789 - val_loss: 15.9805\n",
      "Epoch 132/200\n",
      "5/5 [==============================] - 0s 95ms/step - loss: 15.9675 - val_loss: 15.8785\n",
      "Epoch 133/200\n",
      "5/5 [==============================] - 0s 89ms/step - loss: 15.8513 - val_loss: 15.7740\n",
      "Epoch 134/200\n",
      "5/5 [==============================] - 0s 95ms/step - loss: 15.7227 - val_loss: 15.6189\n",
      "Epoch 135/200\n",
      "5/5 [==============================] - 0s 84ms/step - loss: 15.5817 - val_loss: 15.5082\n",
      "Epoch 136/200\n",
      "5/5 [==============================] - 0s 90ms/step - loss: 15.4791 - val_loss: 15.4075\n",
      "Epoch 137/200\n",
      "5/5 [==============================] - 0s 89ms/step - loss: 15.3787 - val_loss: 15.2902\n",
      "Epoch 138/200\n",
      "5/5 [==============================] - 0s 98ms/step - loss: 15.2472 - val_loss: 15.1604\n",
      "Epoch 139/200\n",
      "5/5 [==============================] - 0s 90ms/step - loss: 15.0920 - val_loss: 14.9878\n",
      "Epoch 140/200\n",
      "5/5 [==============================] - 0s 90ms/step - loss: 14.9367 - val_loss: 14.8258\n",
      "Epoch 141/200\n",
      "5/5 [==============================] - 0s 90ms/step - loss: 14.7727 - val_loss: 14.6767\n",
      "Epoch 142/200\n",
      "5/5 [==============================] - 0s 90ms/step - loss: 14.6445 - val_loss: 14.5636\n",
      "Epoch 143/200\n",
      "5/5 [==============================] - 0s 88ms/step - loss: 14.5205 - val_loss: 14.4295\n",
      "Epoch 144/200\n",
      "5/5 [==============================] - 0s 94ms/step - loss: 14.3543 - val_loss: 14.2470\n",
      "Epoch 145/200\n",
      "5/5 [==============================] - 0s 87ms/step - loss: 14.2155 - val_loss: 14.1810\n",
      "Epoch 146/200\n",
      "5/5 [==============================] - 0s 91ms/step - loss: 14.1532 - val_loss: 14.1187\n",
      "Epoch 147/200\n",
      "5/5 [==============================] - 0s 91ms/step - loss: 13.9812 - val_loss: 13.6725\n",
      "Epoch 148/200\n",
      "5/5 [==============================] - 0s 90ms/step - loss: 13.6953 - val_loss: 13.5032\n",
      "Epoch 149/200\n",
      "5/5 [==============================] - 0s 91ms/step - loss: 13.4704 - val_loss: 13.3916\n",
      "Epoch 150/200\n",
      "5/5 [==============================] - 0s 95ms/step - loss: 13.3544 - val_loss: 13.2840\n",
      "Epoch 151/200\n",
      "5/5 [==============================] - 0s 95ms/step - loss: 13.2461 - val_loss: 13.1705\n",
      "Epoch 152/200\n",
      "5/5 [==============================] - 0s 93ms/step - loss: 13.1463 - val_loss: 13.1076\n",
      "Epoch 153/200\n",
      "5/5 [==============================] - 0s 86ms/step - loss: 13.0783 - val_loss: 13.0428\n",
      "Epoch 154/200\n",
      "5/5 [==============================] - 0s 91ms/step - loss: 13.0228 - val_loss: 12.9918\n",
      "Epoch 155/200\n",
      "5/5 [==============================] - 0s 89ms/step - loss: 12.9648 - val_loss: 12.9141\n",
      "Epoch 156/200\n",
      "5/5 [==============================] - 0s 90ms/step - loss: 12.8966 - val_loss: 12.8727\n",
      "Epoch 157/200\n",
      "5/5 [==============================] - 0s 90ms/step - loss: 12.8511 - val_loss: 12.8199\n",
      "Epoch 158/200\n",
      "5/5 [==============================] - 0s 91ms/step - loss: 12.7975 - val_loss: 12.7467\n",
      "Epoch 159/200\n",
      "5/5 [==============================] - 0s 88ms/step - loss: 12.7056 - val_loss: 12.6528\n",
      "Epoch 160/200\n",
      "5/5 [==============================] - 0s 91ms/step - loss: 12.6281 - val_loss: 12.5733\n",
      "Epoch 161/200\n",
      "5/5 [==============================] - 0s 95ms/step - loss: 12.5327 - val_loss: 12.4805\n",
      "Epoch 162/200\n",
      "5/5 [==============================] - 0s 90ms/step - loss: 12.4622 - val_loss: 12.4585\n",
      "Epoch 163/200\n",
      "5/5 [==============================] - 0s 87ms/step - loss: 12.3981 - val_loss: 12.2891\n",
      "Epoch 164/200\n",
      "5/5 [==============================] - 0s 90ms/step - loss: 12.2729 - val_loss: 12.2202\n",
      "Epoch 165/200\n",
      "5/5 [==============================] - 0s 90ms/step - loss: 12.1975 - val_loss: 12.1658\n",
      "Epoch 166/200\n",
      "5/5 [==============================] - 0s 91ms/step - loss: 12.0886 - val_loss: 12.0252\n",
      "Epoch 167/200\n",
      "5/5 [==============================] - 0s 91ms/step - loss: 12.0176 - val_loss: 11.9853\n",
      "Epoch 168/200\n",
      "5/5 [==============================] - 0s 83ms/step - loss: 11.9263 - val_loss: 11.8517\n",
      "Epoch 169/200\n",
      "5/5 [==============================] - 0s 87ms/step - loss: 11.8395 - val_loss: 11.8183\n",
      "Epoch 170/200\n",
      "5/5 [==============================] - 0s 86ms/step - loss: 11.7495 - val_loss: 11.6137\n",
      "Epoch 171/200\n",
      "5/5 [==============================] - 0s 88ms/step - loss: 11.5911 - val_loss: 11.5198\n",
      "Epoch 172/200\n",
      "5/5 [==============================] - 0s 98ms/step - loss: 11.4522 - val_loss: 11.3824\n",
      "Epoch 173/200\n",
      "5/5 [==============================] - 0s 87ms/step - loss: 11.3339 - val_loss: 11.3006\n",
      "Epoch 174/200\n",
      "5/5 [==============================] - 0s 91ms/step - loss: 11.2629 - val_loss: 11.1754\n",
      "Epoch 175/200\n",
      "5/5 [==============================] - 0s 88ms/step - loss: 11.1247 - val_loss: 11.0753\n",
      "Epoch 176/200\n",
      "5/5 [==============================] - 0s 91ms/step - loss: 11.0462 - val_loss: 11.0030\n",
      "Epoch 177/200\n",
      "5/5 [==============================] - 0s 90ms/step - loss: 10.9845 - val_loss: 10.9463\n",
      "Epoch 178/200\n",
      "5/5 [==============================] - 0s 88ms/step - loss: 10.8984 - val_loss: 10.8400\n",
      "Epoch 179/200\n",
      "5/5 [==============================] - 0s 91ms/step - loss: 10.8281 - val_loss: 10.8150\n",
      "Epoch 180/200\n",
      "5/5 [==============================] - 0s 85ms/step - loss: 10.7749 - val_loss: 10.7165\n",
      "Epoch 181/200\n",
      "5/5 [==============================] - 0s 87ms/step - loss: 10.7004 - val_loss: 10.6846\n",
      "Epoch 182/200\n",
      "5/5 [==============================] - 0s 90ms/step - loss: 10.6428 - val_loss: 10.5888\n",
      "Epoch 183/200\n",
      "5/5 [==============================] - 0s 87ms/step - loss: 10.5633 - val_loss: 10.5379\n",
      "Epoch 184/200\n",
      "5/5 [==============================] - 0s 91ms/step - loss: 10.5213 - val_loss: 10.4951\n",
      "Epoch 185/200\n",
      "5/5 [==============================] - 0s 91ms/step - loss: 10.4712 - val_loss: 10.4421\n",
      "Epoch 186/200\n",
      "5/5 [==============================] - 0s 87ms/step - loss: 10.4138 - val_loss: 10.3850\n",
      "Epoch 187/200\n",
      "5/5 [==============================] - 0s 91ms/step - loss: 10.3661 - val_loss: 10.3575\n",
      "Epoch 188/200\n",
      "5/5 [==============================] - 0s 87ms/step - loss: 10.3642 - val_loss: 10.4812\n",
      "Epoch 189/200\n",
      "5/5 [==============================] - 0s 84ms/step - loss: 10.5298 - val_loss: 10.3479\n",
      "Epoch 190/200\n",
      "5/5 [==============================] - 0s 87ms/step - loss: 10.3330 - val_loss: 10.2870\n",
      "Epoch 191/200\n",
      "5/5 [==============================] - 0s 95ms/step - loss: 10.2678 - val_loss: 10.2550\n",
      "Epoch 192/200\n",
      "5/5 [==============================] - 0s 91ms/step - loss: 10.2267 - val_loss: 10.2018\n",
      "Epoch 193/200\n",
      "5/5 [==============================] - 0s 87ms/step - loss: 10.1898 - val_loss: 10.1864\n",
      "Epoch 194/200\n",
      "5/5 [==============================] - 0s 90ms/step - loss: 10.1659 - val_loss: 10.1611\n",
      "Epoch 195/200\n",
      "5/5 [==============================] - 0s 95ms/step - loss: 10.1484 - val_loss: 10.1502\n",
      "Epoch 196/200\n",
      "5/5 [==============================] - 0s 98ms/step - loss: 10.1383 - val_loss: 10.1436\n",
      "Epoch 197/200\n",
      "5/5 [==============================] - 0s 83ms/step - loss: 10.1300 - val_loss: 10.1246\n",
      "Epoch 198/200\n",
      "5/5 [==============================] - 0s 87ms/step - loss: 10.1102 - val_loss: 10.1086\n",
      "Epoch 199/200\n",
      "5/5 [==============================] - 0s 87ms/step - loss: 10.0939 - val_loss: 10.0922\n",
      "Epoch 200/200\n",
      "5/5 [==============================] - 0s 91ms/step - loss: 10.0744 - val_loss: 10.0758\n",
      "16/16 [==============================] - 0s 8ms/step\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "4/4 [==============================] - 0s 11ms/step\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Combined Transformation Matrix Shape: (17259, 85)\n"
     ]
    }
   ],
   "source": [
    "# Define the directory where the data is stored\n",
    "DATA_DIR = 'C:/Users/ABDULLAHI HAJI/OneDrive/Documents/Thesis_poject/GSE216738_RAW_COUNTS_Abnormal-AML-50bp'\n",
    "\n",
    "# Define the hyperparameters for the autoencoder model\n",
    "params = {\n",
    "    'latent_dim':85,  # The dimensionality of the latent space\n",
    "    'depth': 2,  # The number of hidden layers in the encoder and decoder\n",
    "    'first_layer_size': 500,  # The size of the first hidden layer\n",
    "    'dropout_rate': 0,  # The dropout rate used after each layer\n",
    "    'epochs': 200,  # The number of epochs to train for\n",
    "    'batch_size': 80,  # The batch size used in training\n",
    "    'activation': 'relu',  # The activation function used in each layer\n",
    "}\n",
    "\n",
    "def load_data(filepath):\n",
    "    \"\"\"Load the data from a CSV file.\"\"\"\n",
    "    # Read the CSV file at the given filepath and return it as a pandas DataFrame\n",
    "    data = pd.read_csv(filepath, index_col=0)\n",
    "    print(data.head())  # Print the first few rows of the DataFrame\n",
    "    return data\n",
    "\n",
    "def build_ae(original_dim, params):\n",
    "    \"\"\"Build the autoencoder model.\"\"\"\n",
    "    # Define the input layer\n",
    "    input_layer = Input(shape=(original_dim,))\n",
    "    \n",
    "    # Define the first hidden layer and apply dropout\n",
    "    encoded = Dense(params['first_layer_size'], activation=params['activation'])(input_layer)\n",
    "    encoded = Dropout(params['dropout_rate'])(encoded)\n",
    "    \n",
    "    # Add additional hidden layers with dropout\n",
    "    for _ in range(params['depth'] - 1):\n",
    "        encoded = Dense(params['first_layer_size'], activation=params['activation'])(encoded)\n",
    "        encoded = Dropout(params['dropout_rate'])(encoded)\n",
    "    \n",
    "    # Define the latent space layer and the output layer\n",
    "    encoded = Dense(params['latent_dim'], activation=params['activation'])(encoded)\n",
    "    decoded = Dense(original_dim, activation=params['activation'])(encoded)\n",
    "    \n",
    "    # Define the autoencoder model and compile it\n",
    "    autoencoder = Model(input_layer, decoded)\n",
    "    autoencoder.compile(optimizer='adam', loss='mse')\n",
    "    \n",
    "    # Print the summary of the model\n",
    "    autoencoder.summary()\n",
    "    return autoencoder\n",
    "\n",
    "def train_ae(autoencoder, X_train, X_val, params):\n",
    "    \"\"\"Train the autoencoder model.\"\"\"\n",
    "    # Train the model using the training data (75% of the data) and validate it using the validation data(25% of the data)\n",
    "    history = autoencoder.fit(X_train, X_train,\n",
    "                              epochs=params['epochs'],\n",
    "                              batch_size=params['batch_size'],\n",
    "                              shuffle=True,\n",
    "                              validation_data=(X_val, X_val))\n",
    "    return autoencoder, history\n",
    "\n",
    "def plot_loss(history):\n",
    "    \"\"\"Plot the training and validation loss.\"\"\"\n",
    "    # Create a new figure\n",
    "    plt.figure()\n",
    "    \n",
    "    # Plot the training loss and validation loss\n",
    "    plt.plot(history.history['loss'], label='Train Loss')\n",
    "    plt.plot(history.history['val_loss'], label='Validation Loss')\n",
    "    \n",
    "    # Add a title and labels to the plot\n",
    "    plt.title('Loss over epochs')\n",
    "    plt.ylabel('Loss')\n",
    "    plt.xlabel('Epoch')\n",
    "    plt.legend()\n",
    "    #save plot\n",
    "    plt.savefig('training loss and validation loss plot.png')\n",
    "    # Display the plot\n",
    "    plt.show()\n",
    "\n",
    "def plot_reconstruction_error(autoencoder, X_test):\n",
    "    \"\"\"Plot the reconstruction error on the test set.\"\"\"\n",
    "    # Use the autoencoder to reconstruct the test data\n",
    "    reconstructions = autoencoder.predict(X_test)\n",
    "    \n",
    "    # Calculate the reconstruction error\n",
    "    reconstruction_error = np.mean(np.abs(reconstructions - X_test), axis=1)\n",
    "    \n",
    "    # Create a new figure\n",
    "    plt.figure()\n",
    "    \n",
    "    # Plot a histogram of the reconstruction error\n",
    "    plt.hist(reconstruction_error, bins=50)\n",
    "    \n",
    "    # Add a title and labels to the plot\n",
    "    plt.title('Reconstruction error histogram')\n",
    "    plt.xlabel('Reconstruction error')\n",
    "    plt.ylabel('Number of examples')\n",
    "    \n",
    "    # Display the plot\n",
    "    plt.show()\n",
    "def extract_and_multiply_weights(autoencoder):\n",
    "    \"\"\"Extracts weights from the trained autoencoder and computes the combined transformation matrix.\"\"\"\n",
    "    # Extract weights\n",
    "    weight_1 = autoencoder.layers[1].get_weights()[0]  # Input to First Hidden Layer\n",
    "    weight_2 = autoencoder.layers[3].get_weights()[0]  # First Hidden Layer to Second Hidden Layer\n",
    "    weight_3 = autoencoder.layers[5].get_weights()[0]  # Second Hidden Layer to Latent Layer\n",
    "    \n",
    "    # Perform matrix multiplication\n",
    "    combined_transformation = np.dot(np.dot(weight_1, weight_2), weight_3)\n",
    "    return combined_transformation\n",
    "def main():\n",
    "    \"\"\"Main function to run the pipeline.\"\"\"\n",
    "    # Define the filepath of the data\n",
    "    filepath = os.path.join(DATA_DIR, 'normCounts_res.CSV')\n",
    "    \n",
    "    # Load the data\n",
    "    data = load_data(filepath)\n",
    "    \n",
    "    # Get the original dimension of the data (number of genes)\n",
    "    original_dim = data.shape[0]\n",
    "    \n",
    "    # Transpose the data because the neural network expects samples as rows\n",
    "    data_transposed = data.transpose()\n",
    "    \n",
    "    # Split the data into training and validation sets (75%, 25%)\n",
    "    X_train, X_val = train_test_split(data_transposed, test_size=0.25, random_state=42)\n",
    "    \n",
    "    # Build the autoencoder model\n",
    "    autoencoder = build_ae(original_dim, params)\n",
    "    \n",
    "    # Train the autoencoder model\n",
    "    trained_autoencoder, history = train_ae(autoencoder, X_train, X_val, params)\n",
    "\n",
    "    # Create a separate encoder model from the trained autoencoder\n",
    "    encoder = Model(inputs=trained_autoencoder.input, outputs=trained_autoencoder.layers[-2].output)\n",
    "\n",
    "    # Transform the input data into the latent space to get the latent features\n",
    "    latent_features = encoder.predict(data_transposed)\n",
    "\n",
    "    # Save the latent features to a CSV file\n",
    "    pd.DataFrame(latent_features, index=data_transposed.index).to_csv('latent_features.csv')\n",
    "\n",
    "    # Plot the loss curves and reconstruction error\n",
    "    plot_loss(history)\n",
    "    plot_reconstruction_error(trained_autoencoder, X_val)\n",
    "    \n",
    "\n",
    "    # Save the trained autoencoder and encoder models\n",
    "    trained_autoencoder.save('autoencoder_model.keras')\n",
    "    encoder.save('encoder_model.keras')\n",
    "    \n",
    "    combined_transformation_matrix = extract_and_multiply_weights(trained_autoencoder)\n",
    "    \n",
    "    print(\"Combined Transformation Matrix Shape:\", combined_transformation_matrix.shape)\n",
    "    # Convert the numpy array to a pandas DataFrame\n",
    "    df = pd.DataFrame(combined_transformation_matrix)\n",
    "    # Extract the combined transformation matrix\n",
    "    combined_transformation_matrix = extract_and_multiply_weights(trained_autoencoder)\n",
    "    \n",
    "    # Convert the numpy array to a pandas DataFrame, using gene names as the index\n",
    "    combined_transformation_df = pd.DataFrame(combined_transformation_matrix, index=data.index)\n",
    "\n",
    "    # Save the DataFrame to a CSV file\n",
    "    combined_transformation_df.to_csv('combined_transformation_matrix_with_genes.csv')\n",
    "\n",
    "# Run the main function if the script is run directly (not imported as a module)\n",
    "if __name__ == \"__main__\":\n",
    "    main()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "13c1079e-0738-4b3e-964c-cedb733a629a",
   "metadata": {},
   "source": [
    "In the step below, we first load the dataset. Using a pre-trained autoencoder model, we then generate predictions on the data. Following this, we compute the correlation between the original and predicted data for each sample. To visualize the distribution of these correlations, we create appropriate plots. The aim is to have the R-squared value as close as possible to the loading of the PCA. Achieving this allows us to make a fair comparison between the performance of the autoencoder model and the PCA model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "04b5b530-89cf-45ef-a04e-0203dfec92f5",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                 TDP93C287  TDP93TDPC014  TDP93TDPC068  TDP93TDPC134  \\\n",
      "ENSG00000000419   8.655586      9.319748      9.910092      8.151097   \n",
      "ENSG00000000457   7.536906      8.075948      8.562040      8.864825   \n",
      "ENSG00000000460   7.173539      6.843101      7.104987      7.161539   \n",
      "ENSG00000000938  10.961783      7.433909     10.440624      9.927085   \n",
      "ENSG00000000971   5.773083      7.706707      6.704198      4.321877   \n",
      "\n",
      "                 TDP94C022  TDP94C035  TDP94C079  TDP94C100  TDP94C158  \\\n",
      "ENSG00000000419   9.329317   8.670321   9.291516   8.831926   9.420048   \n",
      "ENSG00000000457   7.450732   8.457329   7.398729   7.958971   8.038661   \n",
      "ENSG00000000460   8.923964   7.818642   6.657649   7.917220   6.311222   \n",
      "ENSG00000000938   6.807541  10.697348  11.882826   7.289811   8.493930   \n",
      "ENSG00000000971   5.031730   8.032692   3.906962   8.108656   5.925756   \n",
      "\n",
      "                 TDP94C240  ...  TDPPS900181  TDPPS900196  TDPPS910112  \\\n",
      "ENSG00000000419   8.795597  ...     9.342703     9.865924     8.360825   \n",
      "ENSG00000000457   8.355285  ...     7.463325     7.786275     7.926907   \n",
      "ENSG00000000460   6.542077  ...     7.371893     7.263690     9.051097   \n",
      "ENSG00000000938   5.854342  ...    10.094467    10.702120     7.903701   \n",
      "ENSG00000000971   6.912095  ...     9.719061     7.071156     5.267116   \n",
      "\n",
      "                 TDPPS920020  TDPPS920022  TDPPS920025  TDPPS920042  \\\n",
      "ENSG00000000419     9.758897     9.151121     8.736171     9.664389   \n",
      "ENSG00000000457     8.208226     7.466056     8.312369     7.572478   \n",
      "ENSG00000000460     6.617400     7.805494     8.259546     7.389162   \n",
      "ENSG00000000938     7.505892    10.150450     9.301983     9.690991   \n",
      "ENSG00000000971     6.356763     5.823421     5.839385     8.130593   \n",
      "\n",
      "                 TDPPS920197  TDPPS920259  TDPPS930024  \n",
      "ENSG00000000419     8.882510     8.201473     9.313138  \n",
      "ENSG00000000457     8.412820     8.209651     8.196317  \n",
      "ENSG00000000460     6.706902     8.221832     8.053330  \n",
      "ENSG00000000938     9.427635     7.970690     6.204927  \n",
      "ENSG00000000971    10.023701     4.770442     6.490266  \n",
      "\n",
      "[5 rows x 506 columns]\n",
      "16/16 [==============================] - 0s 12ms/step\n",
      "Number of correlations computed:  506\n",
      "Average Correlation:  0.519729158792989\n",
      "Average Squared Correlation:  0.2704767624452058\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAjIAAAHHCAYAAACle7JuAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/YYfK9AAAACXBIWXMAAA9hAAAPYQGoP6dpAABGf0lEQVR4nO3deVxWZf7/8feNsokCAgIu4J5L7phLrimJpo6WTWVp6pg2ueRaM06jZjq5p9WYVuOgrY46Lba4hcs05pLmMhniWpAKBi4IyiJcvz/6eX+7ZREQuTn4ej4e90Pu61znnM+5z33Lm3Oucx+bMcYIAADAglycXQAAAEBREWQAAIBlEWQAAIBlEWQAAIBlEWQAAIBlEWQAAIBlEWQAAIBlEWQAAIBlEWQAAIBlEWRgWbVq1dLQoUOdXUaZN3/+fNWpU0flypVTixYtnF1OqbVixQrZbDb9+OOPzi6l2L344ouy2WzFusxt27bJZrNp27Ztxbpc3HkIMigVrv8S2Lt3b67Tu3btqiZNmtzyer788ku9+OKLt7ycO8WmTZv0/PPPq0OHDoqMjNTLL7+cb//PPvtMXbp0UWBgoCpUqKA6derokUce0YYNG0qoYus4cOCABg0apJCQELm7u8vPz0/h4eGKjIxUVlaWs8srNm+88YZWrFjh7DJQhpV3dgFAUcXExMjFpXBZ/Msvv9SSJUsIMwW0ZcsWubi4aPny5XJzc8u374IFC/Tcc8+pS5cumjJliipUqKDjx4/rq6++0qpVq9SzZ88Sqrr0+8c//qE//vGPCgoK0uDBg1W/fn1dvnxZUVFRGj58uM6ePau//OUvzi6zWLzxxhsKCAjIcfS0c+fOunr16k3fV8DNEGRgWe7u7s4uodBSU1Pl5eXl7DIK7Ny5c/L09LzpL5tr165p5syZuv/++7Vp06Zcl1NaXbt2TdnZ2SX2C3XXrl364x//qPbt2+vLL79UpUqV7NPGjx+vvXv36vvvv7/l9WRnZysjI0MeHh45ppWG96GLi0uutQGFxaklWNaNY2QyMzM1Y8YM1a9fXx4eHvL391fHjh21efNmSdLQoUO1ZMkSSZLNZrM/rktNTdWkSZPsh/obNGigBQsW6MYbxF+9elXPPvusAgICVKlSJf3ud7/T6dOnZbPZHI70XB9X8MMPP+jxxx9X5cqV1bFjR0nSoUOHNHToUNWpU0ceHh4KDg7WH/7wByUlJTms6/oyjh49qkGDBsnHx0dVqlTR1KlTZYxRXFyc+vXrJ29vbwUHB2vhwoUFeu2uB4+6devK3d1dtWrV0l/+8help6fb+9hsNkVGRio1NdX+WuV1iiAxMVHJycnq0KFDrtMDAwMdnv/888/q37+/vLy8FBgYqAkTJmjjxo05xkzkNQ6qa9eu6tq1q/15RkaGpk2bprCwMPn4+MjLy0udOnXS1q1bHeb78ccfZbPZtGDBAi1evNi+/T/88IMk6ciRI3r44Yfl5+cnDw8PtW7dWuvWrcux/sOHD6tbt27y9PRUjRo1NGvWLGVnZ+e67TeaMWOGbDab3n//fYcQc13r1q0dtrmg70ubzaYxY8bo/fff19133y13d3dt2LDBftp2+/btGjVqlAIDA1WjRg37fOvXr1enTp3k5eWlSpUqqXfv3jp8+PBNtyMyMlLdunVTYGCg3N3d1bhxYy1dutShT61atXT48GFt377d/h66vt/yGiOzZs0ahYWFydPTUwEBARo0aJBOnz7t0Gfo0KGqWLGiTp8+rf79+6tixYqqUqWKJk+enOO03KpVqxQWFqZKlSrJ29tbTZs21auvvnrT7YN1cEQGpcqlS5eUmJiYoz0zM/Om87744ouaPXu2nnrqKbVp00bJycnau3evvvvuO91///16+umndebMGW3evFnvvvuuw7zGGP3ud7/T1q1bNXz4cLVo0UIbN27Uc889p9OnT2vRokX2vkOHDtXq1as1ePBgtWvXTtu3b1fv3r3zrOv3v/+96tevr5dfftn+y2fz5s06efKkhg0bpuDgYB0+fFhvvfWWDh8+rF27duUYWPnoo4+qUaNGmjNnjr744gvNmjVLfn5+evPNN9WtWzfNnTtX77//viZPnqx77rlHnTt3zve1euqpp7Ry5Uo9/PDDmjRpknbv3q3Zs2crOjpaH3/8sSTp3Xff1VtvvaU9e/boH//4hyTp3nvvzXV5gYGB8vT01GeffaaxY8fKz88vz3VfvXpV3bt3V2xsrJ599llVq1ZN7777rrZs2ZJvzflJTk7WP/7xDw0cOFAjRozQ5cuXtXz5ckVERGjPnj05BilHRkYqLS1NI0eOtI9POXz4sDp06KDq1avrz3/+s7y8vLR69Wr1799f//73v/Xggw9KkuLj43Xffffp2rVr9n5vvfWWPD09b1rnlStXFBUVpc6dOys0NPSm/QvzvpR+PRW4evVqjRkzRgEBAapVq5YOHDggSRo1apSqVKmiadOmKTU1VdKv+3jIkCGKiIjQ3LlzdeXKFS1dulQdO3bU/v37VatWrTxrW7p0qe6++2797ne/U/ny5fXZZ59p1KhRys7O1ujRoyVJixcv1tixY1WxYkW98MILkqSgoKA8l7lixQoNGzZM99xzj2bPnq2EhAS9+uqr2rFjh/bv3y9fX19736ysLEVERKht27ZasGCBvvrqKy1cuFB169bVM888I+nXz9nAgQPVvXt3zZ07V5IUHR2tHTt2aNy4cTd9/WERBigFIiMjjaR8H3fffbfDPDVr1jRDhgyxP2/evLnp3bt3vusZPXq0ye1t/8knnxhJZtasWQ7tDz/8sLHZbOb48ePGGGP27dtnJJnx48c79Bs6dKiRZKZPn25vmz59upFkBg4cmGN9V65cydH24YcfGknmP//5T45ljBw50t527do1U6NGDWOz2cycOXPs7RcuXDCenp4Or0luDhw4YCSZp556yqF98uTJRpLZsmWLvW3IkCHGy8sr3+VdN23aNCPJeHl5mV69epm//e1vZt++fTn6LV682Egyq1evtrelpqaaevXqGUlm69at9vYb9/F1Xbp0MV26dLE/v3btmklPT3foc+HCBRMUFGT+8Ic/2NtOnTplJBlvb29z7tw5h/7du3c3TZs2NWlpafa27Oxsc++995r69evb28aPH28kmd27d9vbzp07Z3x8fIwkc+rUqTxfo4MHDxpJZty4cXn2+a2Cvi+NMUaScXFxMYcPH3boe/2z1bFjR3Pt2jV7++XLl42vr68ZMWKEQ//4+Hjj4+Pj0H79ffhbub2HIyIiTJ06dRza7r77bod9dd3WrVsd9ndGRoYJDAw0TZo0MVevXrX3+/zzz40kM23aNHvbkCFDjCTz0ksvOSyzZcuWJiwszP583Lhxxtvb22G7UfZwagmlypIlS7R58+Ycj2bNmt10Xl9fXx0+fFjHjh0r9Hq//PJLlStXTs8++6xD+6RJk2SM0fr16yXJfvXNqFGjHPqNHTs2z2X/8Y9/zNH227/e09LSlJiYqHbt2kmSvvvuuxz9n3rqKfvP5cqVU+vWrWWM0fDhw+3tvr6+atCggU6ePJlnLdKv2ypJEydOdGifNGmSJOmLL77Id/68zJgxQx988IFatmypjRs36oUXXlBYWJhatWql6Ohoh/VXrVpVDz/8sL2tQoUKGjlyZJHWK/36mlwf45Kdna3z58/r2rVrat26da6v54ABA1SlShX78/Pnz2vLli165JFHdPnyZSUmJioxMVFJSUmKiIjQsWPH7Kc3vvzyS7Vr105t2rSxz1+lShU98cQTN60zOTlZknI9pZSbgr4vr+vSpYsaN26c67JGjBihcuXK2Z9v3rxZFy9e1MCBA+3bm5iYqHLlyqlt27Y5Tsvd6Lfv4etHUrt06aKTJ0/q0qVLBdq+39q7d6/OnTunUaNGOYyd6d27txo2bJjr+/LGz1anTp0c3v++vr5KTU21n15G2cSpJZQqbdq0UevWrXO0V65cOddTTr/10ksvqV+/frrrrrvUpEkT9ezZU4MHDy5QCPrpp59UrVq1HL9gGjVqZJ9+/V8XFxfVrl3boV+9evXyXPaNfaVff3HOmDFDq1atyjEQNrdfAjeehvDx8ZGHh4cCAgJytN84zuZG17fhxpqDg4Pl6+tr39aiGDhwoAYOHKjk5GTt3r1bK1as0AcffKC+ffvq+++/l4eHh3766SfVq1cvx+mzBg0aFHm9krRy5UotXLhQR44ccTgVmdvrf2Pb8ePHZYzR1KlTNXXq1FyXf+7cOVWvXl0//fST2rZtm2N6Qer39vaWJF2+fPmmfaWCvy+vy21b85p2PfB369Yt31rzsmPHDk2fPl07d+7UlStXHKZdunRJPj4++c5/o+vbktvr2LBhQ/33v/91aPPw8HAIo9Kv/09cuHDB/nzUqFFavXq1evXqperVq6tHjx565JFHuIKujCHIoMzo3LmzTpw4oU8//VSbNm3SP/7xDy1atEjLli1zOKJR0nIbO/HII4/om2++0XPPPacWLVqoYsWKys7OVs+ePXMdNPrbv6Tza5OUYxBoXor7C85+y9vbW/fff7/uv/9+ubq6auXKldq9e7e6dOlSqOXkVWNWVpbD9r/33nsaOnSo+vfvr+eee06BgYEqV66cZs+erRMnTuSY/8Z9cv01nzx5siIiInJdZ35htaDq1aun8uXL63//+98tLys3+Y3TyWub3333XQUHB+foX7583r8eTpw4oe7du6thw4Z65ZVXFBISIjc3N3355ZdatGhRgQc+34q83v+/FRgYqAMHDmjjxo1av3691q9fr8jISD355JNauXLlba8RJYMggzLFz89Pw4YN07Bhw5SSkqLOnTvrxRdftAeZvH4x1qxZU1999ZUuX77s8NfvkSNH7NOv/5udna1Tp06pfv369n7Hjx8vcI0XLlxQVFSUZsyYoWnTptnbi3JKrCiub8OxY8fsf9lLUkJCgi5evGjf1uLSunVrrVy5UmfPnrWv//vvv5cxxmF/xMTE5Ji3cuXKunjxYo72n376SXXq1LE/X7t2rerUqaOPPvrIYZnTp08vUI3Xl+Xq6qrw8PB8+9asWTPXfZVb/TeqUKGCunXrpi1btiguLk4hISE3XVdB3pdFUbduXUm//rK/2Tbf6LPPPlN6errWrVvncLQwt9NRBQ3M17clJiYmx1GimJiYIm+rm5ub+vbtq759+yo7O1ujRo3Sm2++qalTpxZLOIXzMUYGZcaNp1QqVqyoevXqOVxSfP27M2785fjAAw8oKytLf//73x3aFy1aJJvNpl69ekmS/a/1N954w6Hf66+/XuA6r/8leeORk8WLFxd4GbfigQceyHV9r7zyiiTlewVWXq5cuaKdO3fmOu36OI7rpwweeOABnTlzRmvXrnWY/6233soxb926dbVr1y5lZGTY2z7//HPFxcU59MvtNd29e3eeNd0oMDBQXbt21ZtvvmkPXL/1yy+/2H9+4IEHtGvXLu3Zs8dh+vvvv1+gdU2fPl3GGA0ePFgpKSk5pu/bt89+tKCg78uiiIiIkLe3t15++eVcrwr87TbfKLfX+9KlS4qMjMzR18vLK9cweqPWrVsrMDBQy5Ytc/jMrl+/XtHR0UV6X974f4KLi4v9VPNv1wFr44gMyozGjRura9euCgsLk5+fn/bu3au1a9dqzJgx9j5hYWGSpGeffVYREREqV66cHnvsMfXt21f33XefXnjhBf34449q3ry5Nm3apE8//VTjx4+3//UaFhamAQMGaPHixUpKSrJffn306FFJBfvr09vbW507d9a8efOUmZmp6tWra9OmTTp16tRteFVyat68uYYMGaK33npLFy9eVJcuXbRnzx6tXLlS/fv313333VfoZV65ckX33nuv2rVrp549eyokJEQXL17UJ598oq+//lr9+/dXy5YtJf066PTvf/+7nnzySe3bt09Vq1bVu+++qwoVKuRY7lNPPaW1a9eqZ8+eeuSRR3TixAm999579v1xXZ8+ffTRRx/pwQcfVO/evXXq1CktW7ZMjRs3zjUs5GbJkiXq2LGjmjZtqhEjRqhOnTpKSEjQzp079fPPP+vgwYOSpOeff17vvvuuevbsqXHjxtkvv65Zs6YOHTp00/Xce++9WrJkiUaNGqWGDRs6fLPvtm3btG7dOs2aNUuSCvy+LApvb28tXbpUgwcPVqtWrfTYY4+pSpUqio2N1RdffKEOHTrkCFDX9ejRw36k4+mnn1ZKSorefvttBQYG5giCYWFhWrp0qWbNmqV69eopMDAw13E5rq6umjt3roYNG6YuXbpo4MCB9suva9WqpQkTJhR6G5966imdP39e3bp1U40aNfTTTz/p9ddfV4sWLRyORsLinHW5FPBb1y8R/fbbb3Od3qVLl5tefj1r1izTpk0b4+vrazw9PU3Dhg3N3/72N5ORkWHvc+3aNTN27FhTpUoVY7PZHC4pvXz5spkwYYKpVq2acXV1NfXr1zfz58832dnZDutNTU01o0ePNn5+fqZixYqmf//+JiYmxkhyuBz6+iWrv/zyS47t+fnnn82DDz5ofH19jY+Pj/n9739vzpw5k+cl3DcuI6/LonN7nXKTmZlpZsyYYWrXrm1cXV1NSEiImTJlisOlx/mtJ7flvf3226Z///6mZs2axt3d3VSoUMG0bNnSzJ8/P8el0T/99JP53e9+ZypUqGACAgLMuHHjzIYNG3Jcfm2MMQsXLjTVq1c37u7upkOHDmbv3r05Lr/Ozs42L7/8sn3dLVu2NJ9//rkZMmSIqVmzpr3f9cuv58+fn+t2nDhxwjz55JMmODjYuLq6murVq5s+ffqYtWvXOvQ7dOiQ6dKli/Hw8DDVq1c3M2fONMuXL7/p5de/tW/fPvP444/b32+VK1c23bt3NytXrjRZWVn2fgV9X0oyo0ePzrGem322tm7daiIiIoyPj4/x8PAwdevWNUOHDjV79+6198nt8ut169aZZs2aGQ8PD1OrVi0zd+5c889//jPHaxAfH2969+5tKlWqZCTZ99uNl19f969//cu0bNnSuLu7Gz8/P/PEE0+Yn3/+2aFPXu/LG+tcu3at6dGjhwkMDDRubm4mNDTUPP300+bs2bO5vhawJpsxBRwZCCBPBw4cUMuWLfXee+8V6DJc5LRt2zbdd9992rp1q8O39gJAfhgjAxTS1atXc7QtXrxYLi4uN/1GXQBA8WKMDFBI8+bN0759+3TfffepfPny9ss6R44cedOrUAAAxYsgAxTSvffeq82bN2vmzJlKSUlRaGioXnzxRfu9ZAAAJYcxMgAAwLIYIwMAACyLIAMAACyrzI+Ryc7O1pkzZ1SpUqXbem8ZAABQfIwxunz5sqpVqyYXl7yPu5T5IHPmzBmuJAEAwKLi4uJUo0aNPKeX+SBz/UZrcXFxN70tPQAAKB2Sk5MVEhLicMPU3JT5IHP9dJK3tzdBBgAAi7nZsBAG+wIAAMsiyAAAAMsiyAAAAMsiyAAAAMsiyAAAAMsiyAAAAMsiyAAAAMsiyAAAAMsiyAAAAMsiyAAAAMsiyAAAAMsiyAAAAMsiyAAAAMsiyAAAAMsq7+wCAAD4rdjYWCUmJubbJyAgQKGhoSVUEUozggwAoNSIjY1Vg4aNlHb1Sr79PDwrKOZINGEGBBkAQOmRmJiotKtX5N9nklz9Q3Ltk5kUp6TPFyoxMZEgA4IMAKD0cfUPkXtwPWeXAQtgsC8AALAsggwAALAsggwAALAsggwAALAsggwAALAsggwAALAsggwAALAsggwAALAsggwAALAspweZ06dPa9CgQfL395enp6eaNm2qvXv32qcbYzRt2jRVrVpVnp6eCg8P17Fjx5xYMQAAKC2cGmQuXLigDh06yNXVVevXr9cPP/yghQsXqnLlyvY+8+bN02uvvaZly5Zp9+7d8vLyUkREhNLS0pxYOQAAKA2ceq+luXPnKiQkRJGRkfa22rVr2382xmjx4sX661//qn79+kmS3nnnHQUFBemTTz7RY489VuI1AwCA0sOpR2TWrVun1q1b6/e//70CAwPVsmVLvf322/bpp06dUnx8vMLDw+1tPj4+atu2rXbu3JnrMtPT05WcnOzwAAAAZZNTg8zJkye1dOlS1a9fXxs3btQzzzyjZ599VitXrpQkxcfHS5KCgoIc5gsKCrJPu9Hs2bPl4+Njf4SE5H4beAAAYH1ODTLZ2dlq1aqVXn75ZbVs2VIjR47UiBEjtGzZsiIvc8qUKbp06ZL9ERcXV4wVAwCA0sSpQaZq1apq3LixQ1ujRo0UGxsrSQoODpYkJSQkOPRJSEiwT7uRu7u7vL29HR4AAKBscmqQ6dChg2JiYhzajh49qpo1a0r6deBvcHCwoqKi7NOTk5O1e/dutW/fvkRrBQAApY9Tr1qaMGGC7r33Xr388st65JFHtGfPHr311lt66623JEk2m03jx4/XrFmzVL9+fdWuXVtTp05VtWrV1L9/f2eWDgAASgGnBpl77rlHH3/8saZMmaKXXnpJtWvX1uLFi/XEE0/Y+zz//PNKTU3VyJEjdfHiRXXs2FEbNmyQh4eHEysHAAClgVODjCT16dNHffr0yXO6zWbTSy+9pJdeeqkEqwIAAFbg9FsUAAAAFBVBBgAAWBZBBgAAWBZBBgAAWJbTB/sCAOAssbGxSkxMzLdPQECAQkNDS6giFBZBBgBwR4qNjVWDho2UdvVKvv08PCso5kg0YaaUIsgAAO5IiYmJSrt6Rf59JsnVP/cbDGcmxSnp84VKTEwkyJRSBBkAwB3N1T9E7sH1nF0GiojBvgAAwLIIMgAAwLIIMgAAwLIIMgAAwLIIMgAAwLIIMgAAwLIIMgAAwLIIMgAAwLIIMgAAwLIIMgAAwLIIMgAAwLIIMgAAwLIIMgAAwLIIMgAAwLIIMgAAwLIIMgAAwLIIMgAAwLIIMgAAwLLKO7sAAACKIjo6Ot/pAQEBCg0NLaFq4CwEGQCApWSlXJBsNg0aNCjffh6eFRRzJJowU8YRZAAAlpKdniIZI/8+k+TqH5Jrn8ykOCV9vlCJiYkEmTKOIAMAsCRX/xC5B9dzdhlwMgb7AgAAyyLIAAAAyyLIAAAAyyLIAAAAyyLIAAAAyyLIAAAAyyLIAAAAyyLIAAAAyyLIAAAAyyLIAAAAyyLIAAAAyyLIAAAAyyLIAAAAy+Lu1wCAMis6OrpI02AdBBkAQJmTlXJBstk0aNAgZ5eC28ypQebFF1/UjBkzHNoaNGigI0eOSJLS0tI0adIkrVq1Sunp6YqIiNAbb7yhoKAgZ5QLALCI7PQUyRj595kkV/+QXPtcPblXl75+r4QrQ3Fz+hGZu+++W1999ZX9efny/1fShAkT9MUXX2jNmjXy8fHRmDFj9NBDD2nHjh3OKBUAYDGu/iFyD66X67TMpLgSrga3g9ODTPny5RUcHJyj/dKlS1q+fLk++OADdevWTZIUGRmpRo0aadeuXWrXrl1JlwoAAEoZp1+1dOzYMVWrVk116tTRE088odjYWEnSvn37lJmZqfDwcHvfhg0bKjQ0VDt37sxzeenp6UpOTnZ4AACAssmpQaZt27ZasWKFNmzYoKVLl+rUqVPq1KmTLl++rPj4eLm5ucnX19dhnqCgIMXHx+e5zNmzZ8vHx8f+CAnJ/dwoAACwPqeeWurVq5f952bNmqlt27aqWbOmVq9eLU9PzyItc8qUKZo4caL9eXJyMmEGAIAyyumnln7L19dXd911l44fP67g4GBlZGTo4sWLDn0SEhJyHVNznbu7u7y9vR0eAACgbCpVQSYlJUUnTpxQ1apVFRYWJldXV0VFRdmnx8TEKDY2Vu3bt3dilQAAoLRw6qmlyZMnq2/fvqpZs6bOnDmj6dOnq1y5cho4cKB8fHw0fPhwTZw4UX5+fvL29tbYsWPVvn17rlgCAACSnBxkfv75Zw0cOFBJSUmqUqWKOnbsqF27dqlKlSqSpEWLFsnFxUUDBgxw+EI8AAAAyclBZtWqVflO9/Dw0JIlS7RkyZISqggAAFhJqRojAwAAUBgEGQAAYFkEGQAAYFkEGQAAYFkEGQAAYFkEGQAAYFkEGQAAYFkEGQAAYFkEGQAAYFkEGQAAYFkEGQAAYFkEGQAAYFkEGQAAYFkEGQAAYFkEGQAAYFkEGQAAYFkEGQAAYFkEGQAAYFkEGQAAYFkEGQAAYFkEGQAAYFkEGQAAYFkEGQAAYFkEGQAAYFkEGQAAYFkEGQAAYFkEGQAAYFkEGQAAYFkEGQAAYFkEGQAAYFkEGQAAYFkEGQAAYFkEGQAAYFkEGQAAYFkEGQAAYFkEGQAAYFkEGQAAYFnlnV0AAODOERsbq8TExDynR0dHl2A1KAsIMgCAEhEbG6sGDRsp7eoVZ5eCMoQgAwAoEYmJiUq7ekX+fSbJ1T8k1z5XT+7Vpa/fK+HKYGUEGQBAiXL1D5F7cL1cp2UmxZVwNbA6BvsCAADLIsgAAADLIsgAAADLIsgAAADLKjVBZs6cObLZbBo/fry9LS0tTaNHj5a/v78qVqyoAQMGKCEhwXlFAgCAUqVUBJlvv/1Wb775ppo1a+bQPmHCBH322Wdas2aNtm/frjNnzuihhx5yUpUAAKC0cXqQSUlJ0RNPPKG3335blStXtrdfunRJy5cv1yuvvKJu3bopLCxMkZGR+uabb7Rr1y4nVgwAAEoLpweZ0aNHq3fv3goPD3do37dvnzIzMx3aGzZsqNDQUO3cuTPP5aWnpys5OdnhAQAAyianfiHeqlWr9N133+nbb7/NMS0+Pl5ubm7y9fV1aA8KClJ8fHyey5w9e7ZmzJhR3KUCAIBSyGlHZOLi4jRu3Di9//778vDwKLblTpkyRZcuXbI/4uL4lkgAAMoqpwWZffv26dy5c2rVqpXKly+v8uXLa/v27XrttddUvnx5BQUFKSMjQxcvXnSYLyEhQcHBwXku193dXd7e3g4PAABQNjnt1FL37t31v//9z6Ft2LBhatiwof70pz8pJCRErq6uioqK0oABAyRJMTExio2NVfv27Z1RMgAAKGWcFmQqVaqkJk2aOLR5eXnJ39/f3j58+HBNnDhRfn5+8vb21tixY9W+fXu1a9fOGSUDAIBSplTf/XrRokVycXHRgAEDlJ6eroiICL3xxhvOLgsAAJQSRQoyJ0+eVJ06dYq7Fm3bts3huYeHh5YsWaIlS5YU+7oAAID1FWmwb7169XTffffpvffeU1paWnHXBAAAUCBFCjLfffedmjVrpokTJyo4OFhPP/209uzZU9y1AQAA5KtIQaZFixZ69dVXdebMGf3zn//U2bNn1bFjRzVp0kSvvPKKfvnll+KuEwAAIIdb+h6Z8uXL66GHHtKaNWs0d+5cHT9+XJMnT1ZISIiefPJJnT17trjqBAAAyOGWgszevXs1atQoVa1aVa+88oomT56sEydOaPPmzTpz5oz69etXXHUCAADkUKSrll555RVFRkYqJiZGDzzwgN555x098MADcnH5NRfVrl1bK1asUK1atYqzVgAAAAdFCjJLly7VH/7wBw0dOlRVq1bNtU9gYKCWL19+S8UBAADkp0hB5tixYzft4+bmpiFDhhRl8QAAAAVSpDEykZGRWrNmTY72NWvWaOXKlbdcFAAAQEEUKcjMnj1bAQEBOdoDAwP18ssv33JRAAAABVGkIBMbG6vatWvnaK9Zs6ZiY2NvuSgAAICCKFKQCQwM1KFDh3K0Hzx4UP7+/rdcFAAAQEEUKcgMHDhQzz77rLZu3aqsrCxlZWVpy5YtGjdunB577LHirhEAACBXRbpqaebMmfrxxx/VvXt3lS//6yKys7P15JNPMkYGAACUmCIFGTc3N/3rX//SzJkzdfDgQXl6eqpp06aqWbNmcdcHAACQpyIFmevuuusu3XXXXcVVCwAAQKEUKchkZWVpxYoVioqK0rlz55Sdne0wfcuWLcVSHAAAQH6KFGTGjRunFStWqHfv3mrSpIlsNltx1wUAAHBTRQoyq1at0urVq/XAAw8Udz0AAAAFVqTLr93c3FSvXr3irgUAAKBQihRkJk2apFdffVXGmOKuBwAAoMCKdGrpv//9r7Zu3ar169fr7rvvlqurq8P0jz76qFiKAwAAyE+Rgoyvr68efPDB4q4FAACgUIoUZCIjI4u7DgAAgEIr0hgZSbp27Zq++uorvfnmm7p8+bIk6cyZM0pJSSm24gAAAPJTpCMyP/30k3r27KnY2Filp6fr/vvvV6VKlTR37lylp6dr2bJlxV0nAABADkU6IjNu3Di1bt1aFy5ckKenp739wQcfVFRUVLEVBwAAkJ8iHZH5+uuv9c0338jNzc2hvVatWjp9+nSxFAYAAHAzRToik52draysrBztP//8sypVqnTLRQEAABREkYJMjx49tHjxYvtzm82mlJQUTZ8+ndsWAACAElOkU0sLFy5URESEGjdurLS0ND3++OM6duyYAgIC9OGHHxZ3jQAAALkqUpCpUaOGDh48qFWrVunQoUNKSUnR8OHD9cQTTzgM/gUAALidihRkJKl8+fIaNGhQcdYCAABQKEUKMu+8806+05988skiFQMAAFAYRQoy48aNc3iemZmpK1euyM3NTRUqVCDIAACAElGkq5YuXLjg8EhJSVFMTIw6duzIYF8AAFBiinyvpRvVr19fc+bMyXG0BgAA4HYp8mDfXBdWvrzOnDlTnIsEAMDpoqOj850eEBCg0NDQEqoGv1WkILNu3TqH58YYnT17Vn//+9/VoUOHYikMAABny0q5INlsN71K18OzgmKORBNmnKBIQaZ///4Oz202m6pUqaJu3bpp4cKFxVEXAABOl52eIhkj/z6T5OofkmufzKQ4JX2+UImJiQQZJyhSkMnOzi7uOgAAKLVc/UPkHlzP2WUgF8U22BcAAKCkFemIzMSJEwvc95VXXinKKgAAAG6qSEFm//792r9/vzIzM9WgQQNJ0tGjR1WuXDm1atXK3s9msxVPlQAAALkoUpDp27evKlWqpJUrV6py5cqSfv2SvGHDhqlTp06aNGlSsRYJAACQmyKNkVm4cKFmz55tDzGSVLlyZc2aNatQVy0tXbpUzZo1k7e3t7y9vdW+fXutX7/ePj0tLU2jR4+Wv7+/KlasqAEDBighIaEoJQMAgDKoSEEmOTlZv/zyS472X375RZcvXy7wcmrUqKE5c+Zo37592rt3r7p166Z+/frp8OHDkqQJEybos88+05o1a7R9+3adOXNGDz30UFFKBgAAZVCRTi09+OCDGjZsmBYuXKg2bdpIknbv3q3nnnuuUEGjb9++Ds//9re/aenSpdq1a5dq1Kih5cuX64MPPlC3bt0kSZGRkWrUqJF27dqldu3aFaV0AABQhhQpyCxbtkyTJ0/W448/rszMzF8XVL68hg8frvnz5xepkKysLK1Zs0apqalq37699u3bp8zMTIWHh9v7NGzYUKGhodq5c2eeQSY9PV3p6en258nJyUWqBwDwf2JjY5WYmJhvn/T0dLm7u+c5/WZf8w8URZGCTIUKFfTGG29o/vz5OnHihCSpbt268vLyKvSy/ve//6l9+/ZKS0tTxYoV9fHHH6tx48Y6cOCA3Nzc5Ovr69A/KChI8fHxeS5v9uzZmjFjRqHrAADkLjY2Vg0aNlLa1Sv5d7S5SIYvTEXJuqWbRp49e1Znz55V586d5enpKWNMoS+5btCggQ4cOKBLly5p7dq1GjJkiLZv317kmqZMmeLwPTfJyckKCcn9a6UBADeXmJiotKtX8v2a/qsn9+rS1+8VqA9QnIoUZJKSkvTII49o69atstlsOnbsmOrUqaPhw4ercuXKhbpyyc3NTfXq/fq1z2FhYfr222/16quv6tFHH1VGRoYuXrzocFQmISFBwcHBeS7P3d0930ObAICiye9r+jOT4grcByhORbpqacKECXJ1dVVsbKwqVKhgb3/00Ue1YcOGWyooOztb6enpCgsLk6urq6KiouzTYmJiFBsbq/bt29/SOgAAQNlQpCMymzZt0saNG1WjRg2H9vr16+unn34q8HKmTJmiXr16KTQ0VJcvX9YHH3ygbdu2aePGjfLx8dHw4cM1ceJE+fn5ydvbW2PHjlX79u25YgkAAEgqYpBJTU11OBJz3fnz5wt1WufcuXN68skndfbsWfn4+KhZs2bauHGj7r//fknSokWL5OLiogEDBig9PV0RERF64403ilIyAAAog4oUZDp16qR33nlHM2fOlPTrPZWys7M1b9483XfffQVezvLly/Od7uHhoSVLlmjJkiVFKRMAAJRxRQoy8+bNU/fu3bV3715lZGTo+eef1+HDh3X+/Hnt2LGjuGsEAADIVZEG+zZp0kRHjx5Vx44d1a9fP6Wmpuqhhx7S/v37Vbdu3eKuEQAAIFeFPiKTmZmpnj17atmyZXrhhRduR00AAAAFUugjMq6urjp06NDtqAUAAKBQinRqadCgQTcdqAsAAHC7FWmw77Vr1/TPf/5TX331lcLCwnLcY+mVV14pluIAAADyU6ggc/LkSdWqVUvff/+9WrVqJUk6evSoQ5/C3msJAACgqAoVZOrXr6+zZ89q69atkn69JcFrr72moKCg21IcAABAfgo1RsYY4/B8/fr1Sk1NLdaCAAAACqpIg32vuzHYAAAAlKRCBRmbzZZjDAxjYgAAgLMUaoyMMUZDhw613xgyLS1Nf/zjH3NctfTRRx8VX4UAAAB5KFSQGTJkiMPzQYMGFWsxAAAAhVGoIBMZGXm76gAAACi0In0hHgAAKJzY2FglJibm2ycgIEChoaElVFHZQJABAOA2i42NVYOGjZR29Uq+/Tw8KyjmSDRhphAIMgAA3GaJiYlKu3pF/n0mydU/JNc+mUlxSvp8oRITEwkyhUCQAQCghLj6h8g9uJ6zyyhTbukL8QAAAJyJIAMAACyLU0sAABSD6OjoIk3DrSHIAABwC7JSLkg2G18S6yQEGQAAbkF2eopkTL5XJF09uVeXvn6vhCu7MxBkAAAoBvldkZSZFFfC1dw5GOwLAAAsiyADAAAsiyADAAAsiyADAAAsiyADAAAsiyADAAAsiyADAAAsiyADAAAsiyADAAAsiyADAAAsiyADAAAsiyADAAAsiyADAAAsi7tfAwBQikRHR+c7PSAgQKGhoSVUTelHkAEAoBTISrkg2WwaNGhQvv08PCso5kg0Yeb/I8gAAFAKZKenSMbIv88kufqH5NonMylOSZ8vVGJiIkHm/yPIAABQirj6h8g9uJ6zy7AMBvsCAADLIsgAAADLIsgAAADLIsgAAADLcmqQmT17tu655x5VqlRJgYGB6t+/v2JiYhz6pKWlafTo0fL391fFihU1YMAAJSQkOKliAABQmjg1yGzfvl2jR4/Wrl27tHnzZmVmZqpHjx5KTU2195kwYYI+++wzrVmzRtu3b9eZM2f00EMPObFqAABQWjj18usNGzY4PF+xYoUCAwO1b98+de7cWZcuXdLy5cv1wQcfqFu3bpKkyMhINWrUSLt27VK7du2cUTYAACglStUYmUuXLkmS/Pz8JEn79u1TZmamwsPD7X0aNmyo0NBQ7dy5M9dlpKenKzk52eEBAADKplITZLKzszV+/Hh16NBBTZo0kSTFx8fLzc1Nvr6+Dn2DgoIUHx+f63Jmz54tHx8f+yMkJPdvRwQAANZXaoLM6NGj9f3332vVqlW3tJwpU6bo0qVL9kdcXFwxVQgAAEqbUnGLgjFjxujzzz/Xf/7zH9WoUcPeHhwcrIyMDF28eNHhqExCQoKCg4NzXZa7u7vc3d1vd8kAAKAUcOoRGWOMxowZo48//lhbtmxR7dq1HaaHhYXJ1dVVUVFR9raYmBjFxsaqffv2JV0uAAAoZZx6RGb06NH64IMP9Omnn6pSpUr2cS8+Pj7y9PSUj4+Phg8frokTJ8rPz0/e3t4aO3as2rdvzxVLAADAuUFm6dKlkqSuXbs6tEdGRmro0KGSpEWLFsnFxUUDBgxQenq6IiIi9MYbb5RwpQAAoDRyapAxxty0j4eHh5YsWaIlS5aUQEUAAMBKSs1VSwAAAIVFkAEAAJZFkAEAAJZFkAEAAJZFkAEAAJZFkAEAAJZFkAEAAJZFkAEAAJZFkAEAAJZFkAEAAJZFkAEAAJZFkAEAAJZFkAEAAJZFkAEAAJZFkAEAAJZFkAEAAJZFkAEAAJZFkAEAAJZFkAEAAJZFkAEAAJZV3tkFAACAwomOjs53ekBAgEJDQ0uoGuciyAAAYBFZKRckm02DBg3Kt5+HZwXFHIm+I8IMQQYAAIvITk+RjJF/n0ly9Q/JtU9mUpySPl+oxMREggwAACh9XP1D5B5cz9lllAoM9gUAAJZFkAEAAJZFkAEAAJZFkAEAAJZFkAEAAJZFkAEAAJZFkAEAAJZFkAEAAJZFkAEAAJZFkAEAAJZFkAEAAJZFkAEAAJZFkAEAAJbF3a8B4A4XGxurxMTEPKdHR0eXYDVA4RBkAOAOFhsbqwYNGynt6hVnlwIUCUEGAO5giYmJSrt6Rf59JsnVPyTXPldP7tWlr98r4cqAgiHIAADk6h8i9+B6uU7LTIor4WqAgmOwLwAAsCyCDAAAsCyCDAAAsCyCDAAAsCynBpn//Oc/6tu3r6pVqyabzaZPPvnEYboxRtOmTVPVqlXl6emp8PBwHTt2zDnFAgCAUsepQSY1NVXNmzfXkiVLcp0+b948vfbaa1q2bJl2794tLy8vRUREKC0trYQrBQAApZFTL7/u1auXevXqles0Y4wWL16sv/71r+rXr58k6Z133lFQUJA++eQTPfbYYyVZKgAAKIVK7RiZU6dOKT4+XuHh4fY2Hx8ftW3bVjt37sxzvvT0dCUnJzs8AABA2VRqg0x8fLwkKSgoyKE9KCjIPi03s2fPlo+Pj/0REpL7N1UCAADrK7VBpqimTJmiS5cu2R9xcXwjJQAAZVWpDTLBwcGSpISEBIf2hIQE+7TcuLu7y9vb2+EBAADKplIbZGrXrq3g4GBFRUXZ25KTk7V79261b9/eiZUBAIDSwqlXLaWkpOj48eP256dOndKBAwfk5+en0NBQjR8/XrNmzVL9+vVVu3ZtTZ06VdWqVVP//v2dVzQAABYQHR2d7/SAgACFhoaWUDW3j1ODzN69e3XffffZn0+cOFGSNGTIEK1YsULPP/+8UlNTNXLkSF28eFEdO3bUhg0b5OHh4aySAQAo1bJSLkg2mwYNGpRvPw/PCoo5Em35MOPUINO1a1cZY/KcbrPZ9NJLL+mll14qwaoAALCu7PQUyRj595kkV//cr9zNTIpT0ucLlZiYSJABAAClj6t/iNyD6zm7jNuu1A72BQAAuBmCDAAAsCxOLQEAcIcqC1c2EWQAALjDlKUrmwgyAADcYcrSlU0EGQAA7lBl4comBvsCAADLIsgAAADL4tQSAJRhsbGxSkxMzHP6za5aAUo7ggwAlFGxsbFq0LCR0q5ecXYpwG1DkAGAMioxMVFpV6/ke2XK1ZN7denr90q4MqD4EGQAoIzL78qUzKS4Eq4GKF4M9gUAAJbFERkAAJCn0n4bA4IMAADIwSq3MSDIAACAHKxyGwOCDAAAyFNpv40Bg30BAIBlEWQAAIBlEWQAAIBlEWQAAIBlEWQAAIBlEWQAAIBlEWQAAIBlEWQAAIBlEWQAAIBlEWQAAIBlEWQAAIBlEWQAAIBlEWQAAIBlcfdrAChhsbGxSkxMzLdPenq63N3db6lPdHR0keoDrIQgAwAlKDY2Vg0aNlLa1Sv5d7S5SCb71vsAZRxBBgBKUGJiotKuXpF/n0ly9Q/Jtc/Vk3t16ev3iq0PUJYRZADACVz9Q+QeXC/XaZlJccXaByjLGOwLAAAsiyADAAAsi1NLAKDiu5IoICBAoaGhxVkagHwQZADc8YrzSiIPzwqKORJNmAFKCEEGwB2vuK4kykyKU9LnC5WYmEiQAUoIQQYA/r9bvUoIQMljsC8AALAsjsjcgoIMDmTgH3B7FccgXb7KH7AugkwRFXRwIAP/gNunWL/uH4AlWSLILFmyRPPnz1d8fLyaN2+u119/XW3atHFqTQUZHMjAP+D2Ku6v+wdgPaU+yPzrX//SxIkTtWzZMrVt21aLFy9WRESEYmJiFBgY6OzyGPgHlAJ8lT9w5yr1g31feeUVjRgxQsOGDVPjxo21bNkyVahQQf/85z+dXRoAAHCyUh1kMjIytG/fPoWHh9vbXFxcFB4erp07dzqxMgAAUBqU6lNLiYmJysrKUlBQkEN7UFCQjhw5kus86enpSk9Ptz+/dOmSJCk5OblYa0tJSfl1ffHHlZ2RlmufzPM/S5L27dtn758bFxcXZWfffCBiQfrRhz53Up+YmBhJN/kc/v/TRiXSpwCf+VJXM33oUwzv+ZSUlGL/PXt9ecaY/DuaUuz06dNGkvnmm28c2p977jnTpk2bXOeZPn26kcSDBw8ePHjwKAOPuLi4fLNCqT4iExAQoHLlyikhIcGhPSEhQcHBwbnOM2XKFE2cONH+PDs7W+fPn5e/v79sNtttrfe3kpOTFRISori4OHl7e5fYelEw7J/Sjf1TerFvSreytH+MMbp8+bKqVauWb79SHWTc3NwUFhamqKgo9e/fX9KvwSQqKkpjxozJdR53d/ccX3zl6+t7myvNm7e3t+XfTGUZ+6d0Y/+UXuyb0q2s7B8fH5+b9inVQUaSJk6cqCFDhqh169Zq06aNFi9erNTUVA0bNszZpQEAACcr9UHm0Ucf1S+//KJp06YpPj5eLVq00IYNG3IMAAYAAHeeUh9kJGnMmDF5nkoqrdzd3TV9+vR87+8C52H/lG7sn9KLfVO63Yn7x2bMza5rAgAAKJ1K9RfiAQAA5IcgAwAALIsgAwAALIsgAwAALIsgUwhLlixRrVq15OHhobZt22rPnj159j18+LAGDBigWrVqyWazafHixbe8TOStuPfNiy++KJvN5vBo2LDhbdyCsq0w++ftt99Wp06dVLlyZVWuXFnh4eE5+htjNG3aNFWtWlWenp4KDw/XsWPHbvdmlFnFvX+GDh2a4/PTs2fP270ZZVJh9s1HH32k1q1by9fXV15eXmrRooXeffddhz5l8bNDkCmgf/3rX5o4caKmT5+u7777Ts2bN1dERITOnTuXa/8rV66oTp06mjNnTp63UyjsMpG727FvJOnuu+/W2bNn7Y///ve/t2sTyrTC7p9t27Zp4MCB2rp1q3bu3KmQkBD16NFDp0+ftveZN2+eXnvtNS1btky7d++Wl5eXIiIilJaW+43tkLfbsX8kqWfPng6fnw8//LAkNqdMKey+8fPz0wsvvKCdO3fq0KFDGjZsmIYNG6aNGzfa+5TJz05x3NzxTtCmTRszevRo+/OsrCxTrVo1M3v27JvOW7NmTbNo0aJiXSb+z+3YN9OnTzfNmzcvxirvXLf6Pr927ZqpVKmSWblypTHGmOzsbBMcHGzmz59v73Px4kXj7u5uPvzww+It/g5Q3PvHGGOGDBli+vXrV9yl3nGK43dEy5YtzV//+ldjTNn97HBEpgAyMjK0b98+hYeH29tcXFwUHh6unTt3lppl3olu5+t47NgxVatWTXXq1NETTzyh2NjYWy33jlMc++fKlSvKzMyUn5+fJOnUqVOKj493WKaPj4/atm3LZ6eQbsf+uW7btm0KDAxUgwYN9MwzzygpKalYay/rbnXfGGMUFRWlmJgYde7cWVLZ/ewQZAogMTFRWVlZOW6LEBQUpPj4+FKzzDvR7Xod27ZtqxUrVmjDhg1aunSpTp06pU6dOuny5cu3WvIdpTj2z5/+9CdVq1bN/p/v9fn47Ny627F/pF9PK73zzjuKiorS3LlztX37dvXq1UtZWVnFWn9ZVtR9c+nSJVWsWFFubm7q3bu3Xn/9dd1///2Syu5nxxK3KABKWq9evew/N2vWTG3btlXNmjW1evVqDR8+3ImV3VnmzJmjVatWadu2bfLw8HB2ObhBXvvnscces//ctGlTNWvWTHXr1tW2bdvUvXt3Z5R6x6hUqZIOHDiglJQURUVFaeLEiapTp466du3q7NJuG47IFEBAQIDKlSunhIQEh/aEhIR8B4uW9DLvRCX1Ovr6+uquu+7S8ePHi22Zd4Jb2T8LFizQnDlztGnTJjVr1szefn0+Pju37nbsn9zUqVNHAQEBfH4Koaj7xsXFRfXq1VOLFi00adIkPfzww5o9e7aksvvZIcgUgJubm8LCwhQVFWVvy87OVlRUlNq3b19qlnknKqnXMSUlRSdOnFDVqlWLbZl3gqLun3nz5mnmzJnasGGDWrdu7TCtdu3aCg4OdlhmcnKydu/ezWenkG7H/snNzz//rKSkJD4/hVBc/7dlZ2crPT1dUhn+7Dh7tLFVrFq1yri7u5sVK1aYH374wYwcOdL4+vqa+Ph4Y4wxgwcPNn/+85/t/dPT083+/fvN/v37TdWqVc3kyZPN/v37zbFjxwq8TBTM7dg3kyZNMtu2bTOnTp0yO3bsMOHh4SYgIMCcO3euxLfP6gq7f+bMmWPc3NzM2rVrzdmzZ+2Py5cvO/Tx9fU1n376qTl06JDp16+fqV27trl69WqJb5/VFff+uXz5spk8ebLZuXOnOXXqlPnqq69Mq1atTP369U1aWppTttGqCrtvXn75ZbNp0yZz4sQJ88MPP5gFCxaY8uXLm7ffftvepyx+dggyhfD666+b0NBQ4+bmZtq0aWN27dpln9alSxczZMgQ+/NTp04ZSTkeXbp0KfAyUXDFvW8effRRU7VqVePm5maqV69uHn30UXP8+PES3KKypTD7p2bNmrnun+nTp9v7ZGdnm6lTp5qgoCDj7u5uunfvbmJiYkpwi8qW4tw/V65cMT169DBVqlQxrq6upmbNmmbEiBH8gVZEhdk3L7zwgqlXr57x8PAwlStXNu3btzerVq1yWF5Z/OzYjDGmZI8BAQAAFA/GyAAAAMsiyAAAAMsiyAAAAMsiyAAAAMsiyAAAAMsiyAAAAMsiyAAAAMsiyABwuq5du2r8+PHOLqPQiqtuq24/UBoQZAAL+eWXX/TMM88oNDRU7u7uCg4OVkREhHbs2OHs0m67jIwMzZs3T82bN1eFChUUEBCgDh06KDIyUpmZmc4ur0C2bdsmm82mixcvOrR/9NFHmjlzpnOKAiyuvLMLAFBwAwYMUEZGhlauXKk6deooISFBUVFRSkpKcnZpysjIkJub221bdkREhA4ePKiZM2eqQ4cO8vb21q5du7RgwQK1bNlSLVq0KPRyjTHKyspS+fKO/xXezm3JjZ+fX4mtCyhznHyLBAAFdOHCBSPJbNu2Ld9+R48eNZ06dTLu7u6mUaNGZtOmTUaS+fjjj40xxmzdutVIMhcuXLDPs3//fiPJnDp1yhhjTGJionnsscdMtWrVjKenp2nSpIn54IMPHNbTpUsXM3r0aDNu3Djj7+9vunbtaowx5n//+5/p2bOn8fLyMoGBgWbQoEHml19+sc+XkpJiBg8ebLy8vExwcLBZsGCB6dKlixk3blye2zR37lzj4uJivvvuuxzTMjIyTEpKijHGmLS0NDN27FhTpUoV4+7ubjp06GD27Nlj73t927/88kvTqlUr4+rqarZu3Vrkbbmx7nfeeceEhYWZihUrmqCgIDNw4ECTkJBgjMn9Hl/X75Nz43LOnz9vBg8ebHx9fY2np6fp2bOnOXr0qH16ZGSk8fHxMRs2bDANGzY0Xl5eJiIiwpw5cybP1xAoqzi1BFhExYoVVbFiRX3yySdKT0/PtU92drYeeughubm5affu3Vq2bJn+9Kc/FXpdaWlpCgsL0xdffKHvv/9eI0eO1ODBg7Vnzx6HfitXrpSbm5t27NihZcuW6eLFi+rWrZtatmypvXv3asOGDUpISNAjjzxin+e5557T9u3b9emnn2rTpk3atm2bvvvuu3zref/99xUeHq6WLVvmmObq6iovLy9J0vPPP69///vfWrlypb777jvVq1dPEREROn/+vMM8f/7znzVnzhxFR0erWbNmRd6WG2VmZmrmzJk6ePCgPvnkE/34448aOnSoJCkkJET//ve/JUkxMTE6e/asXn311VyXM3ToUO3du1fr1q3Tzp07ZYzRAw884HAK7cqVK1qwYIHeffdd/ec//1FsbKwmT56c7+sIlEnOTlIACm7t2rWmcuXKxsPDw9x7771mypQp5uDBg/bpGzduNOXLlzenT5+2t61fv77QR2Ry07t3bzNp0iT78y5dupiWLVs69Jk5c6bp0aOHQ1tcXJyRZGJiYszly5eNm5ubWb16tX16UlKS8fT0zPeIjKenp3n22WfznG7Mr0d6XF1dzfvvv29vy8jIMNWqVTPz5s0zxvzftn/yyScO8xZlW67Pl1/d3377rZFkLl++7LD+3772Ny7n6NGjRpLZsWOHfXpiYqLx9PS0v26RkZFGksMd2ZcsWWKCgoLyrAUoqzgiA1jIgAEDdObMGa1bt049e/bUtm3b1KpVK61YsUKSFB0drZCQEFWrVs0+T/v27Qu9nqysLM2cOVNNmzaVn5+fKlasqI0bNyo2NtahX1hYmMPzgwcPauvWrfajRxUrVlTDhg0lSSdOnNCJEyeUkZGhtm3b2ufx8/NTgwYN8q3HGHPTmk+cOKHMzEx16NDB3ubq6qo2bdooOjraoW/r1q1zzF/YbcnNvn371LdvX4WGhqpSpUrq0qWLJOV43fITHR2t8uXLO7xG/v7+atCggcN2VKhQQXXr1rU/r1q1qs6dO1fg9QBlBYN9AYvx8PDQ/fffr/vvv19Tp07VU089penTp9tPYdyMi8uvf7/8NhzceNXP/Pnz9eqrr2rx4sVq2rSpvLy8NH78eGVkZDj0u35K57qUlBT17dtXc+fOzbHeqlWr6vjx4wWq8UZ33XWXjhw5UqR5c3Nj3bm13WxbbpSamqqIiAhFRETo/fffV5UqVRQbG6uIiIgcr1txcHV1dXhus9kKFPiAsoYjMoDFNW7cWKmpqZKkRo0aKS4uTmfPnrVP37Vrl0P/KlWqSJJDnwMHDjj02bFjh/r166dBgwapefPmqlOnjo4ePXrTWlq1aqXDhw+rVq1aqlevnsPDy8tLdevWlaurq3bv3m2f58KFCzdd9uOPP66vvvpK+/fvzzEtMzNTqampqlu3rn2My2+nffvtt2rcuPFNay/sttzoyJEjSkpK0pw5c9SpUyc1bNgwxxGS61dCZWVl5bneRo0a6dq1aw6vUVJSkmJiYoq0HUBZR5ABLCIpKUndunXTe++9p0OHDunUqVNas2aN5s2bp379+kmSwsPDddddd2nIkCE6ePCgvv76a73wwgsOy6lXr55CQkL04osv6tixY/riiy+0cOFChz7169fX5s2b9c033yg6OlpPP/20EhISblrj6NGjdf78eQ0cOFDffvutTpw4oY0bN2rYsGHKyspSxYoVNXz4cD333HPasmWLvv/+ew0dOtR+lCgv48ePV4cOHdS9e3ctWbJEBw8e1MmTJ7V69Wq1a9dOx44dk5eXl5555hk999xz2rBhg3744QeNGDFCV65c0fDhwwv5at98W24UGhoqNzc3vf766zp58qTWrVuX47thatasKZvNps8//1y//PKLUlJSciynfv366tevn0aMGKH//ve/OnjwoAYNGqTq1avb9zOA/0OQASyiYsWKatu2rRYtWqTOnTurSZMmmjp1qkaMGKG///3vkn49bfTxxx/r6tWratOmjZ566in97W9/c1iOq6urPvzwQx05ckTNmjXT3LlzNWvWLIc+f/3rX9WqVStFRESoa9euCg4OVv/+/W9aY7Vq1bRjxw5lZWWpR48eatq0qcaPHy9fX197WJk/f746deqkvn37Kjw8XB07dswxPuVG7u7u2rx5s55//nm9+eabateune655x699tprevbZZ9WkSRNJ0pw5czRgwAANHjxYrVq10vHjx7Vx40ZVrly5oC9zobblt6pUqaIVK1ZozZo1aty4sebMmaMFCxY49KlevbpmzJihP//5zwoKCtKYMWNyXXdkZKTCwsLUp08ftW/fXsYYffnllzlOJwGQbIaTqkCZZ7PZ9PHHHxcojACAlXBEBgAAWBZBBgAAWBaXXwN3AM4gAyirOCIDAAAsiyADAAAsiyADAAAsiyADAAAsiyADAAAsiyADAAAsiyADAAAsiyADAAAsiyADAAAs6/8BPnq54mLXjYYAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Define the directory where the data is stored\n",
    "DATA_DIR = 'C:/Users/ABDULLAHI HAJI/OneDrive/Documents/Thesis_poject/GSE216738_RAW_COUNTS_Abnormal-AML-50bp'\n",
    "\n",
    "def load_data(filepath):\n",
    "    \"\"\"Load the data from a CSV file.\"\"\"\n",
    "    # Read the CSV file and print the first few rows\n",
    "    data = pd.read_csv(filepath, index_col=0)\n",
    "    print(data.head())\n",
    "    return data\n",
    "\n",
    "# Construct the full path to the CSV file\n",
    "filepath = os.path.join(DATA_DIR, 'normCounts_res.CSV')\n",
    "# Load the data\n",
    "data = load_data(filepath)\n",
    "# Get the number of genes (rows)\n",
    "original_dim = data.shape[0]  \n",
    "\n",
    "# Transpose data because the neural network expects samples as rows\n",
    "data_transposed = data.transpose()\n",
    "\n",
    "# Load the pre-trained autoencoder model\n",
    "loaded_model = load_model('autoencoder_model.keras')\n",
    "\n",
    "# Make predictions using the model\n",
    "predicted = loaded_model.predict(data_transposed)\n",
    "\n",
    "# Initialize a list to store the correlations and squared correlations\n",
    "correlations = []\n",
    "squared_correlations = []\n",
    "\n",
    "# Iterate over each row in the DataFrame\n",
    "for row in range(data_transposed.shape[0]):\n",
    "    # Get the original and predicted data for this row (feature)\n",
    "    original_data = data_transposed.iloc[row, :].values.flatten()\n",
    "    predicted_data = predicted[row, :].flatten()\n",
    "\n",
    "    # Check for NaN or infinite values and handle them\n",
    "    if np.any(np.isnan(original_data)) or np.any(np.isnan(predicted_data)):\n",
    "        print(f\"Skipping row {row} due to NaN values\")\n",
    "        continue\n",
    "    if np.any(np.isinf(original_data)) or np.any(np.isinf(predicted_data)):\n",
    "        print(f\"Skipping row {row} due to infinite values\")\n",
    "        continue\n",
    "\n",
    "    # Check for zero standard deviation and handle it\n",
    "    if np.std(original_data) == 0 or np.std(predicted_data) == 0:\n",
    "        print(f\"Skipping row {row} due to zero standard deviation\")\n",
    "        continue\n",
    "\n",
    "    # Compute the correlation for this row (feature)\n",
    "    correlation, _ = scipy.stats.pearsonr(original_data, predicted_data)  # Changed to Pearson correlation\n",
    "    correlations.append(correlation)\n",
    "    squared_correlations.append(correlation**2)  # Add the squared correlation\n",
    "\n",
    "# Print the number of correlations computed\n",
    "print(\"Number of correlations computed: \", len(correlations))\n",
    "\n",
    "# Compute the average correlation and average squared correlation\n",
    "average_correlation = np.mean(correlations)\n",
    "average_squared_correlation = np.mean(squared_correlations)\n",
    "\n",
    "# Print the average correlation and average squared correlation\n",
    "print(\"Average Correlation: \", average_correlation)\n",
    "print(\"Average Squared Correlation: \", average_squared_correlation)\n",
    "\n",
    "# Create a histogram of the correlations\n",
    "plt.hist(correlations, bins=50, edgecolor='black')\n",
    "plt.title('Histogram of Correlations')\n",
    "plt.xlabel('Correlation')\n",
    "plt.ylabel('Frequency')\n",
    "plt.show()\n",
    "\n",
    "# Create a histogram of the squared correlations\n",
    "plt.hist(squared_correlations, bins=50, edgecolor='black')\n",
    "plt.title('Histogram of Squared Correlations')\n",
    "plt.xlabel('Squared Correlation')\n",
    "plt.ylabel('Frequency')\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "befaea97-2e9c-49c5-8495-5eb306c882bb",
   "metadata": {},
   "source": [
    "# Identifying Top 500 Genes from Combined Weights Matrix ( from the AE model)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "75a6b04e-f008-46f8-b2a8-98389876a58f",
   "metadata": {},
   "source": [
    "In his step we identify the top 500 genes based on their importance scores. we begins by loading a combined weights matrix from a specified file path. Each gene's importance score is calculated by summing the absolute values of its weights across different features. The genes are then sorted in descending order of their importance scores. The top 500 genes are selected and optionally saved to a new CSV file for further analysis. Finally, the names and scores of the top 500 genes are printed."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "766bc002-1030-4bd6-8932-bb2bf42556c2",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ENSG00000265369    1.489900\n",
      "ENSG00000164125    1.463906\n",
      "ENSG00000153560    1.440316\n",
      "ENSG00000000938    1.418589\n",
      "ENSG00000136521    1.418577\n",
      "                     ...   \n",
      "ENSG00000266049    1.220805\n",
      "ENSG00000140464    1.220662\n",
      "ENSG00000131023    1.220605\n",
      "ENSG00000109390    1.220520\n",
      "ENSG00000181163    1.220266\n",
      "Length: 500, dtype: float64\n"
     ]
    }
   ],
   "source": [
    "# Load the combined weights matrix\n",
    "file_path = 'C:/Users/ABDULLAHI HAJI/combined_transformation_matrix_with_genes.csv'\n",
    "weights_df = pd.read_csv(file_path, index_col=0)  # The first column contains gene names\n",
    "\n",
    "# Calculate an overall importance score for each gene\n",
    "# Here, I'm using the sum of weights across different features .\n",
    "gene_importance = weights_df.abs().sum(axis=1)\n",
    "\n",
    "# Sort genes by their importance score in descending order\n",
    "sorted_genes = gene_importance.sort_values(ascending=False)\n",
    "\n",
    "# Select the top 500 genes\n",
    "top_500_genes = sorted_genes.head(500)\n",
    "\n",
    "# Optionally, save the top 500 genes to a new CSV file for further analysis\n",
    "top_500_genes.to_csv('top_500_genes_from_autoencoder.csv')\n",
    "\n",
    "# Print or output the top 500 genes\n",
    "print(top_500_genes)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "655c2310-ef4a-4439-8de5-2156e5c47c34",
   "metadata": {},
   "source": [
    "# PCA Implementation "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cf755715-6d40-4a52-bc2b-d1107f1033b0",
   "metadata": {},
   "source": [
    "Next step was  performing Principal Component Analysis (PCA) on a dataset to identify the top 500 genes based on their contributions to each principal component. The data is loaded, transposed, and standardized before PCA is applied.Once the data was appropriately preprocessed, we applied PCA. The number of principal components to retain was determined by examining a scree plot.Based on the scree plot, we decided to focus on the first four principal components.We then calculated the combined loadings of each gene across these four principal components. Loadings represent the weights of each gene in the principal components and can be interpreted as the correlation between the genes and the principal components. By combining the loadings from the first four principal components, we were able to capture the genes that contribute most significantly across all four components.\n",
    "\n",
    "From this analysis, we selected the top 500 genes with the highest combined loadings. These genes are considered to be the most influential in our dataset based on the first four principal components. For further analysis and potential future reference, we saved these top 500 genes to a CSV file.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "f8f3ddd2-e14b-462b-a863-123f17857697",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 800x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "PC1 explains 10.98% of the variance.\n",
      "PC2 explains 9.50% of the variance.\n",
      "PC3 explains 7.85% of the variance.\n",
      "PC4 explains 4.95% of the variance.\n",
      "PC5 explains 4.57% of the variance.\n",
      "PC6 explains 3.58% of the variance.\n",
      "PC7 explains 3.17% of the variance.\n",
      "PC8 explains 2.29% of the variance.\n",
      "PC9 explains 2.21% of the variance.\n",
      "PC10 explains 1.90% of the variance.\n",
      "                PC1       PC2       PC3           PC4       PC5           PC6  \\\n",
      "0      1.017973e-05  0.000078  0.000248  8.299329e-07  0.000059  1.468086e-05   \n",
      "1      4.622399e-06  0.000030  0.000156  6.003047e-05  0.000259  3.743201e-06   \n",
      "2      1.525806e-04  0.000001  0.000129  1.381657e-05  0.000001  4.004638e-05   \n",
      "3      5.450408e-07  0.000338  0.000075  6.574233e-06  0.000119  3.220481e-05   \n",
      "4      1.599617e-05  0.000002  0.000016  4.288254e-05  0.000003  7.150155e-09   \n",
      "...             ...       ...       ...           ...       ...           ...   \n",
      "17254  3.117600e-05  0.000094  0.000035  6.217472e-06  0.000017  5.808498e-04   \n",
      "17255  4.507403e-05  0.000101  0.000011  1.199072e-08  0.000096  7.942625e-05   \n",
      "17256  9.945392e-05  0.000046  0.000062  4.804842e-05  0.000068  1.831544e-05   \n",
      "17257  7.471705e-05  0.000008  0.000036  4.443824e-08  0.000142  5.886024e-06   \n",
      "17258  1.512989e-05  0.000255  0.000023  1.932165e-05  0.000009  6.538486e-06   \n",
      "\n",
      "            PC7           PC8           PC9          PC10  \n",
      "0      0.000021  3.171085e-05  1.338162e-06  1.767508e-05  \n",
      "1      0.000231  4.999449e-05  7.018441e-05  1.793663e-05  \n",
      "2      0.000103  9.606334e-06  1.411864e-04  2.196596e-05  \n",
      "3      0.000008  2.168574e-07  1.404425e-05  2.101577e-05  \n",
      "4      0.000193  4.468968e-05  2.464049e-05  7.552295e-05  \n",
      "...         ...           ...           ...           ...  \n",
      "17254  0.000006  9.871690e-05  1.417685e-05  2.966430e-06  \n",
      "17255  0.000050  7.781191e-05  1.761409e-06  8.061133e-07  \n",
      "17256  0.000006  3.154542e-06  4.030687e-05  5.409266e-07  \n",
      "17257  0.000005  8.595606e-05  2.419322e-05  2.076504e-06  \n",
      "17258  0.000021  1.251845e-05  5.452505e-07  8.069559e-05  \n",
      "\n",
      "[17259 rows x 10 columns]\n",
      "                 Combined Loadings\n",
      "ENSG00000115084           0.045162\n",
      "ENSG00000267198           0.044767\n",
      "ENSG00000204130           0.044636\n",
      "ENSG00000260744           0.043142\n",
      "ENSG00000264853           0.043112\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1200x1000 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Load the data\n",
    "data_path = 'C:/Users/ABDULLAHI HAJI/OneDrive/Documents/Thesis_poject/GSE216738_RAW_COUNTS_Abnormal-AML-50bp/normCounts_res.CSV'\n",
    "data = pd.read_csv(data_path, index_col=0)\n",
    "\n",
    "# Transpose the data so that genes are columns and samples are rows\n",
    "data_transposed = data.T\n",
    "\n",
    "# Standardize the data\n",
    "scaler = StandardScaler()\n",
    "scaled_data = scaler.fit_transform(data_transposed)\n",
    "\n",
    "# Perform PCA\n",
    "pca = PCA(n_components=10)\n",
    "principalComponents = pca.fit_transform(scaled_data)\n",
    "loadings = pca.components_.T\n",
    "\n",
    "# Convert to DataFrame for easier handling\n",
    "principalDf = pd.DataFrame(data=principalComponents,\n",
    "                           columns=['PC' + str(i) for i in range(1, 11)])\n",
    "\n",
    "# Calculate eigenvalues\n",
    "eigenvalues = pca.explained_variance_\n",
    "\n",
    "# Create a Scree plot\n",
    "plt.figure(figsize=(8, 6))\n",
    "plt.plot(range(1, len(eigenvalues) + 1), eigenvalues, 'ro-', linewidth=2)\n",
    "plt.title('Scree Plot')\n",
    "plt.xlabel('Principal Component')\n",
    "plt.ylabel('Eigenvalue')\n",
    "plt.axhline(y=1, color='r', linestyle='--')\n",
    "plt.savefig('Scree plot.png')\n",
    "plt.show()\n",
    "\n",
    "# Specify the number of top contributing genes to display in the loading plots\n",
    "num_top_genes = 1000\n",
    "\n",
    "# Print the top 1000 contributing genes for each principal component\n",
    "for i, component in enumerate(pca.components_):\n",
    "    sorted_indices = np.argsort(np.abs(component))[::-1]\n",
    "    top_feature_indices = sorted_indices[:num_top_genes]\n",
    "    top_features = [data_transposed.columns[j] for j in top_feature_indices]\n",
    "    \"\"\"print(f\"Top 10 contributing genes for PC{i + 1}: {top_features}\")\"\"\"\n",
    "\n",
    "# Calculate and print the explained variance ratio for each principal component\n",
    "explained_variance_ratio = pca.explained_variance_ratio_\n",
    "for i, ratio in enumerate(explained_variance_ratio):\n",
    "    print(f\"PC{i + 1} explains {ratio * 100:.2f}% of the variance.\")\n",
    "\n",
    "\n",
    "# Square the loadings\n",
    "squared_loadings = np.square(loadings)\n",
    "\n",
    "# Create a DataFrame for squared loadings\n",
    "squared_loadings_df = pd.DataFrame(squared_loadings, columns=['PC' + str(i) for i in range(1, 11)])\n",
    "\n",
    "# Print the DataFrame\n",
    "print(squared_loadings_df)\n",
    "\n",
    "# Combine the loadings from PCA1, PCA2,  PCA3 and PCA4\n",
    "combined_loadings = np.abs(loadings[:, 0]) + np.abs(loadings[:, 1]) + np.abs(loadings[:, 2] + np.abs(loadings[:, 3]))\n",
    "\n",
    "# Create a DataFrame for easier handling\n",
    "loadings_df = pd.DataFrame(combined_loadings, index=data_transposed.columns, columns=['Combined Loadings'])\n",
    "\n",
    "#print(loadings_df.head())\n",
    "# Sort the DataFrame based on the combined loadings\n",
    "sorted_loadings_df = loadings_df.sort_values(by='Combined Loadings', ascending=False)\n",
    "\n",
    "# Get the top 500 most influential genes\n",
    "top_500_genes = sorted_loadings_df.head(500)\n",
    "\n",
    "print (top_500_genes.head())\n",
    "\n",
    "# Save the DataFrame to a CSV file\n",
    "top_500_genes.to_csv('top_500_genes_from_PCA.csv')\n",
    "\n",
    "\n",
    "# Create a histogram of the loadings for each principal component\n",
    "for i in range(pca.n_components_):\n",
    "    plt.figure(figsize=(10, 6))\n",
    "    plt.hist(loadings[:, i], bins=50, edgecolor='black')\n",
    "    plt.title(f'Histogram of Loadings of PC{i + 1}')\n",
    "    plt.xlabel('Loading')\n",
    "    plt.ylabel('Frequency')\n",
    "    plt.show()\n",
    "\n",
    "# Generate a correlation matrix plot for loadings\n",
    "correlation_matrix = np.corrcoef(loadings.T)\n",
    "plt.figure(figsize=(12, 10))\n",
    "sns.heatmap(correlation_matrix, annot=True, cmap='coolwarm', fmt='.2f')\n",
    "plt.title('Correlation Matrix of Loadings')\n",
    "plt.xlabel('Principal Components')\n",
    "plt.ylabel('Principal Components')\n",
    "plt.show()\n",
    "\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.10.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}