{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# PBMC inter-dataset\n", "\n", "Warning: vignette for scHPL v. 0.0.2, this should be updated" ] }, { "cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [], "source": [ "import os\n", "import pandas as pd\n", "import numpy as np\n", "import time as tm\n", "from scHPL import progressive_learning, predict, evaluate" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "During this vignette we will repeat the PBMC inter-dataset experiment. We use three datasets to construct a classification tree and use this tree to predict the labels of a fourth dataset. The aligned datasets and labels can be downloaded from https://doi.org/10.5281/zenodo.4557712" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Read the data\n", "\n", "We start with reading the different datasets, corresponding labels and them to a list.\n", "\n", "In the datasets, the rows represent different cells and columns represent the genes" ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [], "source": [ "data0 = 'Data_EQTL.csv'\n", "labels0 = 'Labels_EQTL.csv'\n", "\n", "data1 = 'Data_10Xv2.csv'\n", "labels1 = 'Labels_10Xv2.csv'\n", "\n", "data2 = 'Data_FACS.csv'\n", "labels2 = 'Labels_FACS.csv'\n", "\n", "data = []\n", "labels = []\n", "\n", "data.append((pd.read_csv(data0, index_col=0, sep=',')))\n", "labels.append(pd.read_csv(labels0, header=0, index_col=None, sep=','))\n", "\n", "data.append((pd.read_csv(data1, index_col=0, sep=',')))\n", "labels.append(pd.read_csv(labels1, header=0, index_col=None, sep=','))\n", "\n", "data.append((pd.read_csv(data2, index_col=0, sep=',')))\n", "labels.append(pd.read_csv(labels2, header=0, index_col=None, sep=','))" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Construct and train the classification tree\n", "\n", "Next, we use hierarchical progressive learning to construct and train a classification tree. After each iteration, an updated tree will be printed. If two labels have a perfect match, one of the labels will not be visible in the tree. Therefore, we will also indicate these perfect matches using a print statement\n", "\n", "During this experiment, we used the linear SVM, didn't apply dimensionality reduction and used the default threshold of 0.25. In you want to use a one-class SVM instead of a linear, the following can be used: classifier = 'svm_occ'" ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Iteration 1 \n", "\n", "Perfect match: B cell - B-10Xv2 is now: B cell - eQTL\n", "Perfect match: Megakaryocyte - B-10Xv2 is now: Megakaryocyte - eQTL\n", "Perfect match: CD4+ T cell - B-10Xv2 is now: CD4+ T cell - eQTL\n", "Perfect match: CD8+ T cell - B-10Xv2 is now: CD8+ T cell - eQTL\n", "\n", "Updated tree:\n", "root\n", "\tB cell - eQTL\n", "\tCD4+ T cell - eQTL\n", "\tCD8+ T cell - eQTL\n", "\tMegakaryocyte - eQTL\n", "\tpDC - eQTL\n", "\tMonocyte - B-10Xv2\n", "\t\tCD14+ Monocyte - eQTL\n", "\t\tCD16+ Monocyte - eQTL\n", "\t\tmDC - eQTL\n", "\tNK cell - B-10Xv2\n", "\t\tCD56+ bright NK cell - eQTL\n", "\t\tCD56+ dim NK cell - eQTL\n", "Iteration 2 \n", "\n", "Perfect match: B cell - FACS is now: B cell - eQTL\n", "Perfect match: CD14+ Monocyte - FACS is now: Monocyte - B-10Xv2\n", "Perfect match: CD34+ cell - FACS is now: pDC - eQTL\n", "\n", "Updated tree:\n", "root\n", "\tB cell - eQTL\n", "\tCD4+ T cell - eQTL\n", "\t\tCD4+/CD25 T Reg - FACS\n", "\t\tCD4+/CD45RA+/CD25- Naive T - FACS\n", "\t\tCD4+/CD45RO+ Memory - FACS\n", "\t\tCD8+/CD45RA+ Naive Cytotoxic - FACS\n", "\tMegakaryocyte - eQTL\n", "\tpDC - eQTL\n", "\tMonocyte - B-10Xv2\n", "\t\tCD14+ Monocyte - eQTL\n", "\t\tCD16+ Monocyte - eQTL\n", "\t\tmDC - eQTL\n", "\tNK cell - FACS\n", "\t\tCD8+ T cell - eQTL\n", "\t\tNK cell - B-10Xv2\n", "\t\t\tCD56+ bright NK cell - eQTL\n", "\t\t\tCD56+ dim NK cell - eQTL\n", "Time to train scHPL on a normal desktop: 632.8667893409729\n" ] } ], "source": [ "start = tm.time()\n", "\n", "classifier = 'svm'\n", "dimred = False\n", "threshold = 0.25\n", "tree = progressive_learning.learn_tree(data, labels, \n", " classifier = classifier, \n", " dimred = dimred, \n", " threshold = threshold)\n", "\n", "training_time = tm.time()-start\n", "print('Time to train scHPL on a normal desktop:', training_time)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Predict the labels of the fourth dataset\n", "\n", "In this last step, we use the learned tree to predict the labels of another dataset" ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Time to make predictions with scHPL on a normal desktop: 4.310481309890747\n" ] } ], "source": [ "data3 = 'Data_10Xv3.csv'\n", "data_test = pd.read_csv(data3, index_col=0, sep=',')\n", "\n", "start = tm.time()\n", "pred_test = predict.predict_labels(data_test, tree)\n", "pred_time = tm.time()-start\n", "print('Time to make predictions with scHPL on a normal desktop:', pred_time)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "We compare these true and predicted labels by constructing a confusion matrix" ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [], "source": [ "labels3 = 'Labels_10Xv3.csv'\n", "y_true = pd.read_csv(labels3, header=0, index_col=None, sep=',')\n", "y_pred = pd.DataFrame(data = pred_test)\n", "\n", "confmatrix = evaluate.confusion_matrix(y_true, y_pred)\n", "confmatrix = confmatrix / np.sum(confmatrix.values, axis = 1, keepdims=True) #Normalize" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "This confusion matrix can be visualized using a heatmap.\n", "\n", "In this heatmap, we notice that the predictions of the CD16+ Monocytes and mDC are switched, which is caused by the mislabeling of the cells in the eQTL dataset." ] }, { "cell_type": "code", "execution_count": 7, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "import seaborn as sns\n", "import matplotlib.pyplot as plt\n", "\n", "plt.figure(figsize=(12,4.5))\n", "sns.heatmap(round(confmatrix,2), vmin = 0, vmax = 1, annot=True)\n", "plt.show()" ] } ], "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": "3.8.3" } }, "nbformat": 4, "nbformat_minor": 4 }