{ "cells": [ { "cell_type": "markdown", "metadata": { "id": "hVDoD2OiwPvJ" }, "source": [ "# Gromov-Wasserstein for multi-omics" ] }, { "cell_type": "markdown", "metadata": { "id": "BB8VjJrVsuuG" }, "source": [ "A [variety of single-cell measurements](https://en.wikipedia.org/wiki/Single-cell_analysis) can help explore cell characteristics that are helpful to understand biological mechanisms. These measurements can for instance [describe epigenetic changes](https://en.wikipedia.org/wiki/Single_cell_epigenomics) (DNA methylation, chromatin accessibility, histone modifications, chromosome conformation), the genome itself, as well as the proteins present in the cell ([single cell sequencing](https://en.wikipedia.org/wiki/Single_cell_sequencing). However, performing measures of different natures rises a major challenge: that of establishing an alignment across two (possibly several) measurement spaces that are unrelated, in the sense that no biological-based theory allows to construct such correspondences between them. In the absence of supervised information, the alignment can be constructed from first-hand principles, such as that of preserving geometry (i.e. an isomorphism) between the two measurement spaces. Indeed, since the population of cells measured is (statistically) the same across measurements, we expect that cells with similar genomes will be mapped to cells with similar transcriptomes, proteomes and epigenetic changes. \n", "\n", "The Gromov-Wasserstein optimal transport framework, implemented in OTT, is a relevant candidate to perform such an unsupervised cell alignment. \n", "This approach was proposed by {cite}`demetci:22`, who called it SCOT ([GitHub repo](https://github.com/rsinghlab/SCOT)), from which this notebook is adapted.\n", "\n", "The original SCOT code code uses [Python Optimal Transport (POT)](https://pythonot.github.io/). Here, we propose a slight modification of the SCOT code to use the {class}`~ott.solvers.quadratic.gromov_wasserstein.GromovWasserstein` solver, which we have found to be faster than the POT implementation of {func}`~ot.gromov.entropic_gromov_wasserstein` on GPU (see [](#alignment-and-evaluation)). We then use this OTT version of the SCOT algorithm to perform cell alignment for the SNARE-seq dataset {cite}`chen:19`, which contains measures of two natures:\n", "\n", " - Chromatin accessibility ([scATAC-seq data](https://en.wikipedia.org/wiki/ATAC-seq))\n", " - Gene expression ([scRNA-seq data](https://en.wikipedia.org/wiki/Single_cell_sequencing#scRNA-Seq))" ] }, { "cell_type": "markdown", "metadata": { "id": "BYAvtCwhsuuJ" }, "source": [ "## Imports and dataset loading " ] }, { "cell_type": "markdown", "metadata": { "id": "pJXAdeCCsuuJ" }, "source": [ "We must clone the SCOT repo within the folder that contains this notebook. For later access to data present in the cloned repo, only relative paths are used." ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "import sys\n", "\n", "if \"google.colab\" in sys.modules:\n", " !pip install -q git+https://github.com/ott-jax/ott@main\n", " !pip install -q seaborn\n", " !git clone -q https://github.com/rsinghlab/SCOT\n", " !pip install -r SCOT/src/requirements.txt" ] }, { "cell_type": "code", "execution_count": 1, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "5j2OA14TsuuL", "outputId": "e968bcf1-c15e-4887-d571-b3a83a35d273" }, "outputs": [], "source": [ "import time\n", "\n", "from IPython import display\n", "from ot.gromov import gwloss, init_matrix\n", "from SCOT.src import evals\n", "from SCOT.src.scot import SCOT\n", "\n", "import numpy as np\n", "import pandas as pd\n", "from sklearn.decomposition import PCA\n", "\n", "import matplotlib.pyplot as plt\n", "import seaborn as sn\n", "from matplotlib import animation\n", "\n", "import ott\n", "from ott.problems.quadratic import quadratic_problem\n", "from ott.solvers.quadratic import gromov_wasserstein" ] }, { "cell_type": "code", "execution_count": 2, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "PqR1ju-rwPvV", "outputId": "18bb0f44-442a-4761-fdc3-fdf875a3e0eb" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Dimensions of input datasets:\n", "X = (1047, 19) => ie 1047 samples belonging to a chromatin accessibility feature space of dimension 19\n", "y = (1047, 10) => ie 1047 samples belonging to a gene expression feature space of dimension 10\n" ] } ], "source": [ "X = np.load(\"SCOT/data/scatac_feat.npy\")\n", "y = np.load(\"SCOT/data/scrna_feat.npy\")\n", "\n", "print(\"Dimensions of input datasets:\")\n", "print(\n", " \"X =\",\n", " X.shape,\n", " \"=> ie\",\n", " X.shape[0],\n", " \"samples belonging to a chromatin accessibility feature space of dimension\",\n", " X.shape[1],\n", ")\n", "print(\n", " \"y =\",\n", " y.shape,\n", " \"=> ie\",\n", " y.shape[0],\n", " \"samples belonging to a gene expression feature space of dimension\",\n", " y.shape[1],\n", ")" ] }, { "cell_type": "markdown", "metadata": { "id": "G8nC8PpHsuuS" }, "source": [ "## Using `gromov_wasserstein` from OTT" ] }, { "cell_type": "markdown", "metadata": { "id": "m9uzbXpZsuuT" }, "source": [ "The following `OTTSCOT` class inherits from the `SCOT` class but overrides the `find_correspondences` method in order to use OTT instead of POT. The matrix `T` is the optimal transport matrix, mapping $X$ to $y$." ] }, { "cell_type": "code", "execution_count": 5, "metadata": { "id": "NM0r8q7DKXUJ" }, "outputs": [], "source": [ "class OTTSCOT(SCOT):\n", " def find_correspondences(self, e: float, verbose: bool = True) -> None:\n", " geom_xx = ott.geometry.Geometry(self.Cx)\n", " geom_yy = ott.geometry.Geometry(self.Cy)\n", " prob = quadratic_problem.QuadraticProblem(\n", " geom_xx, geom_yy, a=self.p, b=self.q\n", " )\n", "\n", " solver = gromov_wasserstein.GromovWasserstein(\n", " epsilon=e, threshold=1e-9, max_iterations=1000\n", " )\n", "\n", " T = solver(prob).matrix\n", "\n", " constC, hC1, hC2 = init_matrix(\n", " self.Cx, self.Cy, self.p, self.q, loss_fun=\"square_loss\"\n", " )\n", " self.gwdist = gwloss(constC, hC1, hC2, np.array(T))\n", " self.coupling = T\n", "\n", " if (\n", " np.isnan(self.coupling).any()\n", " or np.any(~self.coupling.any(axis=1))\n", " or np.any(~self.coupling.any(axis=0))\n", " or sum(sum(self.coupling)) < 0.95\n", " ):\n", " self.flag = False\n", " else:\n", " self.flag = True" ] }, { "cell_type": "markdown", "metadata": { "id": "brfhPBHWsuuV" }, "source": [ "In the Gromov-Wassertein optimal transport, we have two hyperparameters to tune : \n", "- $\\epsilon$, which controls entropy in the regularized optimization problem \n", "- $k$, that is the number of nearest neighbours that we will consider for building the initial graphs that will allow us to define a notion of closeness between points from the same domain\n", "\n", "The `SCOT` class implements an unsupervized hyperparmeter search method which for our `OTTSCOT` outputs optimal values of `1e-3` for $\\epsilon$ and `40` for the number of neighbors $k$ . We will retain these values for the rest of the exploration of the SNARE-seq dataset:" ] }, { "cell_type": "code", "execution_count": 6, "metadata": { "id": "_QR1q7HWsuuW" }, "outputs": [], "source": [ "k = 40\n", "epsilon = 1e-3" ] }, { "cell_type": "markdown", "metadata": { "id": "KtvRj6RosuuY" }, "source": [ "Besides, note that for a fair comparison between `POT` and `OTT`, we have set the maximum number of iterations for OTT's Gromov to be equal to `1000` and the error threshold at `1e-9`, the default values of the POT version of Gromov used in SCOT." ] }, { "cell_type": "markdown", "metadata": { "id": "gmS0MXbwsuuZ" }, "source": [ "## Alignment and evaluation" ] }, { "cell_type": "markdown", "metadata": { "id": "ZbiuslH0suua" }, "source": [ "We now perform the alignment for our dataset and evaluate its execution time (on Colab, using a Tesla P100 GPU):" ] }, { "cell_type": "code", "execution_count": 10, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "sR_cmUAOzqgo", "outputId": "28928890-5df2-4e32-a663-373a72e83373" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "GPU 0: Tesla P100-PCIE-16GB (UUID: GPU-b22fbca2-02fb-5ce3-8143-b4aafdc5de09)\n" ] } ], "source": [ "!nvidia-smi -L" ] }, { "cell_type": "code", "execution_count": 7, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "medFuBx-suub", "outputId": "96600959-48b9-497c-f17b-c00c1562c85c" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Execution time: 18.29 s\n" ] } ], "source": [ "ottscot = OTTSCOT(X, y)\n", "\n", "start = time.time()\n", "X_shifted, y_shifted = ottscot.align(\n", " k=k, e=epsilon, normalize=True, norm=\"l2\", verbose=False\n", ") # OTT\n", "end = time.time()\n", "\n", "print(\"Execution time: \", round(end - start, 2), \"s\")" ] }, { "cell_type": "markdown", "metadata": { "id": "LNp6t08Ysuuc" }, "source": [ "For comparison purposes, we also evaluate excecution time for the original SCOT algorithm using POT (on the same GPU):" ] }, { "cell_type": "code", "execution_count": 8, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "vwj64WVmsuud", "outputId": "69fbd6ba-b367-494c-b7e9-3881bd882131" }, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "/usr/local/lib/python3.7/dist-packages/ot/bregman.py:517: UserWarning: Sinkhorn did not converge. You might want to increase the number of iterations `numItermax` or the regularization parameter `reg`.\n", " warnings.warn(\"Sinkhorn did not converge. You might want to \"\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Execution time: 145.98 s\n" ] } ], "source": [ "potscot = SCOT(X, y)\n", "\n", "start = time.time()\n", "X_shifted_pot, y_shifted_pot = potscot.align(\n", " k=k, e=epsilon, normalize=True, norm=\"l2\", verbose=False\n", ") # POT\n", "end = time.time()\n", "\n", "print(\"Execution time: \", round(end - start, 2), \"s\")" ] }, { "cell_type": "markdown", "metadata": { "id": "Lo-CPjPWsuue" }, "source": [ "For this run on GPU, OTT's {class}`~ott.solvers.quadratic.gromov_wasserstein.GromovWasserstein` solver is about 8 times faster than POT's solver. OTT's GW greatly benefits from parallelization, so its advantage over POT's GW should also be tested on CPU." ] }, { "cell_type": "markdown", "metadata": { "id": "kWJSgPtCsuuf" }, "source": [ "\n", "We are provided with a ground truth alignment, since we have the identity of each cell for the two domains. The fraction of samples closer than the true match (FOSCTTM) is a metric used in SCOT that leverages this supervision information to evaluate alignments (higher is better). We can use it to further compare alignments obtained with OTT and POT :" ] }, { "cell_type": "code", "execution_count": 19, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "fX2d-UEpsuug", "outputId": "a0b9f13d-606f-4928-8a12-976ef7d99c93" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "The average FOSCTTM for the alignment of X (chromatine accessibility domain) on y (gene expression domain) is: 0.2255\n" ] } ], "source": [ "fractions = evals.calc_domainAveraged_FOSCTTM(X_shifted, y_shifted) # OTT\n", "print(\n", " \"The average FOSCTTM for the alignment of X (chromatine accessibility domain)\"\n", " \" on y (gene expression domain) is: \",\n", " np.mean(fractions).round(4),\n", ")" ] }, { "cell_type": "code", "execution_count": 20, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "7m1aI_Fvsuuh", "outputId": "f5339535-ef26-4df8-e79e-b0750572cc88" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "The average FOSCTTM for the alignment of X (chromatine accessibility domain) on y (gene expression domain) is: 0.2247\n" ] } ], "source": [ "fractions_pot = evals.calc_domainAveraged_FOSCTTM(\n", " X_shifted_pot, y_shifted_pot\n", ") # POT\n", "print(\n", " \"The average FOSCTTM for the alignment of X (chromatine accessibility domain)\"\n", " \" on y (gene expression domain) is: \",\n", " np.mean(fractions_pot).round(4),\n", ")" ] }, { "cell_type": "markdown", "metadata": { "id": "UtcqjKXC26uG" }, "source": [ "FOSCTTM are very close for the two alignements, and the OTT version is even slightly better." ] }, { "cell_type": "markdown", "metadata": { "id": "svoMyh1dwPvg" }, "source": [ "## Visualisations" ] }, { "cell_type": "markdown", "metadata": { "id": "wiYuwaeKsuui" }, "source": [ "We begin by taking a look at the 2-Principal Components Analyses of both domains :" ] }, { "cell_type": "code", "execution_count": 21, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 436 }, "id": "leC1mTscsuuj", "outputId": "77a46670-dd2f-4a02-ce56-4cdee53f10dd" }, "outputs": [ { "data": { "text/plain": [ "Text(0.5, 1.0, 'PCA of gene expression before alignment, \\n colored according to cell type')" ] }, "execution_count": 21, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "cellTypes_atac = np.loadtxt(\"SCOT/data/SNAREseq_atac_types.txt\")\n", "cellTypes_rna = np.loadtxt(\"SCOT/data/SNAREseq_rna_types.txt\")\n", "\n", "fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(12, 6))\n", "\n", "pca = PCA(n_components=2)\n", "X_pca = pca.fit_transform(ottscot.X)\n", "\n", "cell_types = list(set(cellTypes_atac))\n", "cell_types_names = [\"H1\", \"GM\", \"BJ\", \"K562\"]\n", "colors = [\"blue\", \"purple\", \"red\", \"green\"]\n", "\n", "df1 = pd.DataFrame(\n", " {\n", " \"x\": np.flip(X_pca[:, 0]),\n", " \"y\": np.flip(X_pca[:, 1]),\n", " \"cellTypes\": np.flip(\n", " [cell_types_names[int(type) - 1] for type in cellTypes_atac]\n", " ),\n", " }\n", ")\n", "\n", "sn.scatterplot(\n", " data=df1,\n", " x=\"x\",\n", " y=\"y\",\n", " hue=\"cellTypes\",\n", " s=45,\n", " alpha=0.6,\n", " edgecolors=\"none\",\n", " ax=ax1,\n", ")\n", "ax1.legend()\n", "ax1.set_title(\n", " \"PCA of chromatin accessibility before alignment, \\n colored according to cell type\"\n", ")\n", "\n", "pca = PCA(n_components=2)\n", "y_pca = pca.fit_transform(ottscot.y)\n", "df1 = pd.DataFrame(\n", " {\n", " \"x\": np.flip(y_pca[:, 0]),\n", " \"y\": np.flip(y_pca[:, 1]),\n", " \"cellTypes\": np.flip(\n", " [cell_types_names[int(type) - 1] for type in cellTypes_rna]\n", " ),\n", " }\n", ")\n", "\n", "sn.scatterplot(\n", " data=df1,\n", " x=\"x\",\n", " y=\"y\",\n", " hue=\"cellTypes\",\n", " s=45,\n", " alpha=0.6,\n", " edgecolors=\"none\",\n", " ax=ax2,\n", ")\n", "\n", "ax2.legend()\n", "ax2.set_title(\n", " \"PCA of gene expression before alignment, \\n colored according to cell type\"\n", ")" ] }, { "cell_type": "markdown", "metadata": { "id": "ZruK0o4Gsuul" }, "source": [ "We visualize the superposition of chromatin accessibility points mapped to gene expression domain to the original points cloud of gene expression data :" ] }, { "cell_type": "code", "execution_count": 22, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 744 }, "id": "HSJUb32Ksuun", "outputId": "b998ca17-4bc7-48c9-8cda-fe53825507e9" }, "outputs": [ { "data": { "text/html": [ "\n", "\n", "\n", "\n", "\n", "\n", "
\n", " \n", "
\n", " \n", "
\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
\n", "
\n", " \n", " \n", " \n", " \n", " \n", " \n", "
\n", "
\n", "
\n", "\n", "\n", "\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "fig = plt.figure(figsize=(9, 9))\n", "(line,) = plt.plot([], [])\n", "\n", "n_samples = len(X)\n", "pca = PCA(n_components=2)\n", "\n", "Xy_pca = pca.fit_transform(np.concatenate((X_shifted, y_shifted), axis=0))\n", "\n", "cell_types = list(set(cellTypes_atac))\n", "cell_types_names = [\"H1\", \"GM\", \"BJ\", \"K562\"]\n", "cellTypes_atac_rna = np.concatenate(\n", " (\n", " [cell_types_names[int(type) - 1] for type in cellTypes_atac],\n", " [cell_types_names[int(type) - 1] for type in cellTypes_rna],\n", " ),\n", " axis=0,\n", ")\n", "original_domain_type = np.concatenate(\n", " (\n", " np.full(n_samples, \"Chromatin accessibility\"),\n", " np.full(n_samples, \"Gene expression\"),\n", " ),\n", " axis=0,\n", ")\n", "\n", "df = pd.DataFrame(\n", " {\n", " \"x\": np.flip(Xy_pca[:, 0]),\n", " \"y\": np.flip(Xy_pca[:, 1]),\n", " \"cellTypes\": np.flip(cellTypes_atac_rna),\n", " \"original_domain\": np.flip(original_domain_type),\n", " }\n", ")\n", "\n", "\n", "def animate(i):\n", " plt.clf()\n", " if i == 0:\n", " sn.scatterplot(\n", " data=df1,\n", " x=\"x\",\n", " y=\"y\",\n", " hue=\"cellTypes\",\n", " s=70,\n", " alpha=0.6,\n", " edgecolors=\"none\",\n", " )\n", " plt.title(\n", " \"PCA of gene expression before alignment, \\n colored according to cell type\"\n", " )\n", " else:\n", " sn.scatterplot(\n", " data=df,\n", " x=\"x\",\n", " y=\"y\",\n", " hue=\"cellTypes\",\n", " s=70,\n", " style=\"original_domain\",\n", " alpha=0.6,\n", " edgecolors=\"none\",\n", " )\n", " plt.title(\n", " \"PCA of chromatin accessibility points mapped to gene expression domain,\\n\"\n", " \"along with original gene expression points,\\n\"\n", " \"colored according to cell type\"\n", " )\n", " return (line,)\n", "\n", "\n", "def init():\n", " line.set_data([], [])\n", " return (line,)\n", "\n", "\n", "anim = animation.FuncAnimation(\n", " fig,\n", " animate,\n", " init_func=init,\n", " frames=[0, 1],\n", " interval=1500,\n", " blit=True,\n", ")\n", "\n", "html = display.HTML(anim.to_jshtml())\n", "display.display(html)\n", "plt.close()" ] }, { "cell_type": "markdown", "metadata": { "id": "7y-DPRLCsuuo" }, "source": [ "We can perform many more visualisations with animated plots. An example provided below explores the visual evolution of the optimal transport when we vary the hyperparameter $k$ (the number of neighbours):" ] }, { "cell_type": "code", "execution_count": 23, "metadata": { "id": "_sCIy_F0suup" }, "outputs": [], "source": [ "k_values = [10, 20, 40, 80, 100]\n", "pointclouds_pairs = []\n", "for k in k_values:\n", " X_new, y_new = ottscot.align(k=k, e=1e-3, normalize=True, norm=\"l2\")\n", " pointclouds_pairs.append((X_new, y_new))" ] }, { "cell_type": "code", "execution_count": 24, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 744 }, "id": "fsv8n60ltkXh", "outputId": "240c583a-5d99-48a1-a150-d1c379ff1bc7" }, "outputs": [ { "data": { "text/html": [ "\n", "\n", "\n", "\n", "\n", "\n", "
\n", " \n", "
\n", " \n", "
\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
\n", "
\n", " \n", " \n", " \n", " \n", " \n", " \n", "
\n", "
\n", "
\n", "\n", "\n", "\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "fig = plt.figure(figsize=(9, 9))\n", "(line,) = plt.plot([], [])\n", "\n", "\n", "def animate(i):\n", " plt.clf()\n", " k = k_values[i]\n", " (X_new, y_new) = pointclouds_pairs[i]\n", " pca = PCA(n_components=2)\n", " Xy_pca = pca.fit_transform(np.concatenate((X_new, y_new), axis=0))\n", "\n", " df_new = pd.DataFrame(\n", " {\n", " \"x\": np.flip(Xy_pca[:, 0]),\n", " \"y\": np.flip(Xy_pca[:, 1]),\n", " \"cellTypes\": np.flip(cellTypes_atac_rna),\n", " \"original_domain\": np.flip(original_domain_type),\n", " }\n", " )\n", "\n", " sn.scatterplot(\n", " data=df_new,\n", " x=\"x\",\n", " y=\"y\",\n", " hue=\"cellTypes\",\n", " s=70,\n", " style=\"original_domain\",\n", " alpha=0.6,\n", " edgecolors=\"none\",\n", " )\n", "\n", " plt.title(\n", " \"PCA of chromatin accessibility points mapped to gene expression domain, \\n \\\n", " along with original gene expression points for k=\"\n", " + str(k)\n", " )\n", "\n", " return (line,)\n", "\n", "\n", "def init():\n", " line.set_data([], [])\n", " return (line,)\n", "\n", "\n", "anim = animation.FuncAnimation(\n", " fig,\n", " animate,\n", " init_func=init,\n", " frames=list(range(5)),\n", " interval=1500,\n", " blit=True,\n", ")\n", "\n", "\n", "html = display.HTML(anim.to_jshtml())\n", "display.display(html)\n", "plt.close()" ] } ], "metadata": { "accelerator": "GPU", "colab": { "collapsed_sections": [], "name": "gromov_wasserstein_multiomics.ipynb", "provenance": [] }, "kernelspec": { "display_name": "ott", "language": "python", "name": "ott" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.10.6" } }, "nbformat": 4, "nbformat_minor": 1 }