{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Evaluating convergence\n", "\n", "Here we evaluate the convergence of a trajectory using the clustering ensemble similarity method and the dimensionality reduction ensemble similarity methods.\n", "\n", "**Last updated:** December 2022 with MDAnalysis 2.4.0-dev0\n", "\n", "**Last updated:** December 2022\n", "\n", "**Minimum version of MDAnalysis:** 1.0.0\n", "\n", "**Packages required:**\n", " \n", "* MDAnalysis (Michaud-Agrawal *et al.*, 2011, Gowers *et al.*, 2016)\n", "* MDAnalysisTests\n", "* [scikit-learn](https://scikit-learn.org/stable/)\n", " \n", "**Optional packages for visualisation:**\n", "\n", "* [matplotlib](https://matplotlib.org)\n", "\n", "\n", "
\n", " \n", "**Note**\n", "\n", "The metrics and methods in the `encore` module are from (Tiberti *et al.*, 2015). Please cite them when using the ``MDAnalysis.analysis.encore`` module in published work.\n", "\n", "
" ] }, { "cell_type": "code", "execution_count": 1, "metadata": { "execution": { "iopub.execute_input": "2021-05-19T05:59:55.164483Z", "iopub.status.busy": "2021-05-19T05:59:55.163573Z", "iopub.status.idle": "2021-05-19T05:59:56.540827Z", "shell.execute_reply": "2021-05-19T05:59:56.541182Z" } }, "outputs": [], "source": [ "import numpy as np\n", "import matplotlib.pyplot as plt\n", "%matplotlib inline\n", "\n", "import MDAnalysis as mda\n", "from MDAnalysis.tests.datafiles import PSF, DCD\n", "from MDAnalysis.analysis import encore\n", "from MDAnalysis.analysis.encore.clustering import ClusteringMethod as clm\n", "from MDAnalysis.analysis.encore.dimensionality_reduction import DimensionalityReductionMethod as drm" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Loading files" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "The test files we will be working with here feature adenylate kinase (AdK), a phosophotransferase enzyme. (Beckstein *et al.*, 2009)" ] }, { "cell_type": "code", "execution_count": 2, "metadata": { "execution": { "iopub.execute_input": "2021-05-19T05:59:56.546291Z", "iopub.status.busy": "2021-05-19T05:59:56.545647Z", "iopub.status.idle": "2021-05-19T05:59:56.785608Z", "shell.execute_reply": "2021-05-19T05:59:56.786089Z" } }, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "/home/pbarletta/mambaforge/envs/mda-user-guide/lib/python3.9/site-packages/MDAnalysis/coordinates/DCD.py:165: DeprecationWarning: DCDReader currently makes independent timesteps by copying self.ts while other readers update self.ts inplace. This behaviour will be changed in 3.0 to be the same as other readers\n", " warnings.warn(\"DCDReader currently makes independent timesteps\"\n" ] } ], "source": [ "u = mda.Universe(PSF, DCD)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Evaluating convergence with similarity measures" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "The convergence of the trajectory is evaluated by the similarity of the conformation ensembles in windows of the trajectory. The trajectory is divided into windows that increase by `window_size` frames. For example, if your trajectory had 13 frames and you specified a `window_size=3`, your windows would be:\n", "\n", " - Window 1: ---\n", " - Window 2: ------\n", " - Window 3: ---------\n", " - Window 4: -------------\n", " \n", "Where `-` represents 1 frame.\n", "\n", "These are compared using either the similarity of their clusters (`ces_convergence`) or their reduced dimension coordinates (`dres_convergence`). The rate at which the similarity values drop to 0 is indicative of how much the trajectory keeps on resampling the same regions of the conformational space, and therefore is the rate of convergence. " ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Using default arguments with clustering ensemble similarity" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "See [clustering_ensemble_similarity.ipynb](clustering_ensemble_similarity.ipynb#Calculating-clustering-similarity-with-default-settings) for an introduction to comparing trajectories via clustering. See the [API documentation for ces_convergence](https://docs.mdanalysis.org/stable/documentation_pages/analysis/encore/similarity.html#MDAnalysis.analysis.encore.similarity.ces_convergence) for more information." ] }, { "cell_type": "code", "execution_count": 3, "metadata": { "execution": { "iopub.execute_input": "2021-05-19T05:59:56.791388Z", "iopub.status.busy": "2021-05-19T05:59:56.790617Z", "iopub.status.idle": "2021-05-19T05:59:58.964292Z", "shell.execute_reply": "2021-05-19T05:59:58.964732Z" } }, "outputs": [], "source": [ "ces_conv = encore.ces_convergence(u, # universe\n", " 10, # window size\n", " select='name CA') # default" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "The output is an array of similarity values, with the shape (`number_of_windows`, `number_of_clustering_methods`)." ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ " 0.4819\n", " 0.4028\n", " 0.3170\n", " 0.2522\n", " 0.1983\n", " 0.1464\n", " 0.0991\n", " 0.0567\n", " 0.0000\n" ] } ], "source": [ "for row in ces_conv:\n", " for sim in row:\n", " print(\"{:>7.4f}\".format(sim))" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "This can be easily plotted as a line." ] }, { "cell_type": "code", "execution_count": 5, "metadata": { "execution": { "iopub.execute_input": "2021-05-19T05:59:58.991618Z", "iopub.status.busy": "2021-05-19T05:59:58.990939Z", "iopub.status.idle": "2021-05-19T05:59:59.088137Z", "shell.execute_reply": "2021-05-19T05:59:59.088508Z" }, "scrolled": true }, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "ces_fig, ces_ax = plt.subplots()\n", "plt.plot(ces_conv)\n", "ces_ax.set_xlabel('Window')\n", "ces_ax.set_ylabel('Jensen-Shannon divergence');" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Comparing different clustering methods" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "You may want to try different clustering methods, or use different parameters within the methods. `encore.ces_convergence` allows you to pass a list of `clustering_methods` to be applied, much like [normal clustering ensemble similarity methods](clustering_ensemble_similarity.ipynb#Calculating-clustering-similarity-with-multiple-methods).\n", "\n", "
\n", " \n", "**Note**\n", "\n", "To use the other ENCORE methods available, you need to install [scikit-learn](https://scikit-learn.org/stable/).\n", "\n", "
" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "The KMeans clustering algorithm separates samples into $n$ groups of equal variance, with centroids that minimise the inertia. You must choose how many clusters to partition. [(See the scikit-learn user guide for more information.)](https://scikit-learn.org/stable/modules/clustering.html#k-means)" ] }, { "cell_type": "code", "execution_count": 6, "metadata": { "execution": { "iopub.execute_input": "2021-05-19T05:59:59.094046Z", "iopub.status.busy": "2021-05-19T05:59:59.093346Z", "iopub.status.idle": "2021-05-19T05:59:59.095367Z", "shell.execute_reply": "2021-05-19T05:59:59.094973Z" } }, "outputs": [], "source": [ "km1 = clm.KMeans(12, # no. clusters\n", " init = 'k-means++', # default\n", " algorithm=\"auto\") # default\n", "\n", "km2 = clm.KMeans(6, # no. clusters\n", " init = 'k-means++', # default\n", " algorithm=\"auto\") # default\n", "\n", "km3 = clm.KMeans(3, # no. clusters\n", " init = 'k-means++', # default\n", " algorithm=\"auto\") # default" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "When we pass a list of clustering methods to `encore.ces_convergence`, the similarity values get saved in `ces_conv2` in order." ] }, { "cell_type": "code", "execution_count": 7, "metadata": { "execution": { "iopub.execute_input": "2021-05-19T05:59:59.099109Z", "iopub.status.busy": "2021-05-19T05:59:59.098472Z", "iopub.status.idle": "2021-05-19T06:00:01.094005Z", "shell.execute_reply": "2021-05-19T06:00:01.094590Z" } }, "outputs": [ { "data": { "text/plain": [ "(9, 3)" ] }, "execution_count": 7, "metadata": {}, "output_type": "execute_result" } ], "source": [ "ces_conv2 = encore.ces_convergence(u, # universe\n", " 10, # window size\n", " select='name CA',\n", " clustering_method=[km1, km2, km3]\n", " )\n", "ces_conv2.shape" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "As you can see, the number of clusters partitioned by KMeans has an effect on the resulting rate of convergence." ] }, { "cell_type": "code", "execution_count": 8, "metadata": { "execution": { "iopub.execute_input": "2021-05-19T06:00:01.112454Z", "iopub.status.busy": "2021-05-19T06:00:01.108370Z", "iopub.status.idle": "2021-05-19T06:00:01.241556Z", "shell.execute_reply": "2021-05-19T06:00:01.242120Z" } }, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "labels = ['12 clusters', '6 clusters', '3 clusters']\n", "\n", "ces_fig2, ces_ax2 = plt.subplots()\n", "for data, label in zip(ces_conv2.T, labels):\n", " plt.plot(data, label=label)\n", "ces_ax2.set_xlabel('Window')\n", "ces_ax2.set_ylabel('Jensen-Shannon divergence')\n", "plt.legend();" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Using default arguments with dimension reduction ensemble similarity" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "See [dimension_reduction_ensemble_similarity.ipynb](dimension_reduction_ensemble_similarity.ipynb#Calculating-dimension-reduction-similarity-with-default-settings) for an introduction on comparing trajectories via dimensionality reduction. We now use the `dres_convergence` function ([API docs](https://docs.mdanalysis.org/stable/documentation_pages/analysis/encore/similarity.html#MDAnalysis.analysis.encore.similarity.dres_convergence))." ] }, { "cell_type": "code", "execution_count": 9, "metadata": { "execution": { "iopub.execute_input": "2021-05-19T06:00:01.247319Z", "iopub.status.busy": "2021-05-19T06:00:01.246438Z", "iopub.status.idle": "2021-05-19T06:00:03.689365Z", "shell.execute_reply": "2021-05-19T06:00:03.689783Z" } }, "outputs": [], "source": [ "dres_conv = encore.dres_convergence(u, # universe\n", " 10, # window size\n", " select='name CA') # default" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Much like `ces_convergence`, the output is an array of similarity values." ] }, { "cell_type": "code", "execution_count": 10, "metadata": { "execution": { "iopub.execute_input": "2021-05-19T06:00:03.696587Z", "iopub.status.busy": "2021-05-19T06:00:03.695214Z", "iopub.status.idle": "2021-05-19T06:00:03.699634Z", "shell.execute_reply": "2021-05-19T06:00:03.699083Z" }, "tags": [ "nbval-ignore-output" ] }, "outputs": [ { "data": { "text/plain": [ "array([[0.52851279],\n", " [0.38894634],\n", " [0.278017 ],\n", " [0.22236994],\n", " [0.17581073],\n", " [0.13455005],\n", " [0.07451477],\n", " [0.02843863],\n", " [0. ]])" ] }, "execution_count": 10, "metadata": {}, "output_type": "execute_result" } ], "source": [ "dres_conv" ] }, { "cell_type": "code", "execution_count": 11, "metadata": { "execution": { "iopub.execute_input": "2021-05-19T06:00:03.710016Z", "iopub.status.busy": "2021-05-19T06:00:03.709181Z", "iopub.status.idle": "2021-05-19T06:00:03.815587Z", "shell.execute_reply": "2021-05-19T06:00:03.816008Z" }, "scrolled": true }, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "dres_fig, dres_ax = plt.subplots()\n", "plt.plot(dres_conv)\n", "dres_ax.set_xlabel('Window')\n", "dres_ax.set_ylabel('Jensen-Shannon divergence');" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Comparing different dimensionality reduction methods" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Again, you may want to compare the performance of different methods.\n", "\n", "Principal component analysis uses singular value decomposition to project data onto a lower dimensional space. [(See the scikit-learn user guide for more information.)](https://scikit-learn.org/stable/modules/decomposition.html#pca)\n", "\n", "The method provided by MDAnalysis.encore accepts any of the keyword arguments of [sklearn.decomposition.PCA](https://scikit-learn.org/stable/modules/generated/sklearn.decomposition.PCA.html) *except* `n_components`. Instead, use `dimension` to specify how many components to keep." ] }, { "cell_type": "code", "execution_count": 12, "metadata": { "execution": { "iopub.execute_input": "2021-05-19T06:00:03.820777Z", "iopub.status.busy": "2021-05-19T06:00:03.820060Z", "iopub.status.idle": "2021-05-19T06:00:03.821891Z", "shell.execute_reply": "2021-05-19T06:00:03.822264Z" } }, "outputs": [], "source": [ "pc1 = drm.PrincipalComponentAnalysis(dimension=1,\n", " svd_solver='auto')\n", "pc2 = drm.PrincipalComponentAnalysis(dimension=2,\n", " svd_solver='auto')\n", "pc3 = drm.PrincipalComponentAnalysis(dimension=3,\n", " svd_solver='auto')" ] }, { "cell_type": "code", "execution_count": 13, "metadata": { "execution": { "iopub.execute_input": "2021-05-19T06:00:03.826293Z", "iopub.status.busy": "2021-05-19T06:00:03.825752Z", "iopub.status.idle": "2021-05-19T06:00:05.200317Z", "shell.execute_reply": "2021-05-19T06:00:05.200683Z" }, "scrolled": true }, "outputs": [ { "data": { "text/plain": [ "(9, 3)" ] }, "execution_count": 13, "metadata": {}, "output_type": "execute_result" } ], "source": [ "dres_conv2 = encore.dres_convergence(u, # universe\n", " 10, # window size\n", " select='name CA',\n", " dimensionality_reduction_method=[pc1, pc2, pc3]\n", " )\n", "dres_conv2.shape" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Again, the size of the subspace you choose to include in your similarity comparison, affects the apparent rate of convergence over the trajectory." ] }, { "cell_type": "code", "execution_count": 14, "metadata": { "execution": { "iopub.execute_input": "2021-05-19T06:00:05.216927Z", "iopub.status.busy": "2021-05-19T06:00:05.216371Z", "iopub.status.idle": "2021-05-19T06:00:05.325075Z", "shell.execute_reply": "2021-05-19T06:00:05.325585Z" }, "scrolled": true }, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "labels = ['1D', '2D', '3D']\n", "\n", "dres_fig2, dres_ax2 = plt.subplots()\n", "for data, label in zip(dres_conv2.T, labels):\n", " plt.plot(data, label=label)\n", "dres_ax2.set_xlabel('Window')\n", "dres_ax2.set_ylabel('Jensen-Shannon divergence')\n", "plt.legend();" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## References\n", "\n", "[1] Oliver Beckstein, Elizabeth J. Denning, Juan R. Perilla, and Thomas B. Woolf.\n", "Zipping and Unzipping of Adenylate Kinase: Atomistic Insights into the Ensemble of OpenClosed Transitions.\n", "Journal of Molecular Biology, 394(1):160–176, November 2009.\n", "00107.\n", "URL: https://linkinghub.elsevier.com/retrieve/pii/S0022283609011164, doi:10.1016/j.jmb.2009.09.009.\n", "\n", "[2] Richard J. Gowers, Max Linke, Jonathan Barnoud, Tyler J. E. Reddy, Manuel N. Melo, Sean L. Seyler, Jan Domański, David L. Dotson, Sébastien Buchoux, Ian M. Kenney, and Oliver Beckstein.\n", "MDAnalysis: A Python Package for the Rapid Analysis of Molecular Dynamics Simulations.\n", "Proceedings of the 15th Python in Science Conference, pages 98–105, 2016.\n", "00152.\n", "URL: https://conference.scipy.org/proceedings/scipy2016/oliver_beckstein.html, doi:10.25080/Majora-629e541a-00e.\n", "\n", "[3] Naveen Michaud-Agrawal, Elizabeth J. Denning, Thomas B. Woolf, and Oliver Beckstein.\n", "MDAnalysis: A toolkit for the analysis of molecular dynamics simulations.\n", "Journal of Computational Chemistry, 32(10):2319–2327, July 2011.\n", "00778.\n", "URL: http://doi.wiley.com/10.1002/jcc.21787, doi:10.1002/jcc.21787.\n", "\n", "[4] Matteo Tiberti, Elena Papaleo, Tone Bengtsen, Wouter Boomsma, and Kresten Lindorff-Larsen.\n", "ENCORE: Software for Quantitative Ensemble Comparison.\n", "PLOS Computational Biology, 11(10):e1004415, October 2015.\n", "00031.\n", "URL: https://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1004415, doi:10.1371/journal.pcbi.1004415." ] } ], "metadata": { "kernelspec": { "display_name": "Python 3.9.15 ('mda-user-guide')", "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": "3.9.15" }, "toc": { "base_numbering": 1, "nav_menu": {}, "number_sections": false, "sideBar": true, "skip_h1_title": false, "title_cell": "Table of Contents", "title_sidebar": "Contents", "toc_cell": false, "toc_position": {}, "toc_section_display": true, "toc_window_display": false }, "vscode": { "interpreter": { "hash": "7b52aa17ef4e9358c0e91ff3f0bf50d10667bb8b55636d4b9fb967a2ff94bd8c" } } }, "nbformat": 4, "nbformat_minor": 2 }