templates/4_Post_Clustering.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "2a2fb446-2c56-4936-bd4f-00f936b23285",
   "metadata": {},
   "source": [
    "# Post-clustering tasks\n",
    "This notebook allows the user to inspect and fine-tune the output of the [clustering](https://github.com/angelolab/ark-analysis/blob/main/templates/3_Cluster_Cells.ipynb) notebook. There are three parts of this notebook. \n",
    "1. Clustering cleanup: If there are clusters that were not properly separated during Pixie cell clustering, this provides the option of defining manual thresholds based on marker intensity to combine or separate specific clusters\n",
    "2. Marker thresholding: For markers that were not directly used for clustering, but whose expression is important for phenotyping individual cells, this provides the user with visualization to determine accurate thresholds for positive/negative classification\n",
    "3. Generate additional cell cluster column(s) to represent different resolutions of clustering data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "94732e91-d7c8-4f7c-a94d-cee6139c8d6a",
   "metadata": {
    "tags": [
     "import"
    ]
   },
   "outputs": [],
   "source": [
    "# import needed libraries\n",
    "import os\n",
    "\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "\n",
    "from ark.phenotyping.post_cluster_utils import (create_mantis_project,\n",
    "                                                plot_hist_thresholds, \n",
    "                                                generate_new_cluster_resolution)\n",
    "from ark.utils import example_dataset"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "61f43e17-9980-4a82-8512-518e8cb6678f",
   "metadata": {
    "tags": []
   },
   "source": [
    "## 0. Download the Example Dataset\n",
    "\n",
    "Here we are using the example data located in `/data/example_dataset/input_data`. To modify this notebook to run using your own data, simply change `base_dir` to point to your own sub-directory within the data folder.\n",
    "\n",
    "* `base_dir`: the path to all of your imaging data. This directory will contain all of the data generated by this notebook, as well as the data generated by cell clustering."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "69c8e900",
   "metadata": {
    "tags": [
     "base_dir"
    ]
   },
   "outputs": [],
   "source": [
    "base_dir = '../data/example_dataset/'"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d62f27c9-0d97-4cdf-bed0-dff67013ed14",
   "metadata": {},
   "source": [
    "If you would like to test the features in Ark with an example dataset, run the cell below. It will download a dataset consisting of 11 FOVs with 22 channels. You may find more information about the example dataset in the [README](../README.md#example-dataset).\n",
    "\n",
    "If you are using your own data, skip the cell below.\n",
    "\n",
    "* `overwrite_existing`: If set to `False`, it will not overwrite existing data in the `data/example_dataset`. Recommended setting to `False` if you are running Notebooks 1,2,3 and 4 in succession. Set to `True` if you are just running Notebook 4."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a9d8369f-8f82-4166-82a5-79cb899c7c59",
   "metadata": {
    "tags": [
     "ex_data_download"
    ]
   },
   "outputs": [],
   "source": [
    "example_dataset.get_example_dataset(dataset=\"post_clustering\", save_dir = base_dir, overwrite_existing = False)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6893ee36-47f4-492d-a7ef-70708c77952a",
   "metadata": {},
   "source": [
    "## Set variables\n",
    "* `testing_fovs`: a list of FOVs to use for evaluating the post-clustering tasks\n",
    "* `base_dir`: the path to all of your imaging data. Should be the same one used for the [cell clustering notebook](https://github.com/angelolab/ark-analysis/blob/main/templates_ark/3_Cluster_Cells.ipynb)\n",
    "* `cell_output_dir`: the path to the directory used for cell clustering\n",
    "* `cell_table_name`: the name of the cell table containing the cluster labels\n",
    "* `segmentation_dir`: the path to the directory containing the segmentation masks\n",
    "* `image_dir`: the path to the directory containing the image data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "52bc223f-6e8d-4cae-a53d-2dca4b3c5c87",
   "metadata": {
    "tags": [
     "file_path"
    ]
   },
   "outputs": [],
   "source": [
    "# FOVs that will be used to evaluate accuracy\n",
    "testing_fovs = ['fov0', 'fov1']\n",
    "\n",
    "# paths\n",
    "cell_output_dir = os.path.join(base_dir, 'pixie/example_cell_output_dir')\n",
    "cell_table_name = 'segmentation/cell_table/cell_table_size_normalized_cell_labels.csv'\n",
    "segmentation_dir = os.path.join(base_dir, 'segmentation/deepcell_output')\n",
    "image_dir = os.path.join(base_dir, 'image_data')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ec5cf527-baff-4667-a757-62513184a234",
   "metadata": {},
   "source": [
    "The following directories will be automatically created without any user input required"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "01308ac6-c939-482e-8751-2078a0122727",
   "metadata": {
    "tags": [
     "dir_set"
    ]
   },
   "outputs": [],
   "source": [
    "# create directory to hold outputs\n",
    "post_cluster_dir = os.path.join(base_dir, 'post_clustering')\n",
    "if not os.path.exists(post_cluster_dir):\n",
    "    os.makedirs(post_cluster_dir)\n",
    "    \n",
    "# create directory to hold mantis masks\n",
    "mask_dir = os.path.join(post_cluster_dir, 'masks')\n",
    "if not os.path.exists(mask_dir):\n",
    "    os.makedirs(mask_dir)\n",
    "\n",
    "# create directory to hold full mantis output\n",
    "mantis_dir = os.path.join(base_dir, 'mantis')\n",
    "if not os.path.exists(mantis_dir):\n",
    "    os.makedirs(mantis_dir)\n",
    "    \n",
    "# read in cell table\n",
    "cell_table =  pd.read_csv(os.path.join(base_dir, cell_table_name), header=0)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "74a4f3b4-b1cb-4f7e-b8eb-a980d80a3f00",
   "metadata": {},
   "source": [
    "## 1. Clustering Cleanup\n",
    "You may have noticed during the cell clustering process that certain populations need to be modfified/adjusted. This part of the notebook allows you to identify proportions of a given cell population to be reassigned to a different one. "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "26c3cc59-2e08-4a5c-ae76-06c71c3b2f1c",
   "metadata": {},
   "source": [
    "### Identify thresholds for splitting problematic clusters\n",
    "The cell below will plot a histogram of marker expression across multiple cell population. This is useful for identifying an aproximate threshold which separates positive from negative expression. The first population(s) should be positive for the marker, with the subsequent populations negative. This can be then be used to pull out the subset of cells which express a given marker, but shouldn't, from an individual population\n",
    "\n",
    "* `populations`: list of clustering populations to use for plotting\n",
    "* `marker`: the channel whose expression will be used to determine problematic cells\n",
    "* `threshold`: value that will be plotted with a vertical line for visualization purposes to aid identification of the correct value"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "d6f06ccc-c1b2-4f22-a10c-b53d7203541c",
   "metadata": {
    "tags": [
     "identify_threshold"
    ]
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x660 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# This cell can be re-run to identify different thresholds and different populations that need adjusting\n",
    "populations = ['CD4T', 'CD8T', \"immune_other\"]\n",
    "marker = 'CD8'\n",
    "threshold = 0.001\n",
    "\n",
    "plot_hist_thresholds(cell_table=cell_table, populations=populations, marker=marker, threshold=threshold)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "13eeaba4-1632-4a1e-917b-5b5b3995463c",
   "metadata": {},
   "source": [
    "### Splitting problematic clusters\n",
    "Once you have an identified a threshold which does a good job of separating negative from positive cells, that threshold can be used to pull out specific cells from a population and reassign them. For example, the cell below identifies 'immune other' cells which express CD8, and assigns them to a new population for manual inspection. You'll need to specify the following parameters:\n",
    "* `marker`: the marker used for separating cells, the same one that was used to generate the histograms above\n",
    "* `threshold`: the threshold for separating cells, can be identified using the threshold argument above or by visual inspection\n",
    "* `target_cluster`: the name of the cell_cluster that contains the contamination\n",
    "* `problematic_cluster`: the name of the new population that will be created based on the threshold\n",
    "\n",
    "This process of identifying thresholds and creating new populations can be repeated for each population where a problem was identified during the clustering process. We recommend making a copy of the cell below for each modification so that you have a record of what changes were made. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "074cd479-e6ac-4b48-ba1a-8087133dab5f",
   "metadata": {
    "tags": [
     "split_problematic_clusters"
    ]
   },
   "outputs": [],
   "source": [
    "# The plots above indicate that 0.001 is a reasonable threshold for separating CD3+ cells. \n",
    "marker = 'CD8'\n",
    "threshold = 0.001\n",
    "target_cluster = \"immune_other\"\n",
    "problematic_cluster = \"Tcell_other\"\n",
    "\n",
    "# determine logic for thresholding; e.g. marker > threshold, marker <= threshold, or alternative logic\n",
    "selected_idx = cell_table[marker] > threshold\n",
    "\n",
    "# Create new column in cell table based on above\n",
    "cell_table.loc[np.logical_and(cell_table['cell_meta_cluster'] == target_cluster, selected_idx), 'cell_meta_cluster'] = problematic_cluster"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bc65eff3-d0ad-4abf-80b2-0945e821480a",
   "metadata": {},
   "source": [
    "### Manual inspection of problematic clusters with Mantis\n",
    "Once you've pulled out the problematic cells into new clusters, you'll need to visually inspect them to determine which cluster they belong to. The code below will create a directory to enable easy visualization with Mantis Viewer. See the [Mantis section of the README](../README.md#mantis-viewer) for an overview of how to use Mantis. You can then manually inspect each of the newly created clusters to determine what the true cell type is."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "4c9f1ecd-dc67-47c1-9f72-27ea0d893549",
   "metadata": {
    "tags": [
     "mantis_manual_inspection"
    ]
   },
   "outputs": [],
   "source": [
    "# create directory that can be read into Mantis\n",
    "create_mantis_project(cell_table=cell_table, fovs=testing_fovs, seg_dir=segmentation_dir, pop_col='cell_meta_cluster', \n",
    "                      mask_dir=mask_dir, mantis_dir=mantis_dir, image_dir=image_dir)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cd36a474-3e39-4e43-b179-16da29361bf6",
   "metadata": {},
   "source": [
    "### Determine final cluster assignment\n",
    "Once you've inspected each of the clusters, you'll need to decide what their final assignment should be. You can either merge them back in with their original cluster, combine them with a different existing cluster, or create a new cluster, depending on what you found after manual inspection."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "7f3d31ac-8309-4740-846b-49a7ea94f369",
   "metadata": {
    "tags": [
     "final_cluster_assignment"
    ]
   },
   "outputs": [],
   "source": [
    "# set names of problematic cluster and new cluster\n",
    "problematic_cluster = \"Tcell_other\"\n",
    "final_assignment = 'CD8T'\n",
    "\n",
    "# assign problematic cluster new name\n",
    "cell_table.loc[cell_table['cell_meta_cluster'] == problematic_cluster, 'cell_meta_cluster'] = final_assignment"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "363d9b96-5b02-4cbd-a6d3-1e0963292a51",
   "metadata": {
    "tags": [
     "updated_cell_table"
    ]
   },
   "outputs": [],
   "source": [
    "# Save the updated cell table\n",
    "cell_table.to_csv(os.path.join(post_cluster_dir, 'updated_cell_table.csv'))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ded3c307-b2be-42b7-a4eb-91e8837a0c37",
   "metadata": {},
   "source": [
    "## 2. Marker thresholding\n",
    "For markers like Ki67 or PD1, we generally don't include them in the clustering process. Intead, following clustering, we score each cell as positive or negative for these \"functional markers.\" This requires picking a threshold for each marker to determine positivity and negativity. Looking directly at the histograms can be useful for getting a general range for where the threshold lies, but picking a specific threshold based only on the histogram can be challenging. Instead, what is often most useful is simply looking at images to determine the specific cutoff value."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ee0bba4e-07d1-457a-b025-a4b013a5e44e",
   "metadata": {},
   "source": [
    "### Set up variables\n",
    "To evaluate the marker levels, we use the modified cell table from the previous step. In addition, we select which markers need to be thresholded"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b1f2a1a1-28fe-4386-9b7b-e78579210736",
   "metadata": {
    "tags": [
     "marker_thresholding_vars"
    ]
   },
   "outputs": [],
   "source": [
    "# load modified cell table\n",
    "cell_table = pd.read_csv(os.path.join(post_cluster_dir, 'updated_cell_table.csv'))\n",
    "\n",
    "# specify functional markers that need to be thresholded\n",
    "functional_markers = ['PD1', 'Ki67', 'GLUT1']"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2d570e61-1852-4048-b8cd-c4f62a9d21a8",
   "metadata": {},
   "source": [
    "### Create Mantis project with functional marker counts\n",
    "To identify the appropriate threshold, we will include the counts of the markers in the mantis directory. The code below will include the specified functional markers for mantis to view"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "52cec373-71cf-418a-a16b-a2bacfdf5d04",
   "metadata": {
    "tags": [
     "mantis_marker_counts"
    ]
   },
   "outputs": [],
   "source": [
    "# create directory that can be read into Mantis\n",
    "create_mantis_project(cell_table=cell_table, fovs=testing_fovs, seg_dir=segmentation_dir, pop_col='cell_meta_cluster', \n",
    "                      mask_dir=mask_dir, mantis_dir=mantis_dir, image_dir=image_dir)\n",
    "\n",
    "# create dataframe with counts of the specified markers\n",
    "marker_counts_df = cell_table.loc[np.isin(cell_table['fov'], testing_fovs), ['fov', 'label'] + functional_markers]\n",
    "\n",
    "# save dataframe\n",
    "marker_counts_df.to_csv(os.path.join(mantis_dir, 'marker_counts.csv'), index=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f74cbe10-4160-482d-a75d-4f38919d4471",
   "metadata": {},
   "source": [
    "### Identify plausible range for a given marker threshold\n",
    "To get a rough idea of what range works for a given functional marker, we can plot its expression across cell types that we do, and do not, expect to be positive for it"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a70d705c-a89d-4e92-a0bd-fb60442784f1",
   "metadata": {
    "tags": [
     "marker_threshold_range"
    ]
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x660 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# This cell can be re-run to identify different thresholds across different functional markers\n",
    "populations = [\"M1_macrophage\", \"other\", \"tumor_ck17\"]\n",
    "marker = 'Ki67'\n",
    "threshold = 0.002\n",
    "\n",
    "plot_hist_thresholds(cell_table=cell_table, populations=populations, marker=marker, threshold=threshold)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fe9c4fd5-ad0a-4645-b1f6-e0e89001d194",
   "metadata": {},
   "source": [
    "### Manually inspect images to confirm\n",
    "Once you've identified a plausible threshold for the specified marker, it's time to look at the images in Mantis. In each image, identify the cells that are positive for the specified marker, and determine what their expression level is. Then, do the same thing for the cells that are negative for the marker. By performing this across a representative range of images, it is straightforward to come up with an estimate of what threshold accurately descriminates between positive and negative cells. You'll need to first [load the marker_counts.csv file](../README.md#loading_project_wide_files) into Mantis. Then, once it's been loaded, you can [view the expression](../README.md#view_cell_features) of the specified marker for the cells in the image. "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "26d65ba1-4510-4e31-9664-dd68c0f7c80e",
   "metadata": {},
   "source": [
    "### Record threshold values\n",
    "After manual inspection, you'll need to record the value you selected as a threshold for each marker. Fill in the list below with the values you selected, adding additional brackets as necessary for all of the markers you wish to annotate"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "9ffe19ba-b330-48df-b0fd-7e61efb280f6",
   "metadata": {
    "tags": [
     "threshold_list_vars"
    ]
   },
   "outputs": [],
   "source": [
    "# create list to hold all functional marker, value pairs\n",
    "threshold_list = [['PD1', 0.0005], ['Ki67', 0.002], ['GLUT1', 0.002]]\n",
    "\n",
    "# Save the thresholds as a csv file. Used in the Pairwise Spatial Enrichment Notebook\n",
    "threshold_df = pd.DataFrame({'marker': [x[0] for x in threshold_list],\n",
    "                             'threshold': [x[1] for x in threshold_list]})\n",
    "threshold_df.to_csv(os.path.join(post_cluster_dir, 'marker_thresholds.csv'), index=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "122cd78b-a739-4f43-91f6-b99b71614791",
   "metadata": {},
   "source": [
    "### Update cell table\n",
    "Finally, you'll update your cell table with the thresholded values. The code below will add a new column named `{marker}_threshold` for each of the indicated marker, with a value of `True` if it's greater than or equal to the threshold, and a value of `False` otherwise."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c24ba9cf-a1c8-4005-b1ad-67986173c6c9",
   "metadata": {
    "tags": [
     "cell_table_threshold"
    ]
   },
   "outputs": [],
   "source": [
    "# add columns to cell table\n",
    "for marker, threshold in threshold_list:\n",
    "    cell_table[marker + '_threshold'] = cell_table[marker].values >= threshold\n",
    "\n",
    "# save thresholded cell table\n",
    "cell_table.to_csv(os.path.join(post_cluster_dir, 'cell_table_thresholded.csv'))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d0b1e5b3-dcb5-4fc3-af66-51645993a03a",
   "metadata": {},
   "source": [
    "## 3. Generate Multiple Cluster Resolutions\n",
    "In this section, you can create a new column of cluster labels, in order to represent higher-level cell information. You will need to set the following variables:\n",
    "* `cluster_col`: the column containing the cell clusters you would like to group\n",
    "* `new_cluster_col`: the name of a new column to be created with the higher resolution clusters, cannont be the same as any existing columns in the cell table\n",
    "* `cluster_mapping`: a dictionary detailing the groupings of the original cell clusters, where the keys are the new cluster names and the value lists contain the original cluster names\n",
    "\n",
    "You can re-run this section for as many new cluster columns as you would like in generate."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "12faad12-aca8-4688-be91-d66e3867323c",
   "metadata": {
    "tags": [
     "read_table"
    ]
   },
   "outputs": [],
   "source": [
    "cell_table = pd.read_csv(os.path.join(post_cluster_dir, 'cell_table_thresholded.csv'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b72e8ad8-48d7-40e1-9298-afcc19b93f23",
   "metadata": {
    "tags": [
     "set_cluster_args"
    ]
   },
   "outputs": [],
   "source": [
    "cluster_col = 'cell_meta_cluster'\n",
    "new_cluster_col = 'new_cell_clustering'\n",
    "\n",
    "save_path = os.path.join(post_cluster_dir, 'final_cell_table.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d87ca06a-ae08-4dc3-a8de-fd6806b77fbb",
   "metadata": {},
   "source": [
    "The below code will provide the unique cluster names found in the column provided. Each of these must be included in a group in the `cluster_mapping` dictionary below."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e8b9d175-3006-4096-a1a7-d7902e8f8139",
   "metadata": {
    "tags": [
     "print_clusters"
    ]
   },
   "outputs": [],
   "source": [
    "print(cell_table[cluster_col].drop_duplicates().reset_index(drop=True))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e5499e56-8f65-45e5-a5f1-5a47e17beabb",
   "metadata": {},
   "source": [
    "Using the listed clusters, define the new cluster groups and then generate a new column in the cell table."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "35c798ea-55bc-4699-b45a-bb22dc31124a",
   "metadata": {
    "tags": [
     "cluster_mapping"
    ]
   },
   "outputs": [],
   "source": [
    "cluster_mapping = {'AB': ['A', 'B'], 'C': ['C']}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "54f1d61f-d469-453e-8049-047ec99c45d2",
   "metadata": {
    "tags": [
     "generate_new_clusters"
    ]
   },
   "outputs": [],
   "source": [
    "generate_new_cluster_resolution(cell_table, cluster_col, new_cluster_col, cluster_mapping, save_path)"
   ]
  }
 ],
 "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": "3.11.6"
  },
  "vscode": {
   "interpreter": {
    "hash": "31e90e2a7155cbaa93471c58eaf30b3c41423e5d17fccea222aec5e482389189"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}