NRSC7610_FluourescenceAnalysis.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Biological Imaging Analysis with Python\n",
    "#### 1. Algorithmic cell counting\n",
    "#### 2. Fluorescence quantification\n",
    "#### 3. Automatic figure regeneration\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Some of the code is pre-written, and you will simply execute it. In other cases, you will be prompted to write some code lines to advance towards this goal. **There are going to be 8 such prompts,** plus some bonus challenges if you want to keep going.\n",
    "\n",
    "## Imports of packages we will need for this notebook. \n",
    "#### All of these are standard and most that we have talked about. Other packages will be introduced as needed."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import seaborn as sns\n",
    "import os,sys,glob\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "one packages we haven't talked about is this custom \"package\" ```utils``` i have added to this repository. it only has one function for now, which we will use much later, but it has an example for how to import a local package of your making, for reusing code across notebooks. note that this code is in viewable in the sidebar <--- to the left, in the ```utils``` folder"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "import utils"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 1. Algorithmic cell counting. \n",
    "#### The goal of this section is to develop some code that will automatically count things in an image. Generally speaking, this type of automated image analysis is quite useful - counting how many cells are in a culture dish, or how many nuclei have fragmented (apoptosis), or how many PSD95 (or whatever!) puncta there are in an image. \n",
    "#### First, we'll load an image of some cells in culture. \n",
    "We'll use a loading function from the ```matplotlib``` package:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "image = plt.imread(\"./res/cell-counting_image.png\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "So, we've loaded a pre-provided , we think, no error. \n",
    "<br>**First question of this exercise:<br>Q1. What type of data is ```image```? write some code that will give the answer in the cell below:**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "But what does this image of cells look like, actually? we want to take a look to make sure everything looks as we would expect"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x7fd3804e1a10>"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(image)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "A color image! Great. But what is the information in each color? How does ```plt.imshow(image)``` know to make this a color image? Let's check the shape of ```image```"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "np.shape(image)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's see what is in each color:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "for color in range(np.shape(image)[2]):  # go through each color image\n",
    "    plt.figure()            # make a new figure\n",
    "    plt.imshow(image[:,:,color],cmap=plt.cm.Greys) # show that image, in black and white"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There's not much information in the color, so let's combine them. We'll do it by "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "image_bw = np.sum(image,axis=2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's take a look at what this summed image looks like. \n",
    "<br>**Q2: plot  ```image_bw```**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To keep going towards our goal, we're going to use some very useful image analysius packages: ```scikit-image```, or ```skimage```, and one part of ```scipy```, ```scipy.ndimage```. These are packages that were developed primarily by computer scientists for machine vision purposes, but we can take advantage of them for our task here."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "from skimage.filters import threshold_otsu\n",
    "from scipy import ndimage"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's first get rid of some noise"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "guassian_sigma = 2.5"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "plt.imshow(ndimage.gaussian_filter(image_bw,guassian_sigma))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Nex,t we try label every pixels has having a cell there or not, so binary yes or no:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "thresh = threshold_otsu(ndimage.gaussian_filter(image_bw,guassian_sigma))\n",
    "# thresh = ndimage.gaussian_filter(thresh,9)\n",
    "binary = ndimage.gaussian_filter(image_bw,guassian_sigma) > thresh"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "plt.imshow(binary)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "filled_in = ndimage.binary_fill_holes(binary)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 185,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x7f8991670890>"
      ]
     },
     "execution_count": 185,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(filled_in)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Finally, we count how many things there are in this image"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 186,
   "metadata": {},
   "outputs": [],
   "source": [
    "labeled_cells, b = ndimage.label(filled_in)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Q3: How many cells are there in the image? Assign the number to the variable ```cell_count```** _Hint: google ```ndimage.label```_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2. Comparing fluourescence images\n",
    "#### The goal of this section will be to quantify the overlap between fluoresence images. These raw data have been graciously provided by Helena Barr from the Hughes Lab. Much appreciated! These are images of microglia. \n",
    "Again we'll use the ```skimage``` pacakge, in this case to load the images. We will load the images into a ```numpy <ndarray>```"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "from skimage import io"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "image_stack = io.imread('./res/Microglia_1d.tif')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As you can tell from the variable name, this one file is an image stack."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<br>**Q4. How many images are there in this image stack? Another way to ask this: what is the shape of ```image_stack```? write some code that will give the answer in the cell(s) below. Assign variables ```num_images``` , ```width```, ```length``` with your answer** If you need more cells, go ahead and add them!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 225,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's take a look at the images:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "for image_number in range(num_images):\n",
    "    plt.figure()\n",
    "    plt.imshow(image_stack[image_number,:,:])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "These images look largely the same, but slightly different (they are a time-lapse capture). \n",
    "<br>Let's ask: over these ```num_images```, does the cell increase or decrease fluorescence? How much?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's wrap the preprocessing we did above into a function, which takes in an image and returns that image, after it has been thresholded and filled in. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "def preprocess_image(image,gaussian_sigma=.75,threshold=130):\n",
    "    binary = ndimage.gaussian_filter(image,gaussian_sigma) > threshold\n",
    "    filled_in = ndimage.binary_fill_holes(binary)\n",
    "    return filled_in"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can now go over the stack images, which is going through time, and see how this cell is changing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "for ind in range(num_images-1):\n",
    "    diff_between_images = preprocess_image(image_stack[ind+1,:,:]) ^ preprocess_image(image_stack[ind,:,:])\n",
    "    plt.figure()\n",
    "    plt.imshow(diff_between_images,cmap=plt.cm.Greys)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now, instead of plottting, let's measure the number of pixels that are different\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "pixel_differences = [] # make a list\n",
    "for ind in range(num_images-1):\n",
    "    diff_between_images = preprocess_image(image_stack[ind+1,:,:]) ^ preprocess_image(image_stack[ind,:,:])\n",
    "    number_of_different_pixels = np.sum(diff_between_images)\n",
    "    pixel_differences.extend([number_of_different_pixels])\n",
    "pixel_differences = np.array(pixel_differences)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Great, we have the number of pixels that has changes. But the meaning of this depends on how many pixels there are in the image. Let's convert this to the percentage of pixels that have changed. \n",
    "<br>**Q5: How many pixels are there in each image? Assign this to the variable ```total_number_of_pixels```**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<br>**Q6: Remake the above plot, but with percentage of different pixels instead of total number**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 0, 'time (arbitrary unit)')"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.ylabel('percentage of different pixels')\n",
    "plt.xlabel('time (arbitrary unit)')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's wrap this up by making a big function that makes a bar chart, which shows the mean and standard deviation of the percentage change over time. I'll write this one:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_percent_change_between_image(stacks):\n",
    "    means = np.zeros(len(stacks))\n",
    "    stds = np.zeros(len(stacks))\n",
    "    for i,image_stack in enumerate(stacks):\n",
    "        pixel_differences = [] # make a list\n",
    "        for ind in range(num_images-1):\n",
    "            diff_between_images = preprocess_image(image_stack[ind+1,:,:]) ^ preprocess_image(image_stack[ind,:,:])\n",
    "            number_of_different_pixels = np.sum(diff_between_images)\n",
    "            pixel_differences.extend([number_of_different_pixels])\n",
    "        pixel_differences = np.array(pixel_differences) / total_number_of_pixels\n",
    "\n",
    "        means[i]=np.mean(pixel_differences)\n",
    "        stds[i] =np.std(pixel_differences)\n",
    "\n",
    "    return means, stds"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "mean, std = get_percent_change_between_image([image_stack])\n",
    "\n",
    "plt.bar(1,mean,yerr=std,color='b')\n",
    "plt.xlim(-2,4)\n",
    "plt.ylim(0,0.05)\n",
    "plt.ylabel('percentage difference, mean')\n",
    "plt.xticks([1],['Microglia1d'])\n",
    "\n",
    "utils.cleanAxes(plt.gca(),leftLabels=True,bottomLabels=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 3. Automating Figure production"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "One very convenient application of scripting languages, whether it MATLAB or python or whatever, is the ability to automate your fixed analyses to produce updated plots once you have more data. This could be done with _any_ type of data, but in this final example we will do it with the Hughes Lab microglia data.\n",
    "<br><br>First, let's make a plot of some analysis of one set of data. In this case, we will use ```microglia_1d.tif```. But instead of directly writing a path to this data, we'll search a folder for all data that starts with ```\"Microglia\"```"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [],
   "source": [
    "path_to_data_folder = './res'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "_Note_: in this case for our exercise, we are using a local folder in the JupyterHub folder. But it could be any folder. Your could point it to Desktop, or Documents, or \"C:/folder_where_keep_my_data:\", etc.\n",
    "<br><br>\n",
    "Now we'll get a list of all of the files in this folder that are called ```microglia_1```+any letter or character+```.tif```. we use use two python standard library pacakages: ```glob```, which is great for searching folders in flexible ways, and ```os```, which is critical for cross-platform compatibility when browsing file systems"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['./res/Microglia_1d.tif']\n"
     ]
    }
   ],
   "source": [
    "microglia_image_paths = glob.glob(os.path.join(path_to_data_folder,'Microglia_1*.tif'))\n",
    "print(microglia_image_paths)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "so, we can get this image by loading it the same way we did before:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [],
   "source": [
    "image_stack_1 = io.imread(microglia_image_paths[0])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now, let's imagine you collect some more data. You put it in the same folder, and you call it ```Microglia_2d.tif```"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Q7: Use ```glob``` to find all .tif files that start with \"Microglia\" in the same folder. Call this ```image_paths```**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [],
   "source": [
    "image_paths = glob.glob(os.path.join(path_to_data_folder,'Microglia_*'))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now we get all of these tif images, and "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [],
   "source": [
    "images_all = [io.imread(path) for path in image_paths]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [],
   "source": [
    "means, stds = get_percent_change_between_image(images_all)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['./res/Microglia_2d.tif', './res/Microglia_1d.tif']"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "image_paths"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.bar(range(len(means)),means,yerr=stds,color='b')\n",
    "# plt.xlim(-2,4)\n",
    "plt.ylim(0,0.05)\n",
    "plt.ylabel('percentage difference, mean')\n",
    "\n",
    "utils.cleanAxes(plt.gca(),leftLabels=True,bottomLabels=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Q8: Make the plot again, but make the bars different colors and mak"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.7.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}