02a_remote_files/exploring_bioimage_archive.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "ac8fe9b2-5a79-48ad-8ae2-ced7dd8888dd",
   "metadata": {},
   "source": [
    "# Exploring the BioImage Archive\n",
    "In this notebook we use the [bia-explorer](https://github.com/BioImage-Archive/bia-explorer) project to explore the [S-BIAD634 dataset](https://www.ebi.ac.uk/bioimage-archive/galleries/S-BIAD634-ai.html)  in the [Bio-image Archive](https://www.ebi.ac.uk/bioimage-archive). We will download some images and store them in a local directory."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "3c0e30d0-283c-41d2-b6ed-f6368d83f07b",
   "metadata": {},
   "outputs": [],
   "source": [
    "from bia_explorer import io, biostudies\n",
    "from skimage.io import imread, imsave\n",
    "from IPython.display import display, Markdown\n",
    "import stackview"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "605b72ab-c4e6-4fc7-8e66-dd1128ee9f40",
   "metadata": {},
   "source": [
    "## Accessing meta-data\n",
    "First we access the meta data of the dataets. Here we can for example see what the data can be used for and under which license it can be used."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "46d874f0-002f-4162-89d0-fbdb6f47497a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Title : An annotated fluorescence image dataset for training nuclear segmentation methods\n",
      "Description : This dataset contains annotated fluorescent nuclear images of normal or cancer cells from different tissue origins and sample preparation types, and can be used to train machine-learning based nuclear image segmentation algorithms. It consists of 79 expert-annotated fluorescence images of immuno and DAPI stained samples containing 7813 nuclei in total. In addition, the dataset is heterogenous in aspects such as type of preparation, imaging modality, magnification, signal-to-noise ratio and other technical aspects. Relevant parameters, e.g. diagnosis, magnification, signal-to-noise ratio and modality with respect to the type of preparation are provided in the file list. The images are derived from one Schwann cell stroma-rich tissue (from a ganglioneuroblastoma) cryosection (10 images/2773 nuclei), seven neuroblastoma (NB) patients (19 images/931 nuclei), one Wilms patient (1 image/102 nuclei), two NB cell lines (CLB-Ma, STA-NB10) (8 images/1785 nuclei) and a human keratinocyte cell line (HaCaT) (41 images/2222 nuclei).\n",
      "Keywords : AI\n",
      "Keywords : segmentation\n",
      "Keywords : nucleus\n",
      "Keywords : fluorescence\n",
      "License : CC0\n",
      "Funding statement : This work was facilitated by an EraSME grant (project TisQuant) under the grant no. 844198 and by a COIN grant (project VISIOMICS) under the grant no. 861750, both grants kindly provided by the Austrian Research Promotion Agency (FFG), and the St. Anna Kinderkrebsforschung. Partial funding was further provided by BMK, BMDW, and the Province of Upper Austria in the frame of the COMET Programme managed by FFG.\n"
     ]
    }
   ],
   "source": [
    "accession = 'S-BIAD634'\n",
    "submission = biostudies.load_submission(accession)\n",
    "\n",
    "for a in submission.section.attributes:\n",
    "    name = a.name\n",
    "    short_value = str(a.value)\n",
    "    \n",
    "    print(f\"{name} : {short_value}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "84e01875-523e-4f02-86d0-78226809ec79",
   "metadata": {},
   "source": [
    "We can also see how many images are in the dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "b7feae3a-e0e9-4b07-a339-5f0448806e57",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "388"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "study = io.load_bia_study(accession)\n",
    "\n",
    "len(study.images)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d3aac11c-ccc5-49ad-8603-a05cf7009fba",
   "metadata": {},
   "source": [
    "## Visualizing images\n",
    "\n",
    "A single image can be loaded and shown like this ([See also](https://github.com/BioImage-Archive/bia-explorer/issues/3))."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "687bdf50-76b0-447f-8f62-cc217f5c564e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "BIAImage(uri='https://www.ebi.ac.uk/biostudies/files/S-BIAD634/dataset\\\\groundtruth\\\\Ganglioneuroblastoma_0.tif', size=2239668, fpath=WindowsPath('dataset/groundtruth/Ganglioneuroblastoma_0.tif'))"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "image = study.images[0]\n",
    "image"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "0447fc37-240e-435e-8b81-a62961aee9e6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table>\n",
       "<tr>\n",
       "<td>\n",
       "<img src=\"data:image/png;base64,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\"></img>\n",
       "</td>\n",
       "<td style=\"text-align: center; vertical-align: top;\">\n",
       "\n",
       "<table>\n",
       "<tr><td>shape</td><td>(914,&nbsp;1225)</td></tr>\n",
       "<tr><td>dtype</td><td>uint16</td></tr>\n",
       "<tr><td>size</td><td>2.1 MB</td></tr>\n",
       "<tr><td>min</td><td>0</td></tr><tr><td>max</td><td>300</td></tr>\n",
       "</table>\n",
       "<img src=\"data:image/png;base64,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\"></img>\n",
       "</td>\n",
       "</tr>\n",
       "</table>"
      ],
      "text/plain": [
       "StackViewNDArray([[0, 0, 0, ..., 0, 0, 0],\n",
       "                  [0, 0, 0, ..., 0, 0, 0],\n",
       "                  [0, 0, 0, ..., 0, 0, 0],\n",
       "                  ...,\n",
       "                  [0, 0, 0, ..., 0, 0, 0],\n",
       "                  [0, 0, 0, ..., 0, 0, 0],\n",
       "                  [0, 0, 0, ..., 0, 0, 0]], dtype=uint16)"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "uri = image.uri.replace(\"\\\\\", \"/\")\n",
    "image_data = imread(uri)\n",
    "stackview.insight(image_data)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5c00bb0c-148b-4852-995c-520256e08783",
   "metadata": {},
   "source": [
    "To get an idea about the folder structure within the datasets, we can print out paths on the server."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "350503d1-2616-428e-a7bd-40e416643155",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "dataset\\groundtruth\\Ganglioneuroblastoma_0.tif\n",
      "dataset\\groundtruth\\Ganglioneuroblastoma_1.tif\n",
      "dataset\\groundtruth\\Ganglioneuroblastoma_10.tif\n",
      "dataset\\groundtruth\\Ganglioneuroblastoma_2.tif\n",
      "dataset\\groundtruth\\Ganglioneuroblastoma_3.tif\n",
      "dataset\\rawimages\\otherspecimen_5.tif\n",
      "dataset\\rawimages\\otherspecimen_6.tif\n",
      "dataset\\rawimages\\otherspecimen_7.tif\n",
      "dataset\\rawimages\\otherspecimen_8.tif\n",
      "dataset\\rawimages\\otherspecimen_9.tif\n"
     ]
    }
   ],
   "source": [
    "# print out filenames of some images\n",
    "for image in study.images[:5] + study.images[-5:]:\n",
    "    print(str(image.fpath))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "996e3ab1-24b5-4f44-8bca-579aed840f8e",
   "metadata": {},
   "source": [
    "## Downloading data\n",
    "Before we download selected images, we need to ensure that the folder exists where we want to store the data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "81fa7b52-cdb1-4b62-9f37-82c0277fc57a",
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "\n",
    "def ensure_folder_exists(folder_path):\n",
    "    if not os.path.exists(folder_path):\n",
    "        os.makedirs(folder_path)\n",
    "\n",
    "base_folder = f\"data/{accession}\"\n",
    "raw_folder = f\"data/{accession}/images\"\n",
    "groundtruth_folder = f\"data/{accession}/groundtruth\"\n",
    "\n",
    "ensure_folder_exists(base_folder)\n",
    "ensure_folder_exists(raw_folder)\n",
    "ensure_folder_exists(groundtruth_folder)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d6af74f1-9927-4510-b5a9-954cb3b59d0d",
   "metadata": {},
   "source": [
    "Next we download all raw images, and ground-truth annotations of all datasets containing \"Ganglioneuroblastoma\" in their name. We also only download the files in case they are not downloaded yet."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "f022ddc4-7deb-4ad2-a2ea-59220ed76930",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\haase\\AppData\\Local\\Temp\\ipykernel_20112\\2091960599.py:14: UserWarning: data/S-BIAD634/groundtruth/Ganglioneuroblastoma_0.tif is a low contrast image\n",
      "  imsave(target_file, image_data)\n",
      "C:\\Users\\haase\\AppData\\Local\\Temp\\ipykernel_20112\\2091960599.py:14: UserWarning: data/S-BIAD634/groundtruth/Ganglioneuroblastoma_1.tif is a low contrast image\n",
      "  imsave(target_file, image_data)\n",
      "C:\\Users\\haase\\AppData\\Local\\Temp\\ipykernel_20112\\2091960599.py:14: UserWarning: data/S-BIAD634/groundtruth/Ganglioneuroblastoma_10.tif is a low contrast image\n",
      "  imsave(target_file, image_data)\n",
      "C:\\Users\\haase\\AppData\\Local\\Temp\\ipykernel_20112\\2091960599.py:14: UserWarning: data/S-BIAD634/groundtruth/Ganglioneuroblastoma_2.tif is a low contrast image\n",
      "  imsave(target_file, image_data)\n",
      "C:\\Users\\haase\\AppData\\Local\\Temp\\ipykernel_20112\\2091960599.py:14: UserWarning: data/S-BIAD634/groundtruth/Ganglioneuroblastoma_3.tif is a low contrast image\n",
      "  imsave(target_file, image_data)\n",
      "C:\\Users\\haase\\AppData\\Local\\Temp\\ipykernel_20112\\2091960599.py:14: UserWarning: data/S-BIAD634/groundtruth/Ganglioneuroblastoma_4.tif is a low contrast image\n",
      "  imsave(target_file, image_data)\n",
      "C:\\Users\\haase\\AppData\\Local\\Temp\\ipykernel_20112\\2091960599.py:14: UserWarning: data/S-BIAD634/groundtruth/Ganglioneuroblastoma_6.tif is a low contrast image\n",
      "  imsave(target_file, image_data)\n",
      "C:\\Users\\haase\\AppData\\Local\\Temp\\ipykernel_20112\\2091960599.py:14: UserWarning: data/S-BIAD634/groundtruth/Ganglioneuroblastoma_7.tif is a low contrast image\n",
      "  imsave(target_file, image_data)\n",
      "C:\\Users\\haase\\AppData\\Local\\Temp\\ipykernel_20112\\2091960599.py:14: UserWarning: data/S-BIAD634/groundtruth/Ganglioneuroblastoma_8.tif is a low contrast image\n",
      "  imsave(target_file, image_data)\n",
      "C:\\Users\\haase\\AppData\\Local\\Temp\\ipykernel_20112\\2091960599.py:14: UserWarning: data/S-BIAD634/groundtruth/Ganglioneuroblastoma_9.tif is a low contrast image\n",
      "  imsave(target_file, image_data)\n"
     ]
    }
   ],
   "source": [
    "for image in study.images:\n",
    "    if \"Ganglioneuroblastoma\" in str(image.fpath):\n",
    "        uri = image.uri.replace(\"\\\\\", \"/\")\n",
    "        filename = uri.split(\"/\")[-1]\n",
    "        if \"\\\\rawimages\\\\\" in str(image.fpath):\n",
    "            target_file = raw_folder + \"/\" + filename\n",
    "            if not os.path.exists(target_file):\n",
    "                image_data = imread(uri)\n",
    "                imsave(target_file, image_data)\n",
    "        if \"\\\\groundtruth\\\\\" in str(image.fpath):\n",
    "            target_file = groundtruth_folder + \"/\" + filename\n",
    "            if not os.path.exists(target_file):\n",
    "                image_data = imread(uri)\n",
    "                imsave(target_file, image_data)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "381f39fd-1b97-4917-bcfc-490b54feb1c6",
   "metadata": {},
   "source": [
    "We can then check which files arrived."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "3491c7a6-bc61-4d5a-8b28-378705fe5410",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Ganglioneuroblastoma_0.tif\n",
      "Ganglioneuroblastoma_1.tif\n",
      "Ganglioneuroblastoma_10.tif\n",
      "Ganglioneuroblastoma_2.tif\n",
      "Ganglioneuroblastoma_3.tif\n",
      "Ganglioneuroblastoma_4.tif\n",
      "Ganglioneuroblastoma_6.tif\n",
      "Ganglioneuroblastoma_7.tif\n",
      "Ganglioneuroblastoma_8.tif\n",
      "Ganglioneuroblastoma_9.tif\n"
     ]
    }
   ],
   "source": [
    "for f in os.listdir(raw_folder):\n",
    "    print(f)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "ad066bf1-9955-4de0-9218-165da84ff32f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Ganglioneuroblastoma_0.tif\n",
      "Ganglioneuroblastoma_1.tif\n",
      "Ganglioneuroblastoma_10.tif\n",
      "Ganglioneuroblastoma_2.tif\n",
      "Ganglioneuroblastoma_3.tif\n",
      "Ganglioneuroblastoma_4.tif\n",
      "Ganglioneuroblastoma_6.tif\n",
      "Ganglioneuroblastoma_7.tif\n",
      "Ganglioneuroblastoma_8.tif\n",
      "Ganglioneuroblastoma_9.tif\n"
     ]
    }
   ],
   "source": [
    "for f in os.listdir(groundtruth_folder):\n",
    "    print(f)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "448c021b-bbdb-4f5c-a409-50e30b5946bf",
   "metadata": {},
   "source": [
    "## Exercise\n",
    "Download all images with \"Neuroblastoma\" in their name and upload them to a folder in the Speicherwolke / Nextcloud. Do not download and upload files which already exist."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "84571723-0fef-492f-aa67-753b2b2827bd",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.9.16"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}