Adding mia examples

.gitignore

@@ -16,10 +16,16 @@ src/docs/*.html
 src/*.html
 src/notebooks/**/*.html
 src/*-listing.json
+renv/
+renv.lock
+.Rhistory
 
 *.parquet
 !**/example-data/**/*.parquet
 *.sig
 ko*.pathview.png
 src/notebooks/R Examples/*.tsv
-src/notebooks/R Examples/*.txt
+src/notebooks/R Examples/*.txt
+.Rproj.user
+.RData
+.Rprofile

src/notebooks/R Mia Examples/Fetch-Analyses-metadata-for-a-Study.ipynb

@@ -0,0 +1,143 @@
+{
+  "cells": [
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "---\n",
+        "title: \"Fetch analyses metadata for a study\"\n",
+        "author:\n",
+        "  - name: Noah de Gunst\n",
+        "    affiliation:\n",
+        "      - id: mia\n",
+        "        name: Department of Computing, University of Turku, Finland\n",
+        "  - name: Sam Hillman\n",
+        "    affiliation:\n",
+        "      - id: mia\n",
+        "        name: Department of Computing, University of Turku, Finland\n",
+        "categories: [R]\n",
+        "execute: \n",
+        "  eval: true\n",
+        "---\n",
+        "\n",
+        "::: {style=\"max-width:1200px\"}\n",
+        "![](./../_resources/mgnify_logo.png)\n",
+        "\n",
+        "# Fetch a study using MGnifyR; download the metadata for all of its analyses\n",
+        "\n",
+        "The [MGnify API](https://www.ebi.ac.uk/metagenomics/api/v1) returns data and \n",
+        "relationships as JSON. [MGnifyR](https://www.bioconductor.org/packages/release/bioc/html/MGnifyR.html) \n",
+        "is a package to help you read MGnify data into your R analyses.\n",
+        "\n",
+        "You can find all of the other \"API endpoints\" using the [Browsable API interface in your web browser](https://www.ebi.ac.uk/metagenomics/api/v1).\n",
+        "\n",
+        "This is an interactive code notebook (a Jupyter Notebook). To run this code, click \n",
+        "into each cell and press the ▶ button in the top toolbar, or press `shift+enter`.\n",
+        "\n",
+        "------------------------------------------------------------------------\n",
+        ":::\n",
+        "\n",
+        "#### Setting the access code\n",
+        "First, we need to specify the accession number of the study we're working with. \n",
+        "This can be done by setting the `mgnify_study_accession` variable. The accession \n",
+        "number uniquely identifies the study in the MGnify database.\n",
+        "\n",
+        "```{r}\n",
+        "#| output: false\n",
+        "\n",
+        "source(\"./utils/variable_utils.r\")\n",
+        "\n",
+        "mgnify_study_accession <- get_variable_from_link_or_input('MGYS', 'Study Accession', \n",
+        "                                                          'MGYS00005116')\n",
+        "\n",
+        "# You can also just directly set the accession variable in code, like this:\n",
+        "# mgnify_study_accession <- \"MGYS00005292\"\n",
+        "```\n",
+        "\n",
+        "#### Constructing a MgnifyClient object to access the database\n",
+        "To interact with the MGnify database, we need to create an MgnifyClient object. \n",
+        "This object allows us to fetch data from MGnify, and we can configure it to use \n",
+        "a cache for efficiency. \n",
+        "\n",
+        "```{r}\n",
+        "#| output: false\n",
+        "# Importing the libraries\n",
+        "library(vegan)\n",
+        "library(ggplot2)\n",
+        "library(mia)\n",
+        "library(MGnifyR)\n",
+        "\n",
+        "# Check if the cache directory exists, if not, create it\n",
+        "if (!dir.exists(\"./.mgnify_cache\")) {\n",
+        "  dir.create(\"./.mgnify_cache\", recursive = TRUE)\n",
+        "}\n",
+        "\n",
+        "# Create the MgnifyClient object with caching enabled\n",
+        "mg <- MgnifyClient(usecache = TRUE, cache_dir = \"./.mgnify_cache\")\n",
+        "```\n",
+        "\n",
+        "#### Displaying the help file\n",
+        "\n",
+        "```{r}\n",
+        "#| output: false\n",
+        "library(IRdisplay)\n",
+        "display_markdown(file = '../_resources/mgnifyr_help.md')\n",
+        "```\n",
+        "\n",
+        "## Fetch a list of the Analyses for the Study\n",
+        "Using the MgnifyClient object, we can search for all analyses associated with the \n",
+        "study accession number we set earlier. This will return a list of analysis accession \n",
+        "numbers.\n",
+        "\n",
+        "```{r}\n",
+        "#| output: false\n",
+        "analyses_accessions <- searchAnalysis(mg, \"studies\", mgnify_study_accession)\n",
+        "analyses_accessions\n",
+        "```\n",
+        "\n",
+        "## Download metadata for the first 10 Analyses\n",
+        "...and put it into a dataframe.\n",
+        "\n",
+        "```{r}\n",
+        "#| output: false\n",
+        "analyses_metadata_df <- getMetadata(mg, head(analyses_accessions, 10))\n",
+        "```\n",
+        "\n",
+        "## Display metadata\n",
+        "The table could be big, so let's look at a sample of it (`head`).\n",
+        "\n",
+        "```{r}\n",
+        "#| output: false\n",
+        "t(head(analyses_metadata_df))\n",
+        "```\n",
+        "\n",
+        "## Download the data to a multi-assay data object\n",
+        "\n",
+        "> [mia](https://microbiome.github.io/mia/) is a Bioconductor package designed to \n",
+        "import, store and analyze microbiome data using an object called a `TreeSummarizedExperiment.` \n",
+        "This is a tailored data container optimized for microbiome data analysis.Being \n",
+        "built on the `SummarizedExperiment` class, miaverse seamlessly integrates into the \n",
+        "extensive `SummarizedExperiment` ecosystem. In this example we download the MGnifyR \n",
+        "data to an MAE, which contains multiple `TreeSummarizedExperiment` objects.\n",
+        "\n",
+        "\n",
+        "```{r}\n",
+        "#| output: false\n",
+        "mae <- getResult(mg, accession = analyses_accessions)\n",
+        "```\n",
+        "\n",
+        "You use `MGnifyR` features further, for example to download data. Check the Cheat \n",
+        "Sheet at the top for more."
+      ]
+    }
+  ],
+  "metadata": {
+    "kernelspec": {
+      "display_name": "Python 3",
+      "language": "python",
+      "name": "python3"
+    }
+  },
+  "nbformat": 4,
+  "nbformat_minor": 4
+}

src/notebooks/R Mia Examples/_resources/mgnifyr_help.md

@@ -0,0 +1,12 @@
+# Help with MGnifyR
+
+MGnifyR is an R package that provides a convenient way for R users to access data from [the MGnify API](https://www.ebi.ac.uk/metagenomics/api/).
+
+Detailed help for each function is available in R using the standard `?function_name` command.
+
+A vignette is available containing a reasonably verbose overview of the main functionality. 
+This can be read either within R with the `vignette("MGnifyR")` command, or [on the bioconductor vignette website](https://www.bioconductor.org/packages/release/bioc/vignettes/MGnifyR/inst/doc/MGnifyR.html)
+
+## MGnifyR Command cheat sheet
+
+For a full list of key MGnifyR functions, please look at the [MGnifyR website](https://ebi-metagenomics.github.io/MGnifyR/reference/index.html).

src/notebooks/R Mia Examples/utils/kegg_pathways_utils.r

@@ -0,0 +1,60 @@
+# Retrieve name and URL for a specific pathway in the KEGG database 
+get_pathway_info <- function(pathway) {
+  pathway <- paste("map", pathway, sep = "")
+  pathway_name <- keggList(pathway)[[1]]
+  pathway_url <- paste("https://www.kegg.jp/pathway/", pathway, sep = "")
+  return(list(pathway_name = pathway_name, pathway_url = pathway_url))
+}
+
+
+# Function to prompt users for pathway selection and return custom pathway IDs
+PathwaysSelection <- function() {
+    display_markdown("#### Pathways Selection :\n\n
+- For the most general & most complete pathways, input 'G'\n\n
+- Press Enter to generate the most complete pathways\n\n
+- To add custom pathways, input pathway numbers (ex: 00053,01220)")
+
+  flush.console()
+  CUSTOM_PATHWAY_IDS <- get_variable_from_link_or_input('CUSTOM_PATHWAY_IDS', name = 'Pathways Accession', default = '')
+
+  if (CUSTOM_PATHWAY_IDS == "") {
+    CUSTOM_PATHWAY_IDS <- list()
+  } else if (CUSTOM_PATHWAY_IDS == "G") {
+    CUSTOM_PATHWAY_IDS <- list("00010", "00020", "00030", "00061", "01232","00240", "00190")
+  } else {
+    CUSTOM_PATHWAY_IDS <- strsplit(CUSTOM_PATHWAY_IDS, ",")[[1]]
+  }
+
+  message(if (length(CUSTOM_PATHWAY_IDS) > 0) {
+    paste("\nUsing", CUSTOM_PATHWAY_IDS, " - ", sapply(CUSTOM_PATHWAY_IDS, function(id) paste(get_pathway_info(id)[1]," : ",get_pathway_info(id)[2])), "as a Custom Pathway")
+  } else {
+    "\nUsing NONE as a Custom Pathway"
+  })
+    return(CUSTOM_PATHWAY_IDS)
+}
+
+
+# Clearing the current working directory and displaying generated figures from `pathway_plots/` directory
+generatePathwayPlots <- function() {
+# Clearing the current working directory
+  if (!dir.exists("pathway_plots")) {
+    dir.create("pathway_plots")
+  }
+
+  file.copy(from = list.files(pattern = "./*pathview.png"), to = "./pathway_plots/", overwrite = TRUE)
+
+  png_files <- list.files(path = ".", pattern = "*.png")
+  xml_files <- list.files(path = ".", pattern = "*.xml")
+  files <- c(png_files, xml_files)
+  output <- capture.output({
+    unlink(files)
+  })
+
+# Accessing the png files and displaying it
+  images <- list.files("pathway_plots", full.names = TRUE)
+
+  for (pathway in images) {
+    display_markdown(get_pathway_info(gsub("[^0-9]", "", basename(pathway)))$pathway_name)
+    display_png(file = pathway)
+  }
+}

src/notebooks/R Mia Examples/utils/variable_utils.r

@@ -0,0 +1,16 @@
+library(glue)
+
+get_variable_from_link_or_input <- function(variable, name = 'accession', default = NA) {
+    # Get a variable value, either from an ENV VAR that would have been set by the jlab_query_params extension, or through direct user input.
+    var <- Sys.getenv(variable, unset = NA)
+    if (!is.na(var)) {
+        print(glue('Using {name} = {var} from the link you followed.'))
+    } else {
+        determiner <- ifelse(grepl(tolower(substr(name, 0, 1)), 'aeiou'), 'an', 'a')
+        var <- readline(prompt = glue("Type {determiner} {name} [default: {default}]"))
+    }
+    var <- ifelse(is.na(var) || var == '', default, var)
+    print(glue('Using "{var}" as {name}'))
+    var
+}
+