WIP Set up import from Zenodo, GitHub hash; remove .downloadZ

R/bugphyzz.R

@@ -12,8 +12,8 @@ utils::globalVariables(c(
 #' tidy data.frames. To learn more about the structure of the data.frames
 #' please check the bugphyzz vignette with `browseVignettes("bugphyzz")`.
 #'
-#' @param version Character string indicating the version.
-#' Options: devel or a zenodo record.
+#' @param version Character string indicating the version. Default is the
+#' latest release on Zenodo.  Options: Zenodo DOI, GitHub commit hash, or devel.
 #' @param force_download Logical value. Force a fresh download of the data or
 #' use the one stored in the cache (if available). Default is FALSE.
 #' @param v Validation value. Default 0.5 (see details).
@@ -54,12 +54,14 @@ utils::globalVariables(c(
 #' names(bp)
 #'
 importBugphyzz <- function(
-    version = 'devel', force_download = FALSE, v = 0.5, exclude_rarely = TRUE
+    version = "d3fd894", force_download = FALSE, v = 0.5, exclude_rarely = TRUE
 
 ) {
-  if (version == 'devel') {
-    output <- .downloadDevel(force_download)
-  }
+
+  ## output is a list of three data.frames
+  ## one of each: binary, multistate, numeric
+  output <- .downloadResource(version, force_download)
+
   ## TODO add release version
   output <- lapply(output, function(x) split(x, x$Attribute))
   output <- purrr::list_flatten(output)
@@ -285,9 +287,9 @@ getTaxonSignatures <- function(tax, bp, ...) {
 }
 
 .validationData <- function() {
-  url <-  "https://raw.githubusercontent.com/waldronlab/taxPProValidation/main/validation_summary.tsv"
+  fname <- system.file("extdata", "validation_summary.tsv", package = "bugphyzz")
   utils::read.table(
-    file = url, header = TRUE, sep = "\t", row.names = NULL
+    file = fname, header = TRUE, sep = "\t", row.names = NULL
   ) |>
     dplyr::mutate(
       value = dplyr::case_when(
@@ -297,30 +299,20 @@ getTaxonSignatures <- function(tax, bp, ...) {
     )
 }
 
-## Import the devel version of bupghyzz
-.downloadDevel <- function(force_download) {
-  types <- c("multistate", "binary", "numeric")
-  urls <- paste0(
-    "https://github.com/waldronlab/bugphyzzExports/raw/main/bugphyzz_",
-    types,
-    ".csv"
-  )
-  names(urls) <- types
-  output <- vector("list", length(urls))
-  for (i in seq_along(output)) {
-    message("Importing ", names(urls)[i], " data...")
-    names(output)[i] <- names(urls)[i]
-    rpath <- .getResource(
-      rname = paste0("bugphyzz_", names(urls)[i], ".tsv"),
-      url = urls[i], verbose = TRUE, force = force_download
-    )
-    output[[i]] <- utils::read.csv(rpath, header = TRUE, skip = 1) |>
-      dplyr::mutate(Attribute = tolower(Attribute))
+## Import a version of bupghyzz
+.downloadResource <- function(version, force_download) {
+  if (stringr::str_detect(version, "^10.5281/zenodo.[0-9]+$")) {
+    suffix <- sub("^10.5281/zenodo\\.", "", version)
+    output <- .downloadZ(suffix, force_download)
+  } else if (version == "devel" || stringr::str_detect(version, stringr::regex("^[:alnum:]{7}$")) ){
+    output <- .downloadGH(version, force_download)
+  } else {
+    stop("Version must be a Zenodo DOI, GitHub commit hash, or 'devel'.")
   }
   return(output)
 }
 
-## TODO update this function when relase is ready
+## Function for downloading data on Zenodo
 .downloadZ <- function(record, force_download) {
   base_url <- paste0("https://zenodo.org/api/records/", record)
   req <- httr2::request(base_url)
@@ -340,9 +332,29 @@ getTaxonSignatures <- function(tax, bp, ...) {
 
   output <- vector("list", length(files))
   for (i in seq_along(output)) {
-    output[[i]] <- utils::read.csv(files[i], header = TRUE)
-    # output[[i]] <- utils::read.csv(files, header = TRUE, skip = 1)
-    # dplyr::mutate(Attribute = tolower(Attribute))
+    output[[i]] <- utils::read.csv(files, header = TRUE, skip = 1) |>
+      dplyr::mutate(Attribute = tolower(Attribute))
   }
   return(output)
 }
+
+## Function for downloading data on GitHub
+.downloadGH <- function(version, force_download) {
+    file_suffix <- c("binary", "multistate", "numeric")
+    urls <- paste0("https://github.com/waldronlab/bugphyzzExports/raw/",
+           version, "/bugphyzz_", file_suffix, ".csv"
+    )
+    names(urls) <-  c("binary", "multistate", "numeric")
+    output <- vector("list", length(urls))
+    for (i in seq_along(output)) {
+      message("Importing ", names(urls)[i], " data...")
+      names(output)[i] <- names(urls)[i]
+      rpath <- .getResource(
+        rname = paste0("bugphyzz_", names(urls)[i], ".csv"),
+        url = urls[i], verbose = TRUE, force = force_download
+      )
+      output[[i]] <- utils::read.csv(rpath, header = TRUE, skip = 1) |>
+        dplyr::mutate(Attribute = tolower(Attribute))
+    }
+    return(output)
+}

README.md

@@ -27,3 +27,7 @@ After installation, check vignettes with:
 ```r
 browseVigenttes("bugphyzz")
 ```
+
+TODO:
+
+- Add scores to signatures

inst/extdata/README.md

@@ -104,6 +104,38 @@ Columns:
 | unit | units used for numeric data. |
 
 
+## [validation_summary.tsv](./validation_summary.tsv)
+
+Results of the validation of the ASR method using the waldronlab/taxPPro package
+
+Hash: e736097
+
+| Column name | Description |
+| ----------- | ----------- |
+| method | ASR method (phytools-ltp or castor-ltp) |
+| rank | Taxonomic rank (all, genus, species, strain). all was used for the final results.|
+| physiology | The name of the attribute group. |
+| attribute | The name of the attribute or attribute value.|
+| mcc_mean | mean of Mathew's correlation coefficient for discrete only. |
+| mcc_sd | standard deviation of Matthew's correlation coefficient for discrete only.|
+| r2_mean | R-squard for numeric only. |
+| r2_sd | R-squared for numeric only. |
+| ltp_bp | Interesction between taxa in bugphyzz (per attribute/attribute value) and the ltp tree. |
+| bp | Taxa in bugphyzz (per attribute/attribute value). |
+| ltp_bp_phys | Intersection between taxa in bugphyz (per attribute group) and the ltp tree. |
+| bp_phys | Total taxa in the bugphyzz (per attribute group).|
+| ltp | Number of taxa in the LTP tree. |
+| nsti_mean | Mean NSTI value. This was calculated for all physiologies, but is only relevant for numeric ones.|
+| nsti_sd |  Standard deviation of the NSTI value. This was calculated for all physiologies, but is only relevant for numeric ones. |
+| ltp_bp_per | Intersection of LTP tree and bugphyzz (per attribute/attribute value) in numbers. |
+| ltp_bp_phys_per | Intersection of LTP tree and bugphyzz (per physiology/attribute group) in percentage. |
+
+
+
+
+
+
+
 
 
 

inst/extdata/attributes.tsv

@@ -170,8 +170,8 @@
 "extreme environment"	NA	"logical"	NA	NA
 "COGEM pathogenicity rating"	"COGEM pathogenicity rating"	"integer"	NA	NA
 "antimicrobial sensitivity"	"antimicrobial sensitivity"	"logical"	NA	NA
-"biofilm formation"	"biofilm forming"	"logical"	"APO:0000159"	"The ability to form a layer on solid surfaces with or without other microorganisms."
-"Butyrate-Producing Bacteria"	"butyrate producing"	"numeric"	"NCIT:C126417"	"Any anaerobic bacteria that metabolizes complex carbohydrates to produce butyrate. These microorganisms may form colonies in the intestines of mammals; their presence may aid the host by catabolizing undigested fiber and starch. Additionally, butyrate is metabolized by colonic epithelial cells and may protect the host against ulcerative colitis and cancer."
+"biofilm formation"	"biofilm formation"	"logical"	"APO:0000159"	"The ability to form a layer on solid surfaces with or without other microorganisms."
+"Butyrate-Producing Bacteria"	"Butyrate-Producing Bacteria"	"numeric"	"NCIT:C126417"	"Any anaerobic bacteria that metabolizes complex carbohydrates to produce butyrate. These microorganisms may form colonies in the intestines of mammals; their presence may aid the host by catabolizing undigested fiber and starch. Additionally, butyrate is metabolized by colonic epithelial cells and may protect the host against ulcerative colitis and cancer."
 "acetate producing"	"acetate producing"	"numeric"	NA	NA
 "lactate producing"	"lactate producing"	"numeric"	NA	NA
 "mutation rate per site per year"	"mutation rates per site per year"	"numeric"	NA	NA
@@ -407,7 +407,7 @@
 "terrestrial biome"	"habitat"	"logical"	NA	NA
 "not host-associated"	"habitat"	"logical"	NA	NA
 "rumen"	"habitat;isolation site"	"logical"	NA	NA
-"Waste and sludge water"	"habitat"	"logical"	NA	NA
+"waste and sludge water"	"habitat"	"logical"	NA	NA
 "blood"	"habitat;isolation site"	"logical"	NA	NA
 "bovine"	"isolation site"	"logical"	NA	NA
 "cerebral spinal fluid"	"habitat"	"logical"	NA	NA
@@ -453,7 +453,7 @@
 "gamma"	"hemolysis"	"logical"	NA	NA
 "beta"	"hemolysis"	"logical"	NA	NA
 "alpha"	"hemolysis"	"logical"	NA	NA
-"CAMP test"	"hemolysis"	"logical"	NA	NA
+"camp test"	"hemolysis"	"logical"	NA	NA
 "biosafety level 1"	"biosafety level"	"logical"	NA	NA
 "biosafety level 2"	"biosafety level"	"logical"	NA	NA
 "biosafety level 3"	"biosafety level"	"logical"	NA	NA