Short term change

DESCRIPTION

@@ -23,18 +23,20 @@ biocViews: Microbiome, Software, Sequencing
 License: Artistic-2.0 | file LICENSE
 Depends:
     R (>= 4.4.0),
-    mia
+    ggplot2,
+    mia,
+    SummarizedExperiment
 Imports:
     dplyr,
     methods,
     S4Vectors,
     SingleCellExperiment,
-    SummarizedExperiment,
-    TreeSummarizedExperiment
+    TreeSummarizedExperiment,
+    ggrepel,
+    reshape2
 Suggests:
     BiocStyle,
     devtools,
-    ggplot2,
     knitr,
     lubridate,
     miaViz,

NAMESPACE

@@ -1,17 +1,30 @@
 # Generated by roxygen2: do not edit by hand
 
 export(addBaselineDivergence)
+export(addShortTermChange)
 export(addStepwiseDivergence)
 export(getBaselineDivergence)
+export(getShortTermChange)
 export(getStepwiseDivergence)
 exportMethods(addBaselineDivergence)
+exportMethods(addShortTermChange)
 exportMethods(addStepwiseDivergence)
 exportMethods(getBaselineDivergence)
+exportMethods(getShortTermChange)
 exportMethods(getStepwiseDivergence)
 import(TreeSummarizedExperiment)
 import(mia)
+importFrom(SummarizedExperiment,colData)
+importFrom(dplyr,"%>%")
 importFrom(dplyr,arrange)
+importFrom(dplyr,as_tibble)
 importFrom(dplyr,group_by)
 importFrom(dplyr,lag)
 importFrom(dplyr,mutate)
+importFrom(dplyr,summarize)
 importFrom(dplyr,ungroup)
+importFrom(ggplot2,aes)
+importFrom(ggplot2,ggplot)
+importFrom(ggrepel,geom_text_repel)
+importFrom(mia,rarefyAssay)
+importFrom(mia,transformAssay)

R/AllGenerics.R

@@ -20,3 +20,15 @@ setGeneric("getStepwiseDivergence", signature = c("x"), function(x, ...)
 #' @export
 setGeneric("addStepwiseDivergence", signature = "x", function(x, ...)
     standardGeneric("addStepwiseDivergence"))
+
+#' @rdname addShortTermChange
+#' @export
+setGeneric("addShortTermChange", signature = "x", function(x, ...)
+    standardGeneric("addShortTermChange"))
+
+#' @rdname addShortTermChange
+#' @export
+setGeneric("getShortTermChange", signature = "x", function(x, ...)
+    standardGeneric("getShortTermChange"))
+
+

R/getShortTermChange.R

@@ -0,0 +1,128 @@
+#' @title Short term Changes in Abundance
+#'
+#' @description Calculates short term changes in abundance of taxa
+#'              using temporal Abundance data.
+#'  
+#' @param x a \code{\link{SummarizedExperiment}} object.
+#' @param assay.type \code{Character scalar}. Specifies the name of assay 
+#'   used in calculation. (Default: \code{"counts"})
+#' @param name \code{Character scalar}. Specifies a name for storing
+#' short term results. (Default: \code{"short_term_change"})
+#' @param ... additional arguments.
+#' 
+#' 
+#' @return \code{getShortTermChange} returns \code{DataFrame} object containing 
+#' the short term change in abundance over time for a microbe. 
+#' \code{addShortTermChange}, on the other hand, returns a
+#' \code{\link[SummarizedExperiment:SummarizedExperiment-class]{SummarizedExperiment}}
+#' object with these results in its \code{metadata}.
+#' 
+#' @details This approach is used by Wisnoski NI and colleagues
+#'          \url{https://github.com/nwisnoski/ul-seedbank}. Their approach is based on
+#'          the following calculation log(present abundance/past abundance).
+#'          Also a compositional version using relative abundance similar to
+#'          Brian WJi, Sheth R et al
+#'          \url{https://www.nature.com/articles/s41564-020-0685-1} can be used.
+#'          This approach is useful for identifying short term growth behaviors of taxa.
+#'          
+#' @name addShortTermChange
+#' 
+#' 
+#' @examples
+#' 
+#' # Load time series data
+#' data(minimalgut)
+#' tse <- minimalgut
+#' 
+#' short_time_labels <- c("74.5h", "173h", "438h", "434h", "390h")
+#' 
+#' # Subset samples by Time_label and StudyIdentifier
+#' tse <- tse[, !(tse$Time_label %in% short_time_labels)]
+#' tse <- tse[, (tse$StudyIdentifier == "Bioreactor A")]
+#' 
+#' # Get short term change
+#' # Case of rarefying counts
+#' tse <- transformAssay(tse, method = "relabundance")
+#' getShortTermChange(tse, assay.type = "relabundance", time.col = "Time.hr")
+#' 
+#' # Case of transforming counts
+#' tse <- rarefyAssay(tse, assay.type = "counts")
+#' getShortTermChange(tse, assay.type = "subsampled", time.col = "Time.hr")
+NULL
+
+#' @rdname addShortTermChange
+#' @export
+#' @importFrom dplyr arrange as_tibble summarize "%>%"
+#' @importFrom ggplot2 ggplot aes
+#' @importFrom ggrepel geom_text_repel
+#' @importFrom mia rarefyAssay transformAssay
+#' @importFrom SummarizedExperiment colData
+setMethod("getShortTermChange", signature = c(x = "SummarizedExperiment"),
+    function(x, assay.type = "counts", ...){
+        ############################## Input check #############################
+        # Check validity of object
+        if (nrow(x) == 0L){
+            stop("No data available in `x` ('x' has nrow(x) == 0L.)",
+                 call. = FALSE)
+        }
+        # Check assay.type
+        .check_assay_present(assay.type, x)
+        ########################### Growth Metrics ############################
+        grwt <- .calculate_growth_metrics(x, assay.type = assay.type, ...)
+        # Clean and format growth metrics
+        grs.all <- .clean_growth_metrics(grwt, ...)
+        return(grs.all)
+    }
+)
+
+#' @rdname addShortTermChange
+#' @export
+setMethod("addShortTermChange", signature = c(x = "SummarizedExperiment"),
+          function(x, assay.type = "counts", name = "short_term_change", ...){
+              # Calculate short term change
+              res <- getShortTermChange(x, ...)
+              # Add to metadata
+              x <- .add_values_to_metadata(x, name, res, ...)
+              return(x)
+          }
+)
+
+################################ HELP FUNCTIONS ################################
+
+# wrapper to calculate growth matrix
+.calculate_growth_metrics <- function(x, assay.type, time.col = NULL, ...) {
+
+    # Reshape data and calculate growth metrics
+    assay_data <- meltSE(x, assay.type = assay.type, 
+                         add.col = time.col, row.name = "Feature_ID")
+    assay_data <- assay_data %>%
+        arrange( !!sym(time.col) ) %>%
+        group_by(Feature_ID) %>%
+        mutate(
+            time_lag = !!sym(time.col) - lag( !!sym(time.col) ), 
+            growth_diff =!!sym(assay.type) - lag(!!sym(assay.type)),
+            growth_rate = (!!sym(assay.type) - lag(!!sym(assay.type))) / lag(!!sym(assay.type)),
+            var_abund = (!!sym(assay.type) - lag(!!sym(assay.type))) / time_lag
+        )
+    return(assay_data)
+}
+
+.clean_growth_metrics <- function(grwt, time.col = NULL, ...) {
+    # Calculate max growth
+    maxgrs <- grwt %>%
+        summarize(max_growth = max(growth_diff, na.rm = TRUE))
+    # Merge growth data with max growth
+    grs.all <- merge(grwt, maxgrs, by = "Feature_ID")
+    # Add 'ismax' column indicating if the growth is the maximum
+    grs.all <- grs.all %>%
+        mutate(ismax = ifelse(growth_diff == max_growth, TRUE, FALSE))
+    # Clean and abbreviate FeatureID names
+    grs.all$Feature_IDabb <- toupper(abbreviate(grs.all$Feature_ID, 
+                                               minlength = 3, 
+                                               method = "both.sides"))
+    # Create 'Feature.time' column combining abbreviation and time information
+    grs.all$Feature_time <- paste0(grs.all$Feature_IDabb, " ", 
+                                   grs.all[[time.col]], "h")
+
+    return(grs.all)
+}

R/utils.R

@@ -166,3 +166,4 @@
 .check_assay_present <- mia:::.check_assay_present
 .add_values_to_colData <- mia:::.add_values_to_colData
 .check_and_get_altExp <- mia:::.check_and_get_altExp
+.add_values_to_metadata <- mia:::.add_values_to_metadata

tests/testthat/test-getShortTermChange.R

@@ -0,0 +1,25 @@
+test_that("getShortTermChange", {
+    library(SummarizedExperiment)
+    # Load dataset
+    data(minimalgut)
+    tse <- minimalgut
+    # Check if the function handles empty input
+    empty_se <- SummarizedExperiment()
+    expect_error(getShortTermChange(empty_se), 
+                 "No data available in `x`")
+
+    # Should still return a dataframe
+    short_time_labels <- c("74.5h", "173h", "438h", "434h", "390h")
+    # Subset samples by Time_label and StudyIdentifier 
+    tse_filtered <- tse[, !(tse$Time_label %in% short_time_labels)]
+    tse_filtered <- tse_filtered[, (tse_filtered$StudyIdentifier == "Bioreactor A")]
+
+    expect_true(all(!(tse_filtered$Time_label %in% short_time_labels)))
+
+    result <- getShortTermChange(tse_filtered, time.col = "Time.hr")
+    # Expected output is a dataframe
+    expect_true(is.data.frame(result))  
+    expect_true("growth_diff" %in% colnames(result))
+    # Test some expected properties (e.g., that growth_diff isn't all NAs)
+    expect_false(all(is.na(result$growth_diff)))
+})