Merge pull request #14 from biobakery/devel

Allow assay_type
biobakery · Jan 30, 2025 · c52f782 · c52f782
2 parents 5be8dba + 76bde7d
commit c52f782
Show file tree

Hide file tree

Showing 7 changed files with 25 additions and 14 deletions.
diff --git a/DESCRIPTION b/DESCRIPTION
@@ -2,7 +2,7 @@ Package: maaslin3
 Title: "Refining and extending generalized multivariate linear models
         for meta-omic association discovery"
 Year: 2024
-Version: 0.99.3
+Version: 1.0.0
 Authors@R: c(
   person("William", "Nickols", email = "[email protected]", role = c("aut", "cre"), comment=c(ORCID="0000-0001-8214-9746")), 
   person("Jacob", "Nearing", email = "[email protected]", role = "aut")

diff --git a/R/maaslin3.R b/R/maaslin3.R
@@ -2659,7 +2659,8 @@ maaslin3 <- function(input_data,
                     save_models = FALSE,
                     save_plots_rds = FALSE,
                     verbosity = 'FINEST',
-                    summary_plot_balanced = FALSE) {
+                    summary_plot_balanced = FALSE,
+                    assay_type = 1) {
     match.arg(verbosity, c("FINEST", "FINER", "FINE", "DEBUG", "INFO",
                             "WARN", "ERROR"))
     logging::logReset()
@@ -2679,7 +2680,7 @@ maaslin3 <- function(input_data,
 
     if (inherits(input_data, "SummarizedExperiment")) {
         summarized_experiment_out <-
-            maaslin_read_summarized_experiment_data(input_data)
+            maaslin_read_summarized_experiment_data(input_data, assay_type)
 
         input_data <- summarized_experiment_out[['data']]
         input_metadata <- summarized_experiment_out[['metadata']]

diff --git a/R/utility_scripts.R b/R/utility_scripts.R
@@ -825,12 +825,12 @@ preprocess_dna_mtx <- function(dna_table, rna_table) {
 # Read from SummarizedExperiment #
 ##################################
 
-maaslin_read_summarized_experiment_data <- function(summarized_experiment) {
+maaslin_read_summarized_experiment_data <- function(summarized_experiment, assay_type = 1) {
     if (!inherits(summarized_experiment, "SummarizedExperiment")) {
         stop("Input must be a SummarizedExperiment object")
     }
 
-    data <- as.data.frame(t(SummarizedExperiment::assay(summarized_experiment)))
+    data <- as.data.frame(t(SummarizedExperiment::assay(summarized_experiment, assay_type)))
     metadata <- as.data.frame(
         SummarizedExperiment::colData(summarized_experiment))
     return(list(

diff --git a/README.md b/README.md
@@ -651,7 +651,7 @@ significant associations.
 * `max_pngs` (default `30`): The top `max_pngs` significant associations
 will be plotted.
 
-#### Technical parameters ####
+#### Technical/miscellaneous parameters ####
 
 * `cores` (default `1`): How many cores to use when fitting models.
 (Using multiple cores will likely be faster only for large datasets or
@@ -661,6 +661,8 @@ save them to an RData file.
 * `verbosity` (default `'FINEST'`): The level of verbosity for the 
 `logging` package.
 * `save_plots_rds` (default `FALSE`): Whether to save the plots as RDS files.
+* `assay_type` (default `1`): A string or index to select the assay when using
+a `SummarizedExperiment` object.
 
 ## Troubleshooting ##
 

diff --git a/man/maaslin3.Rd b/man/maaslin3.Rd
@@ -19,7 +19,7 @@ This wrapper for all MaAsLin 3 steps finds abundance and prevalence
 }
 \usage{
 maaslin3(input_data,
-    input_metadata,
+    input_metadata = NULL,
     output,
     formula = NULL,
     fixed_effects = NULL,
@@ -59,7 +59,8 @@ maaslin3(input_data,
     save_models = FALSE,
     save_plots_rds = FALSE,
     verbosity = 'FINEST',
-    summary_plot_balanced = FALSE)
+    summary_plot_balanced = FALSE,
+    assay_type = 1)
 }
 \arguments{
     \item{input_data}{A data frame of feature abundances or read counts, a
@@ -222,6 +223,8 @@ maaslin3(input_data,
     \code{coef_plot_vars} where N is equal to: 
     \code{ceiling(summary_plot_first_n/length(coef_plot_vars))}. Will error
     if \code{coef_plot_vars} = \code{NULL}}
+    \item{assay_type}{A string or index to select the assay when using
+    a \code{SummarizedExperiment} object}
 }
 \value{
     A list containing the following items:

diff --git a/tests/testthat/test_Maaslin3.R b/tests/testthat/test_Maaslin3.R
@@ -61,12 +61,14 @@ fit_out <- maaslin3(input_data = se,
                     verbosity = 'WARN')
 
 tse <- TreeSummarizedExperiment::TreeSummarizedExperiment(
-    assays = list(taxa_table = t(taxa_table)),
+    assays = list(taxa_table_junk = matrix(0, nrow = ncol(taxa_table), ncol = nrow(taxa_table)),
+                  another_taxa_table = t(taxa_table)),
     colData = metadata
 )
 
+metadata_df <- as(metadata, "DataFrame")
 fit_out <- maaslin3(input_data = tse, 
-                    input_metadata = metadata, 
+                    input_metadata = metadata_df, 
                     output = output_tmp, 
                     normalization = 'TSS', 
                     transform = 'LOG', 
@@ -79,9 +81,9 @@ fit_out <- maaslin3(input_data = tse,
                     median_comparison_abundance = TRUE, 
                     median_comparison_prevalence = FALSE, 
                     cores=1, 
-                    verbosity = 'WARN')
+                    verbosity = 'WARN',
+                    assay_type = 'another_taxa_table')
 
-metadata <- as(metadata, "DataFrame")
 fit_out <- maaslin3(input_data = tse, 
                     input_metadata = metadata, 
                     output = output_tmp, 
@@ -96,6 +98,7 @@ fit_out <- maaslin3(input_data = tse,
                     median_comparison_abundance = TRUE, 
                     median_comparison_prevalence = FALSE, 
                     cores=1, 
-                    verbosity = 'WARN')
+                    verbosity = 'WARN',
+                    assay_type = 2)
 
 unlink(output_tmp, recursive = T)
diff --git a/vignettes/maaslin3_manual.Rmd b/vignettes/maaslin3_manual.Rmd
@@ -889,7 +889,7 @@ significant associations.
 * `max_pngs` (default `30`): The top `max_pngs` significant associations
 will be plotted.
 
-#### Technical parameters ####
+#### Technical/miscellaneous parameters ####
 
 * `cores` (default `1`): How many cores to use when fitting models.
 (Using multiple cores will likely be faster only for large datasets or
@@ -899,6 +899,8 @@ save them to an RData file.
 * `verbosity` (default `'FINEST'`): The level of verbosity for the 
 `logging` package.
 * `save_plots_rds` (default `FALSE`): Whether to save the plots as RDS files.
+* `assay_type` (default `1`): A string or index to select the assay when using
+a `SummarizedExperiment` object.
 
 ## Tool comparison ##