diff --git a/R/classifySingleR.R b/R/classifySingleR.R
index f27e70a..3a3b747 100644
--- a/R/classifySingleR.R
+++ b/R/classifySingleR.R
@@ -12,7 +12,8 @@
 #' @param tune.thresh A numeric scalar specifying the maximum difference from the maximum correlation to use in fine-tuning.
 #' @param sd.thresh Deprecated and ignored.
 #' @param assay.type Integer scalar or string specifying the matrix of expression values to use if \code{test} is a \linkS4class{SummarizedExperiment}.
-#' @param check.missing Logical scalar indicating whether rows should be checked for missing values (and if found, removed).
+#' @param check.missing Logical scalar indicating whether rows should be checked for missing values.
+#' If true and any missing values are found, the rows containing these values are silently removed.
 #' @param prune A logical scalar indicating whether label pruning should be performed.
 #' @param num.threads Integer scalar specifying the number of threads to use for classification.
 #' @param BPPARAM A \linkS4class{BiocParallelParam} object specifying the parallelization scheme to use for \code{NA} scanning, when \code{check.missing=TRUE}.
diff --git a/R/trainSingleR.R b/R/trainSingleR.R
index 49bafff..fe56ec5 100644
--- a/R/trainSingleR.R
+++ b/R/trainSingleR.R
@@ -43,7 +43,8 @@
 #' @param recompute Deprecated and ignored.
 #' @param assay.type An integer scalar or string specifying the assay of \code{ref} containing the relevant expression matrix,
 #' if \code{ref} is a \linkS4class{SummarizedExperiment} object (or is a list that contains one or more such objects).
-#' @param check.missing Logical scalar indicating whether rows should be checked for missing values (and if found, removed).
+#' @param check.missing Logical scalar indicating whether rows should be checked for missing values.
+#' If true and any missing values are found, the rows containing these values are silently removed.
 #' @param BNPARAM Deprecated and ignored.
 #' @param approximate Logical scalar indicating whether a faster approximate method should be used to compute the quantile.
 #' @param num.threads Integer scalar specifying the number of threads to use for index building.
diff --git a/man/classifySingleR.Rd b/man/classifySingleR.Rd
index 218193c..51a71f8 100644
--- a/man/classifySingleR.Rd
+++ b/man/classifySingleR.Rd
@@ -38,7 +38,8 @@ Alternatively, a List of Lists produced by \code{\link{trainSingleR}} for multip
 
 \item{assay.type}{Integer scalar or string specifying the matrix of expression values to use if \code{test} is a \linkS4class{SummarizedExperiment}.}
 
-\item{check.missing}{Logical scalar indicating whether rows should be checked for missing values (and if found, removed).}
+\item{check.missing}{Logical scalar indicating whether rows should be checked for missing values.
+If true and any missing values are found, the rows containing these values are silently removed.}
 
 \item{num.threads}{Integer scalar specifying the number of threads to use for classification.}
 
diff --git a/man/getClassicMarkers.Rd b/man/getClassicMarkers.Rd
index d623471..a04662a 100644
--- a/man/getClassicMarkers.Rd
+++ b/man/getClassicMarkers.Rd
@@ -32,7 +32,8 @@ Each element should contain a character vector or factor specifying the label fo
 \item{assay.type}{An integer scalar or string specifying the assay of \code{ref} containing the relevant expression matrix,
 if \code{ref} is a \linkS4class{SummarizedExperiment} object (or is a list that contains one or more such objects).}
 
-\item{check.missing}{Logical scalar indicating whether rows should be checked for missing values (and if found, removed).}
+\item{check.missing}{Logical scalar indicating whether rows should be checked for missing values.
+If true and any missing values are found, the rows containing these values are silently removed.}
 
 \item{de.n}{An integer scalar specifying the number of DE genes to use.
 Defaults to \code{500 * (2/3) ^ log2(N)} where \code{N} is the number of unique labels.}
diff --git a/man/trainSingleR.Rd b/man/trainSingleR.Rd
index 6701649..af34530 100644
--- a/man/trainSingleR.Rd
+++ b/man/trainSingleR.Rd
@@ -80,7 +80,8 @@ By default, all genes in \code{ref} are used.}
 \item{assay.type}{An integer scalar or string specifying the assay of \code{ref} containing the relevant expression matrix,
 if \code{ref} is a \linkS4class{SummarizedExperiment} object (or is a list that contains one or more such objects).}
 
-\item{check.missing}{Logical scalar indicating whether rows should be checked for missing values (and if found, removed).}
+\item{check.missing}{Logical scalar indicating whether rows should be checked for missing values.
+If true and any missing values are found, the rows containing these values are silently removed.}
 
 \item{approximate}{Logical scalar indicating whether a faster approximate method should be used to compute the quantile.}
 
diff --git a/tests/testthat/test-classify.R b/tests/testthat/test-classify.R
index fb596be..4ad16e0 100644
--- a/tests/testthat/test-classify.R
+++ b/tests/testthat/test-classify.R
@@ -86,7 +86,7 @@ test_that("classifySingleR behaves with missing values", {
     logcounts(sce)[1,1] <- NA
 
     Q <- 0.8
-    expect_warning(out <- classifySingleR(sce, trained, fine.tune=FALSE, quantile=Q), 'missing values')
+    out <- classifySingleR(sce, trained, fine.tune=FALSE, quantile=Q)
     ref <- classifySingleR(test, trained, fine.tune=FALSE, quantile=Q)
     expect_identical(out, ref)
 })
diff --git a/tests/testthat/test-train.R b/tests/testthat/test-train.R
index 027bd04..03ead0f 100644
--- a/tests/testthat/test-train.R
+++ b/tests/testthat/test-train.R
@@ -125,7 +125,7 @@ test_that("trainSingleR strips out NAs", {
     sce <- training
     logcounts(sce)[1,1] <- NA
 
-    expect_warning(out <- trainSingleR(sce, sce$label), "missing values")
+    out <- trainSingleR(sce, sce$label)
     ref <- trainSingleR(sce[-1,], sce$label)
 
     expect_identical(out$ref, ref$ref)