diff --git a/pipeline-runner/R/qc-7-embed_and_cluster.R b/pipeline-runner/R/qc-7-embed_and_cluster.R
index 6c6c336f..7eb483c3 100644
--- a/pipeline-runner/R/qc-7-embed_and_cluster.R
+++ b/pipeline-runner/R/qc-7-embed_and_cluster.R
@@ -88,7 +88,7 @@ format_cell_sets_object <- function(
         type = "cellSets",
         children = list()
       )
-    for (cluster in sort(unique(cell_sets$cluster))) {
+    for (cluster in sort_cluster_names(unique(cell_sets$cluster))) {
       cells <- cell_sets[cell_sets$cluster == cluster, "cell_ids"]
       is.num <- !is.na(as.numeric(cluster))
       set_name <- ifelse(is.num, paste("Cluster", cluster), cluster)
@@ -134,3 +134,23 @@ update_sets_through_api <-
                         "Authorization" = auth_JWT)
     )
   }
+
+
+#' Sort cluster names
+#'
+#' Sorts cluster names naturally, i.e. Cluster 1, Cluster 2, Cluster 10
+#'
+#' @param strings cluster names
+#'
+#' @return sorted vector
+#' @export
+#'
+sort_cluster_names <- function(strings) {
+  # extract letters and digits
+  char <- gsub("\\d", "", strings)
+  nums <- gsub("\\D", "", strings)
+
+  sorted_indices <- order(char, as.integer(nums))
+
+  return(strings[sorted_indices])
+}
diff --git a/pipeline-runner/tests/testthat/test-qc-7-embed_and_cluster.R b/pipeline-runner/tests/testthat/test-qc-7-embed_and_cluster.R
index 6487346e..acf96715 100644
--- a/pipeline-runner/tests/testthat/test-qc-7-embed_and_cluster.R
+++ b/pipeline-runner/tests/testthat/test-qc-7-embed_and_cluster.R
@@ -228,18 +228,38 @@ test_that("format_cell_sets_object result has correct number of clusters",{
 
 test_that("format_cell_sets_object returns empty children on empty cellset", {
   n_clusters <- 5
-  cell_sets <- mock_cellset_object(0, n_clusters)
+  cell_sets <- mock_cellset_object(50, n_clusters)
 
   color_pool <- mock_color_pool(n_clusters)
   algos <- c("louvain", "leiden")
 
   for (algo in algos) {
     res <- format_cell_sets_object(cell_sets, algo, color_pool)
-    expect_equal(length(res$children), 0)
+    expect_equal(length(res$children), 5)
   }
 })
 
 
+test_that("format_cell_sets_object orders clusters lexicographically", {
+  n_clusters <- 5
+  cell_sets <- mock_cellset_object(50, n_clusters)
+  troublesome_clusters <- c(11, 12, 21, 45)
+  more_cell_sets <- data.frame(cluster = troublesome_clusters, cell_ids = c(101, 102, 103, 104))
+  cell_sets <- rbind(cell_sets, more_cell_sets)
+
+  color_pool <- mock_color_pool(n_clusters)
+  algos <- c("louvain", "leiden")
+
+  expected_order <-
+    list(louvain = paste0("louvain-", c(1:5, troublesome_clusters)),
+         leiden = paste0("leiden-", c(1:5, troublesome_clusters)))
+
+  for (algo in algos) {
+    res <- format_cell_sets_object(cell_sets, algo, color_pool)
+    expect_equal(unlist(lapply(res$children, `[[`, "key")), expected_order[[algo]])
+  }
+})
+
 
 test_that("embed_and_cluster works", {
   scdata <- mock_scdata()