diff --git a/js/index.js b/js/index.js
index a4bfac1a..e9d39a06 100644
--- a/js/index.js
+++ b/js/index.js
@@ -29,7 +29,7 @@ export * from "./suggestCrisprQcFilters.js";
 export * from "./filterCells.js";
 
 export * from "./computeClrm1Factors.js";
-export * from "./logNormCounts.js";
+export * from "./normalizeCounts.js";
 
 export * from "./modelGeneVariances.js";
 export * from "./chooseHvgs.js";
diff --git a/js/logNormCounts.js b/js/normalizeCounts.js
similarity index 88%
rename from js/logNormCounts.js
rename to js/normalizeCounts.js
index 1d377a59..2529e15b 100644
--- a/js/logNormCounts.js
+++ b/js/normalizeCounts.js
@@ -12,6 +12,7 @@ import * as wasm from "./wasm.js";
  * This should have length equal to the number of columns in `x`.
  * If centering is required, it should be applied with {@linkcode centerSizeFactors} - no additional centering is performed here.
  * If `null`, size factors are computed from the centered column sums of `x`.
+ * @param {boolean} [options.log=true] - Whether to perform log-transformation.
  * @param {boolean} [options.allowZeros=false] - Whether size factors of zero should be allowed.
  * If `true`, size factors of zero are converted to the smallest non-zero size factor across all cells.
  * If `false`, an error is raised instead.
@@ -19,10 +20,11 @@ import * as wasm from "./wasm.js";
  * If `true`, size factors of infinity or NaN are converted to the largest non-zero size factor in the dataset or 1, respectively.
  * If `false`, an error is raised instead.
  *
- * @return {ScranMatrix} A matrix of the same type as `x` containing log-transformed normalized expression values.
+ * @return {ScranMatrix} A matrix of the same type as `x` containing normalized expression values.
+ * If `log = true`, the values in the matrix are log-transformed.
  */
-export function logNormCounts(x, options = {}) {
-    const { sizeFactors = null, allowZeros = false, allowNonFinite = false, ...others } = options;
+export function normalizeCounts(x, options = {}) {
+    const { sizeFactors = null, log = true, allowZeros = false, allowNonFinite = false, ...others } = options;
     utils.checkOtherOptions(others);
 
     var sf_data;
@@ -41,7 +43,7 @@ export function logNormCounts(x, options = {}) {
         }
 
         output = gc.call(
-            module => module.normalize_counts(x.matrix, sf_data.offset, allowZeros, allowNonFinite),
+            module => module.normalize_counts(x.matrix, sf_data.offset, log, allowZeros, allowNonFinite),
             x.constructor
         );
 
@@ -58,11 +60,11 @@ export function logNormCounts(x, options = {}) {
 
 /**
  * Center size factors in preparation for log-transformation.
- * This is usually called by {@linkcode logNormCounts} internally, but can also be directly called by users to reconstitute the size factors used in the log-normalized matrix.
+ * This is usually called by {@linkcode normalizeCounts} internally, but can also be directly called by users to reconstitute the size factors used in the log-normalized matrix.
  *
  * @param {TypedArray|WasmArray} sizeFactors - Array of non-negative size factors, one per cell.
  * @param {object} [options={}] - Optional parameters.
- * @param {?(Int32WasmArray|Array|TypedArray)} [options.block=null] - Array containing the block assignment for each cell, see {@linkcode logNormCounts}.
+ * @param {?(Int32WasmArray|Array|TypedArray)} [options.block=null] - Array containing the block assignment for each cell, see {@linkcode normalizeCounts}.
  * @param {boolean} [options.asTypedArray=true] - Whether to return a Float64Array.
  * If `false`, a Float64WasmArray is returned instead.
  * @param {?Float64WasmArray} [options.buffer=null] - Buffer in which to store the output size factors.
diff --git a/src/normalize_counts.cpp b/src/normalize_counts.cpp
index fe0351c5..d824f813 100644
--- a/src/normalize_counts.cpp
+++ b/src/normalize_counts.cpp
@@ -25,7 +25,7 @@ void center_size_factors(size_t n, uintptr_t ptr, bool use_blocks, uintptr_t blo
     }
 }
 
-NumericMatrix normalize_counts(const NumericMatrix& mat, uintptr_t size_factors, bool allow_zero, bool allow_non_finite) {
+NumericMatrix normalize_counts(const NumericMatrix& mat, uintptr_t size_factors, bool log, bool allow_zero, bool allow_non_finite) {
     const double* sfptr = reinterpret_cast<const double*>(size_factors);
     std::vector<double> sf(sfptr, sfptr + mat.ncol());
 
@@ -40,6 +40,7 @@ NumericMatrix normalize_counts(const NumericMatrix& mat, uintptr_t size_factors,
     scran_norm::sanitize_size_factors(sf.size(), sf.data(), san_opt);
 
     scran_norm::NormalizeCountsOptions norm_opt;
+    norm_opt.log = log;
     return NumericMatrix(scran_norm::normalize_counts(mat.ptr, std::move(sf), norm_opt));
 }
 
diff --git a/tests/chooseHvgs.test.js b/tests/chooseHvgs.test.js
index 4ef6770c..f5349567 100644
--- a/tests/chooseHvgs.test.js
+++ b/tests/chooseHvgs.test.js
@@ -10,7 +10,7 @@ test("chooseHvgs works correctly", () => {
     var ncells = 100;
 
     var mat = simulate.simulateMatrix(ngenes, ncells);
-    var norm = scran.logNormCounts(mat);
+    var norm = scran.normalizeCounts(mat);
     var res = scran.modelGeneVariances(norm);
 
     var output = scran.chooseHvgs(res, { number: 101 });
diff --git a/tests/modelGeneVariances.test.js b/tests/modelGeneVariances.test.js
index 473bf29a..22c39d74 100644
--- a/tests/modelGeneVariances.test.js
+++ b/tests/modelGeneVariances.test.js
@@ -10,7 +10,7 @@ test("Variance modelling works as expected", () => {
     var ncells = 100;
 
     var mat = simulate.simulateMatrix(ngenes, ncells);
-    var norm = scran.logNormCounts(mat);
+    var norm = scran.normalizeCounts(mat);
     var res = scran.modelGeneVariances(norm);
 
     // Some cursory tests.
@@ -32,7 +32,7 @@ test("Variance modelling works as expected with blocking", () => {
     var ncells = 100;
 
     var mat = simulate.simulateMatrix(ngenes, ncells);
-    var norm = scran.logNormCounts(mat);
+    var norm = scran.normalizeCounts(mat);
 
     var block = new Array(ncells);
     var half = ncells / 2;
diff --git a/tests/logNormCounts.test.js b/tests/normalizeCounts.test.js
similarity index 91%
rename from tests/logNormCounts.test.js
rename to tests/normalizeCounts.test.js
index 2c74fe1e..5844e774 100644
--- a/tests/logNormCounts.test.js
+++ b/tests/normalizeCounts.test.js
@@ -10,7 +10,7 @@ test("Log-normalization works as expected", () => {
     var ncells = 20;
     var mat = simulate.simulateMatrix(ngenes, ncells);
 
-    var norm = scran.logNormCounts(mat);
+    var norm = scran.normalizeCounts(mat);
     expect(norm.constructor.name).toBe(mat.constructor.name);
     expect(norm.numberOfRows()).toBe(mat.numberOfRows());
     expect(norm.numberOfColumns()).toBe(mat.numberOfColumns());
@@ -40,7 +40,7 @@ test("Log-normalization works as expected with pre-supplied size factors", () =>
         sf[i] = Math.random();
     }
 
-    var norm = scran.logNormCounts(mat, { sizeFactors: sf });
+    var norm = scran.normalizeCounts(mat, { sizeFactors: sf });
     expect(norm.constructor.name).toBe(mat.constructor.name);
     expect(norm.numberOfRows()).toBe(mat.numberOfRows());
     expect(norm.numberOfColumns()).toBe(mat.numberOfColumns());
@@ -89,10 +89,10 @@ test("Log-normalization behaves with zeros", () => {
 
     var empty = new Float64Array(ncells);
     empty.fill(0);
-    expect(() => scran.logNormCounts(mat, { sizeFactors: empty })).toThrow("zero");
+    expect(() => scran.normalizeCounts(mat, { sizeFactors: empty })).toThrow("zero");
 
     // Now trying with allowed zeros.
-    let out = scran.logNormCounts(mat, { sizeFactors: empty, allowZeros: true });
+    let out = scran.normalizeCounts(mat, { sizeFactors: empty, allowZeros: true });
     let ocol = out.column(0);
     let rcol = mat.column(0);
     for (var i = 0; i < ngenes; i++) {
diff --git a/tests/scoreGsdecon.test.js b/tests/scoreGsdecon.test.js
index 6ee6e593..ef8ecb4e 100644
--- a/tests/scoreGsdecon.test.js
+++ b/tests/scoreGsdecon.test.js
@@ -9,7 +9,7 @@ test("scoreGsdecon works as expected with Uint8Array inputs", () => {
     var ngenes = 1000;
     var ncells = 20;
     var mat = simulate.simulateMatrix(ngenes, ncells);
-    var norm = scran.logNormCounts(mat);
+    var norm = scran.normalizeCounts(mat);
 
     let features = new Uint8Array(ngenes);
     features.fill(1, 10, 20);
@@ -25,7 +25,7 @@ test("scoreGsdecon gives different results after scaling", () => {
     var ngenes = 1000;
     var ncells = 20;
     var mat = simulate.simulateMatrix(ngenes, ncells);
-    var norm = scran.logNormCounts(mat);
+    var norm = scran.normalizeCounts(mat);
 
     let features = new Uint8Array(ngenes);
     features.fill(1, 10, 20);
@@ -40,7 +40,7 @@ test("scoreGsdecon works with blocking", () => {
     var ngenes = 1000;
     var ncells = 50;
     var mat = simulate.simulateMatrix(ngenes, ncells);
-    var norm = scran.logNormCounts(mat);
+    var norm = scran.normalizeCounts(mat);
 
     let features = new Uint8Array(ngenes);
     features.fill(1, 0, 5);
diff --git a/tests/scoreMarkers.test.js b/tests/scoreMarkers.test.js
index dae564cd..8a6b08d5 100644
--- a/tests/scoreMarkers.test.js
+++ b/tests/scoreMarkers.test.js
@@ -9,7 +9,7 @@ test("scoreMarkers works as expected", () => {
     var ngenes = 1000;
     var ncells = 20;
     var mat = simulate.simulateMatrix(ngenes, ncells);
-    var norm = scran.logNormCounts(mat);
+    var norm = scran.normalizeCounts(mat);
 
     var groups = [];
     for (var i = 0; i < ncells; i++) {
@@ -40,7 +40,7 @@ test("scoreMarkers works with a log-fold change threshold", () => {
     var ngenes = 1000;
     var ncells = 20;
     var mat = simulate.simulateMatrix(ngenes, ncells);
-    var norm = scran.logNormCounts(mat);
+    var norm = scran.normalizeCounts(mat);
 
     var groups = [];
     for (var i = 0; i < ncells; i++) {
@@ -67,7 +67,7 @@ test("scoreMarkers works after turning off AUCs", () => {
     var ngenes = 1000;
     var ncells = 20;
     var mat = simulate.simulateMatrix(ngenes, ncells);
-    var norm = scran.logNormCounts(mat);
+    var norm = scran.normalizeCounts(mat);
 
     var groups = [];
     for (var i = 0; i < ncells; i++) {
@@ -96,7 +96,7 @@ test("scoreMarkers works with maximum and medians", () => {
     var ngenes = 1000;
     var ncells = 20;
     var mat = simulate.simulateMatrix(ngenes, ncells);
-    var norm = scran.logNormCounts(mat);
+    var norm = scran.normalizeCounts(mat);
 
     var groups = [];
     for (var i = 0; i < ncells; i++) {
@@ -118,7 +118,7 @@ test("scoreMarkers works as expected with blocking", () => {
     var ngenes = 1000;
     var ncells = 20;
     var mat = simulate.simulateMatrix(ngenes, ncells);
-    var norm = scran.logNormCounts(mat);
+    var norm = scran.normalizeCounts(mat);
 
     var groups = [];
     for (var i = 0; i < ncells; i++) {