metameq 2026.2.1__py3-none-any.whl → 2026.2.3__py3-none-any.whl

metameq/tests/test_metadata_extender.py

@@ -475,7 +475,7 @@ class TestMetadataExtender(TestCase):
 
             # Verify metadata file contents - includes failed row when remove_internals=False
             result_df = pandas.read_csv(
-                metadata_files[0], sep="\t", keep_default_na=False)
+                metadata_files[0], sep="\t", dtype=str, keep_default_na=False)
             assert_frame_equal(metadata_df, result_df)
 
             # Find the validation errors file (uses comma separator)
@@ -484,7 +484,7 @@ class TestMetadataExtender(TestCase):
             self.assertEqual(1, len(validation_files))
 
             # Verify validation errors file contents
-            result_validation_df = pandas.read_csv(validation_files[0], sep=",")
+            result_validation_df = pandas.read_csv(validation_files[0], sep=",", dtype=str, keep_default_na=False)
             assert_frame_equal(validation_msgs_df, result_validation_df)
 
             # No fails file should be created when remove_internals=False
@@ -513,7 +513,7 @@ class TestMetadataExtender(TestCase):
             self.assertEqual(1, len(metadata_files))
 
             # Verify metadata has internal cols removed and no failures
-            result_df = pandas.read_csv(metadata_files[0], sep="\t")
+            result_df = pandas.read_csv(metadata_files[0], sep="\t", dtype=str, keep_default_na=False)
             expected_df = pandas.DataFrame({
                 SAMPLE_NAME_KEY: ["sample1", "sample3"],
                 "field_a": ["a1", "a3"]
@@ -526,7 +526,7 @@ class TestMetadataExtender(TestCase):
             self.assertEqual(1, len(fails_files))
 
             # Verify fails file contains the failed row
-            fails_df = pandas.read_csv(fails_files[0], sep=",")
+            fails_df = pandas.read_csv(fails_files[0], sep=",", dtype=str, keep_default_na=False)
             expected_fails_df = pandas.DataFrame({
                 SAMPLE_NAME_KEY: ["sample2"],
                 "field_a": ["a2"],
@@ -593,7 +593,7 @@ class TestMetadataExtender(TestCase):
             self.assertEqual(1, len(metadata_files))
 
             # Verify custom internal cols are removed
-            result_df = pandas.read_csv(metadata_files[0], sep="\t")
+            result_df = pandas.read_csv(metadata_files[0], sep="\t", dtype=str, keep_default_na=False)
             expected_df = pandas.DataFrame({
                 SAMPLE_NAME_KEY: ["sample1", "sample2"],
                 "field_a": ["a1", "a2"]
@@ -748,16 +748,15 @@ class TestMetadataExtender(TestCase):
 
     # Tests for _fill_na_if_default
 
-    def test__fill_na_if_default_specific_overrides_settings(self):
+    def test__fill_na_if_default_has_default_in_settings(self):
         """Test that specific_dict default takes precedence over settings_dict."""
         input_df = pandas.DataFrame({
             "field1": ["value1", np.nan, "value3"],
             "field2": [np.nan, "value2", np.nan]
         })
-        specific_dict = {DEFAULT_KEY: "filled"}
-        settings_dict = {DEFAULT_KEY: "unused"}
+        settings_dict = {DEFAULT_KEY: "filled"}
 
-        result = _fill_na_if_default(input_df, specific_dict, settings_dict)
+        result = _fill_na_if_default(input_df, settings_dict)
 
         expected = pandas.DataFrame({
             "field1": ["value1", "filled", "value3"],
@@ -765,18 +764,19 @@ class TestMetadataExtender(TestCase):
         })
         assert_frame_equal(expected, result)
 
-    def test__fill_na_if_default_uses_settings_when_specific_missing(self):
-        """Test that settings_dict default is used when specific_dict has no default."""
+    def test__fill_na_if_default_no_default_in_settings(self):
+        """Test that NaN values are unchanged when no default is in settings."""
         input_df = pandas.DataFrame({
-            "field1": [np.nan]
+            "field1": ["value1", np.nan, "value3"],
+            "field2": [np.nan, "value2", np.nan]
         })
-        specific_dict = {}
-        settings_dict = {DEFAULT_KEY: "settings_default"}
+        settings_dict = {}
 
-        result = _fill_na_if_default(input_df, specific_dict, settings_dict)
+        result = _fill_na_if_default(input_df, settings_dict)
 
         expected = pandas.DataFrame({
-            "field1": ["settings_default"]
+            "field1": ["value1", np.nan, "value3"],
+            "field2": [np.nan, "value2", np.nan]
         })
         assert_frame_equal(expected, result)
 
@@ -1273,14 +1273,13 @@ class TestMetadataExtender(TestCase):
             SAMPLETYPE_SHORTHAND_KEY: ["stool", "stool"],
             QC_NOTE_KEY: ["", ""]
         })
-        global_plus_host_settings_dict = {
-            OVERWRITE_NON_NANS_KEY: False,
-            LEAVE_REQUIREDS_BLANK_KEY: False,
-            DEFAULT_KEY: "not provided"
-        }
+
         # Config is pre-resolved: sample type's metadata_fields already includes
         # host fields merged in, plus sample_type and qiita_sample_type
         host_type_config_dict = {
+            OVERWRITE_NON_NANS_KEY: False,
+            LEAVE_REQUIREDS_BLANK_KEY: False,
+            DEFAULT_KEY: "not provided",
             METADATA_FIELDS_KEY: {
                 "host_field": {
                     DEFAULT_KEY: "host_default",
@@ -1314,7 +1313,7 @@ class TestMetadataExtender(TestCase):
         }
 
         result_df, validation_msgs = _generate_metadata_for_a_sample_type_in_a_host_type(
-            input_df, "stool", global_plus_host_settings_dict, host_type_config_dict)
+            input_df, "stool", host_type_config_dict)
 
         expected_df = pandas.DataFrame({
             SAMPLE_NAME_KEY: ["sample1", "sample2"],
@@ -1337,12 +1336,11 @@ class TestMetadataExtender(TestCase):
             SAMPLETYPE_SHORTHAND_KEY: ["unknown_type"],
             QC_NOTE_KEY: [""]
         })
-        global_plus_host_settings_dict = {
+
+        host_type_config_dict = {
             OVERWRITE_NON_NANS_KEY: False,
             LEAVE_REQUIREDS_BLANK_KEY: False,
-            DEFAULT_KEY: "not provided"
-        }
-        host_type_config_dict = {
+            DEFAULT_KEY: "not provided",
             METADATA_FIELDS_KEY: {},
             SAMPLE_TYPE_SPECIFIC_METADATA_KEY: {
                 "stool": {
@@ -1352,7 +1350,7 @@ class TestMetadataExtender(TestCase):
         }
 
         result_df, validation_msgs = _generate_metadata_for_a_sample_type_in_a_host_type(
-            input_df, "unknown_type", global_plus_host_settings_dict, host_type_config_dict)
+            input_df, "unknown_type", host_type_config_dict)
 
         expected_df = pandas.DataFrame({
             SAMPLE_NAME_KEY: ["sample1"],
@@ -1371,12 +1369,11 @@ class TestMetadataExtender(TestCase):
             SAMPLETYPE_SHORTHAND_KEY: ["stool", "blood", "stool"],
             QC_NOTE_KEY: ["", "", ""]
         })
-        global_plus_host_settings_dict = {
+
+        host_type_config_dict = {
             OVERWRITE_NON_NANS_KEY: False,
             LEAVE_REQUIREDS_BLANK_KEY: False,
-            DEFAULT_KEY: "not provided"
-        }
-        host_type_config_dict = {
+            DEFAULT_KEY: "not provided",
             METADATA_FIELDS_KEY: {},
             SAMPLE_TYPE_SPECIFIC_METADATA_KEY: {
                 "stool": {
@@ -1394,7 +1391,7 @@ class TestMetadataExtender(TestCase):
         }
 
         result_df, validation_msgs = _generate_metadata_for_a_sample_type_in_a_host_type(
-            input_df, "stool", global_plus_host_settings_dict, host_type_config_dict)
+            input_df, "stool", host_type_config_dict)
 
         # Should only have the two stool samples
         self.assertEqual(2, len(result_df))
@@ -1409,12 +1406,11 @@ class TestMetadataExtender(TestCase):
             SAMPLETYPE_SHORTHAND_KEY: ["stool"],
             QC_NOTE_KEY: [""]
         })
-        global_plus_host_settings_dict = {
+
+        host_type_config_dict = {
             OVERWRITE_NON_NANS_KEY: False,
             LEAVE_REQUIREDS_BLANK_KEY: True,
-            DEFAULT_KEY: "not provided"
-        }
-        host_type_config_dict = {
+            DEFAULT_KEY: "not provided",
             METADATA_FIELDS_KEY: {},
             SAMPLE_TYPE_SPECIFIC_METADATA_KEY: {
                 "stool": {
@@ -1429,7 +1425,7 @@ class TestMetadataExtender(TestCase):
         }
 
         result_df, validation_msgs = _generate_metadata_for_a_sample_type_in_a_host_type(
-            input_df, "stool", global_plus_host_settings_dict, host_type_config_dict)
+            input_df, "stool", host_type_config_dict)
 
         self.assertEqual(LEAVE_BLANK_VAL, result_df["required_field"].iloc[0])
 
@@ -1441,12 +1437,11 @@ class TestMetadataExtender(TestCase):
             SAMPLETYPE_SHORTHAND_KEY: ["stool"],
             QC_NOTE_KEY: [""]
         })
-        global_plus_host_settings_dict = {
+
+        host_type_config_dict = {
             OVERWRITE_NON_NANS_KEY: False,
             LEAVE_REQUIREDS_BLANK_KEY: False,
-            DEFAULT_KEY: "global_default"
-        }
-        host_type_config_dict = {
+            DEFAULT_KEY: "global_default",
             METADATA_FIELDS_KEY: {},
             SAMPLE_TYPE_SPECIFIC_METADATA_KEY: {
                 "stool": {
@@ -1461,7 +1456,7 @@ class TestMetadataExtender(TestCase):
         }
 
         result_df, validation_msgs = _generate_metadata_for_a_sample_type_in_a_host_type(
-            input_df, "stool", global_plus_host_settings_dict, host_type_config_dict)
+            input_df, "stool", host_type_config_dict)
 
         # When leave_requireds_blank is False, NaN values get filled with global default
         self.assertEqual("global_default", result_df["required_field"].iloc[0])
@@ -1475,12 +1470,11 @@ class TestMetadataExtender(TestCase):
             QC_NOTE_KEY: [""],
             "existing_field": ["original_value"]
         })
-        global_plus_host_settings_dict = {
+
+        host_type_config_dict = {
             OVERWRITE_NON_NANS_KEY: True,
             LEAVE_REQUIREDS_BLANK_KEY: False,
-            DEFAULT_KEY: "not provided"
-        }
-        host_type_config_dict = {
+            DEFAULT_KEY: "not provided",
             METADATA_FIELDS_KEY: {},
             SAMPLE_TYPE_SPECIFIC_METADATA_KEY: {
                 "stool": {
@@ -1495,7 +1489,7 @@ class TestMetadataExtender(TestCase):
         }
 
         result_df, validation_msgs = _generate_metadata_for_a_sample_type_in_a_host_type(
-            input_df, "stool", global_plus_host_settings_dict, host_type_config_dict)
+            input_df, "stool", host_type_config_dict)
 
         self.assertEqual("new_value", result_df["existing_field"].iloc[0])
 
@@ -1508,12 +1502,11 @@ class TestMetadataExtender(TestCase):
             QC_NOTE_KEY: [""],
             "existing_field": ["original_value"]
         })
-        global_plus_host_settings_dict = {
+
+        host_type_config_dict = {
             OVERWRITE_NON_NANS_KEY: False,
             LEAVE_REQUIREDS_BLANK_KEY: False,
-            DEFAULT_KEY: "not provided"
-        }
-        host_type_config_dict = {
+            DEFAULT_KEY: "not provided",
             METADATA_FIELDS_KEY: {},
             SAMPLE_TYPE_SPECIFIC_METADATA_KEY: {
                 "stool": {
@@ -1528,7 +1521,7 @@ class TestMetadataExtender(TestCase):
         }
 
         result_df, validation_msgs = _generate_metadata_for_a_sample_type_in_a_host_type(
-            input_df, "stool", global_plus_host_settings_dict, host_type_config_dict)
+            input_df, "stool", host_type_config_dict)
 
         self.assertEqual("original_value", result_df["existing_field"].iloc[0])
 
@@ -1540,14 +1533,13 @@ class TestMetadataExtender(TestCase):
             SAMPLETYPE_SHORTHAND_KEY: ["feces"],
             QC_NOTE_KEY: [""]
         })
-        global_plus_host_settings_dict = {
-            OVERWRITE_NON_NANS_KEY: False,
-            LEAVE_REQUIREDS_BLANK_KEY: False,
-            DEFAULT_KEY: "not provided"
-        }
+
         # Config is pre-resolved: alias "feces" has its own metadata_fields
         # that is a copy of "stool"'s resolved fields with sample_type="stool"
         host_type_config_dict = {
+            OVERWRITE_NON_NANS_KEY: False,
+            LEAVE_REQUIREDS_BLANK_KEY: False,
+            DEFAULT_KEY: "not provided",
             METADATA_FIELDS_KEY: {},
             SAMPLE_TYPE_SPECIFIC_METADATA_KEY: {
                 "feces": {
@@ -1590,7 +1582,7 @@ class TestMetadataExtender(TestCase):
         }
 
         result_df, validation_msgs = _generate_metadata_for_a_sample_type_in_a_host_type(
-            input_df, "feces", global_plus_host_settings_dict, host_type_config_dict)
+            input_df, "feces", host_type_config_dict)
 
         self.assertEqual("stool_value", result_df["stool_field"].iloc[0])
         # sample_type should be set to the resolved type "stool"
@@ -1606,17 +1598,15 @@ class TestMetadataExtender(TestCase):
             SAMPLETYPE_SHORTHAND_KEY: ["stool", "stool"],
             QC_NOTE_KEY: ["", ""]
         })
-        settings_dict = {
-            OVERWRITE_NON_NANS_KEY: False,
-            LEAVE_REQUIREDS_BLANK_KEY: False,
-            DEFAULT_KEY: "global_default"
-        }
+
         # Config is pre-resolved: sample type's metadata_fields includes
         # host fields merged in, plus sample_type and qiita_sample_type
         full_flat_config_dict = {
             HOST_TYPE_SPECIFIC_METADATA_KEY: {
                 "human": {
                     DEFAULT_KEY: "human_default",
+                    OVERWRITE_NON_NANS_KEY: False,
+                    LEAVE_REQUIREDS_BLANK_KEY: False,
                     METADATA_FIELDS_KEY: {
                         "host_field": {
                             DEFAULT_KEY: "host_value",
@@ -1652,7 +1642,7 @@ class TestMetadataExtender(TestCase):
         }
 
         result_df, validation_msgs = _generate_metadata_for_a_host_type(
-            input_df, "human", settings_dict, full_flat_config_dict)
+            input_df, "human", full_flat_config_dict)
 
         expected_df = pandas.DataFrame({
             SAMPLE_NAME_KEY: ["sample1", "sample2"],
@@ -1675,14 +1665,13 @@ class TestMetadataExtender(TestCase):
             SAMPLETYPE_SHORTHAND_KEY: ["stool"],
             QC_NOTE_KEY: [""]
         })
-        settings_dict = {
-            OVERWRITE_NON_NANS_KEY: False,
-            LEAVE_REQUIREDS_BLANK_KEY: False,
-            DEFAULT_KEY: "global_default"
-        }
+
         full_flat_config_dict = {
             HOST_TYPE_SPECIFIC_METADATA_KEY: {
                 "human": {
+                    OVERWRITE_NON_NANS_KEY: False,
+                    LEAVE_REQUIREDS_BLANK_KEY: False,
+                    DEFAULT_KEY: "global_default",
                     METADATA_FIELDS_KEY: {},
                     SAMPLE_TYPE_SPECIFIC_METADATA_KEY: {}
                 }
@@ -1690,7 +1679,7 @@ class TestMetadataExtender(TestCase):
         }
 
         result_df, validation_msgs = _generate_metadata_for_a_host_type(
-            input_df, "unknown_host", settings_dict, full_flat_config_dict)
+            input_df, "unknown_host", full_flat_config_dict)
 
         expected_df = pandas.DataFrame({
             SAMPLE_NAME_KEY: ["sample1"],
@@ -1709,14 +1698,13 @@ class TestMetadataExtender(TestCase):
             SAMPLETYPE_SHORTHAND_KEY: ["unknown_sample"],
             QC_NOTE_KEY: [""]
         })
-        settings_dict = {
-            OVERWRITE_NON_NANS_KEY: False,
-            LEAVE_REQUIREDS_BLANK_KEY: False,
-            DEFAULT_KEY: "global_default"
-        }
+
         full_flat_config_dict = {
             HOST_TYPE_SPECIFIC_METADATA_KEY: {
                 "human": {
+                    OVERWRITE_NON_NANS_KEY: False,
+                    LEAVE_REQUIREDS_BLANK_KEY: False,
+                    DEFAULT_KEY: "global_default",
                     METADATA_FIELDS_KEY: {},
                     SAMPLE_TYPE_SPECIFIC_METADATA_KEY: {
                         "stool": {
@@ -1728,7 +1716,7 @@ class TestMetadataExtender(TestCase):
         }
 
         result_df, validation_msgs = _generate_metadata_for_a_host_type(
-            input_df, "human", settings_dict, full_flat_config_dict)
+            input_df, "human", full_flat_config_dict)
 
         expected_df = pandas.DataFrame({
             SAMPLE_NAME_KEY: ["sample1"],
@@ -1747,16 +1735,15 @@ class TestMetadataExtender(TestCase):
             SAMPLETYPE_SHORTHAND_KEY: ["stool", "stool", "stool"],
             QC_NOTE_KEY: ["", "", ""]
         })
-        settings_dict = {
-            OVERWRITE_NON_NANS_KEY: False,
-            LEAVE_REQUIREDS_BLANK_KEY: False,
-            DEFAULT_KEY: "global_default"
-        }
+
         # Config is pre-resolved: sample type's metadata_fields includes
         # host fields merged in, plus sample_type and qiita_sample_type
         full_flat_config_dict = {
             HOST_TYPE_SPECIFIC_METADATA_KEY: {
                 "human": {
+                    OVERWRITE_NON_NANS_KEY: False,
+                    LEAVE_REQUIREDS_BLANK_KEY: False,
+                    DEFAULT_KEY: "global_default",
                     METADATA_FIELDS_KEY: {
                         "human_field": {
                             DEFAULT_KEY: "human_value",
@@ -1785,6 +1772,9 @@ class TestMetadataExtender(TestCase):
                     }
                 },
                 "mouse": {
+                    OVERWRITE_NON_NANS_KEY: False,
+                    LEAVE_REQUIREDS_BLANK_KEY: False,
+                    DEFAULT_KEY: "global_default",
                     METADATA_FIELDS_KEY: {},
                     SAMPLE_TYPE_SPECIFIC_METADATA_KEY: {}
                 }
@@ -1792,7 +1782,7 @@ class TestMetadataExtender(TestCase):
         }
 
         result_df, validation_msgs = _generate_metadata_for_a_host_type(
-            input_df, "human", settings_dict, full_flat_config_dict)
+            input_df, "human", full_flat_config_dict)
 
         expected_df = pandas.DataFrame({
             SAMPLE_NAME_KEY: ["sample1", "sample3"],
@@ -1813,17 +1803,15 @@ class TestMetadataExtender(TestCase):
             SAMPLETYPE_SHORTHAND_KEY: ["stool"],
             QC_NOTE_KEY: [""]
         })
-        settings_dict = {
-            OVERWRITE_NON_NANS_KEY: False,
-            LEAVE_REQUIREDS_BLANK_KEY: False,
-            DEFAULT_KEY: "global_default"
-        }
+
         # Config is pre-resolved: sample type's metadata_fields includes
         # host fields merged in, plus sample_type and qiita_sample_type
         full_flat_config_dict = {
             HOST_TYPE_SPECIFIC_METADATA_KEY: {
                 "human": {
                     DEFAULT_KEY: "human_specific_default",
+                    OVERWRITE_NON_NANS_KEY: False,
+                    LEAVE_REQUIREDS_BLANK_KEY: False,
                     METADATA_FIELDS_KEY: {},
                     SAMPLE_TYPE_SPECIFIC_METADATA_KEY: {
                         "stool": {
@@ -1850,7 +1838,7 @@ class TestMetadataExtender(TestCase):
         }
 
         result_df, validation_msgs = _generate_metadata_for_a_host_type(
-            input_df, "human", settings_dict, full_flat_config_dict)
+            input_df, "human", full_flat_config_dict)
 
         expected_df = pandas.DataFrame({
             SAMPLE_NAME_KEY: ["sample1"],
@@ -1871,17 +1859,14 @@ class TestMetadataExtender(TestCase):
             SAMPLETYPE_SHORTHAND_KEY: ["stool"],
             QC_NOTE_KEY: [""]
         })
-        settings_dict = {
-            OVERWRITE_NON_NANS_KEY: False,
-            LEAVE_REQUIREDS_BLANK_KEY: False,
-            DEFAULT_KEY: "global_default"
-        }
         # Config is pre-resolved: sample type's metadata_fields includes
         # host fields merged in, plus sample_type and qiita_sample_type
         full_flat_config_dict = {
             HOST_TYPE_SPECIFIC_METADATA_KEY: {
                 "human": {
-                    # No DEFAULT_KEY here
+                    OVERWRITE_NON_NANS_KEY: False,
+                    LEAVE_REQUIREDS_BLANK_KEY: False,
+                    DEFAULT_KEY: "global_default",
                     METADATA_FIELDS_KEY: {},
                     SAMPLE_TYPE_SPECIFIC_METADATA_KEY: {
                         "stool": {
@@ -1908,7 +1893,7 @@ class TestMetadataExtender(TestCase):
         }
 
         result_df, validation_msgs = _generate_metadata_for_a_host_type(
-            input_df, "human", settings_dict, full_flat_config_dict)
+            input_df, "human", full_flat_config_dict)
 
         expected_df = pandas.DataFrame({
             SAMPLE_NAME_KEY: ["sample1"],
@@ -1939,6 +1924,9 @@ class TestMetadataExtender(TestCase):
             OVERWRITE_NON_NANS_KEY: False,
             HOST_TYPE_SPECIFIC_METADATA_KEY: {
                 "human": {
+                    DEFAULT_KEY: "global_default",
+                    LEAVE_REQUIREDS_BLANK_KEY: False,
+                    OVERWRITE_NON_NANS_KEY: False,
                     METADATA_FIELDS_KEY: {
                         "host_field": {
                             DEFAULT_KEY: "host_value",
@@ -2005,6 +1993,9 @@ class TestMetadataExtender(TestCase):
             OVERWRITE_NON_NANS_KEY: False,
             HOST_TYPE_SPECIFIC_METADATA_KEY: {
                 "human": {
+                    DEFAULT_KEY: "global_default",
+                    LEAVE_REQUIREDS_BLANK_KEY: False,
+                    OVERWRITE_NON_NANS_KEY: False,
                     METADATA_FIELDS_KEY: {
                         "human_field": {
                             DEFAULT_KEY: "human_value",
@@ -2051,6 +2042,9 @@ class TestMetadataExtender(TestCase):
                     }
                 },
                 "mouse": {
+                    DEFAULT_KEY: "global_default",
+                    LEAVE_REQUIREDS_BLANK_KEY: False,
+                    OVERWRITE_NON_NANS_KEY: False,
                     METADATA_FIELDS_KEY: {
                         "mouse_field": {
                             DEFAULT_KEY: "mouse_value",
@@ -2182,6 +2176,9 @@ class TestMetadataExtender(TestCase):
             OVERWRITE_NON_NANS_KEY: False,
             HOST_TYPE_SPECIFIC_METADATA_KEY: {
                 "human": {
+                    DEFAULT_KEY: "global_default",
+                    LEAVE_REQUIREDS_BLANK_KEY: True,  # This causes required fields to get LEAVE_BLANK_VAL
+                    OVERWRITE_NON_NANS_KEY: False,
                     METADATA_FIELDS_KEY: {},
                     SAMPLE_TYPE_SPECIFIC_METADATA_KEY: {
                         "stool": {
@@ -2506,6 +2503,9 @@ class TestMetadataExtender(TestCase):
             OVERWRITE_NON_NANS_KEY: False,
             HOST_TYPE_SPECIFIC_METADATA_KEY: {
                 "human": {
+                    DEFAULT_KEY: "not provided",
+                    LEAVE_REQUIREDS_BLANK_KEY: False,
+                    OVERWRITE_NON_NANS_KEY: False,
                     METADATA_FIELDS_KEY: {
                         "host_field": {
                             DEFAULT_KEY: "host_value",
@@ -2580,6 +2580,9 @@ class TestMetadataExtender(TestCase):
             },
             HOST_TYPE_SPECIFIC_METADATA_KEY: {
                 "human": {
+                    DEFAULT_KEY: "not provided",
+                    LEAVE_REQUIREDS_BLANK_KEY: False,
+                    OVERWRITE_NON_NANS_KEY: False,
                     METADATA_FIELDS_KEY: {},
                     SAMPLE_TYPE_SPECIFIC_METADATA_KEY: {
                         "stool": {
@@ -2639,6 +2642,9 @@ class TestMetadataExtender(TestCase):
             },
             HOST_TYPE_SPECIFIC_METADATA_KEY: {
                 "human": {
+                    DEFAULT_KEY: "not provided",
+                    LEAVE_REQUIREDS_BLANK_KEY: False,
+                    OVERWRITE_NON_NANS_KEY: False,
                     METADATA_FIELDS_KEY: {},
                     SAMPLE_TYPE_SPECIFIC_METADATA_KEY: {
                         "stool": {
@@ -2687,6 +2693,9 @@ class TestMetadataExtender(TestCase):
             OVERWRITE_NON_NANS_KEY: False,
             HOST_TYPE_SPECIFIC_METADATA_KEY: {
                 "human": {
+                    DEFAULT_KEY: "not provided",
+                    LEAVE_REQUIREDS_BLANK_KEY: False,
+                    OVERWRITE_NON_NANS_KEY: False,
                     METADATA_FIELDS_KEY: {},
                     SAMPLE_TYPE_SPECIFIC_METADATA_KEY: {}
                 }
@@ -2721,6 +2730,9 @@ class TestMetadataExtender(TestCase):
             OVERWRITE_NON_NANS_KEY: False,
             HOST_TYPE_SPECIFIC_METADATA_KEY: {
                 "human": {
+                    DEFAULT_KEY: "not provided",
+                    LEAVE_REQUIREDS_BLANK_KEY: False,
+                    OVERWRITE_NON_NANS_KEY: False,
                     METADATA_FIELDS_KEY: {},
                     SAMPLE_TYPE_SPECIFIC_METADATA_KEY: {
                         "stool": {
@@ -2781,6 +2793,9 @@ class TestMetadataExtender(TestCase):
             },
             HOST_TYPE_SPECIFIC_METADATA_KEY: {
                 "human": {
+                    DEFAULT_KEY: "not provided",
+                    LEAVE_REQUIREDS_BLANK_KEY: False,
+                    OVERWRITE_NON_NANS_KEY: False,
                     METADATA_FIELDS_KEY: {},
                     SAMPLE_TYPE_SPECIFIC_METADATA_KEY: {
                         "stool": {
@@ -3319,7 +3334,7 @@ class TestMetadataExtender(TestCase):
             self.assertEqual(1, len(output_files))
 
             # Read and verify contents (keep_default_na=False preserves empty strings)
-            result_df = pandas.read_csv(output_files[0], sep="\t", keep_default_na=False)
+            result_df = pandas.read_csv(output_files[0], sep="\t", dtype=str, keep_default_na=False)
             expected_df = input_df
             assert_frame_equal(expected_df, result_df)
 
@@ -3343,7 +3358,7 @@ class TestMetadataExtender(TestCase):
             self.assertEqual(1, len(output_files))
 
             # Verify main output has internal cols removed and no failures
-            result_df = pandas.read_csv(output_files[0], sep="\t")
+            result_df = pandas.read_csv(output_files[0], sep="\t", dtype=str, keep_default_na=False)
             expected_df = pandas.DataFrame({
                 SAMPLE_NAME_KEY: ["sample1", "sample3"],
                 "field_a": ["a1", "a3"]
@@ -3355,7 +3370,7 @@ class TestMetadataExtender(TestCase):
             self.assertEqual(1, len(fails_files))
 
             # Verify fails file contains the failed row
-            fails_df = pandas.read_csv(fails_files[0], sep=",")
+            fails_df = pandas.read_csv(fails_files[0], sep=",", dtype=str, keep_default_na=False)
             expected_fails_df = pandas.DataFrame({
                 SAMPLE_NAME_KEY: ["sample2"],
                 "field_a": ["a2"],
@@ -3432,7 +3447,7 @@ class TestMetadataExtender(TestCase):
             self.assertEqual(1, len(output_files))
 
             # Read and verify contents (keep_default_na=False preserves empty strings)
-            result_df = pandas.read_csv(output_files[0], sep=",", keep_default_na=False)
+            result_df = pandas.read_csv(output_files[0], sep=",", dtype=str, keep_default_na=False)
             expected_df = input_df
             assert_frame_equal(expected_df, result_df)
 
@@ -3454,14 +3469,14 @@ class TestMetadataExtender(TestCase):
             # Main output file should have only headers (empty data)
             output_files = glob.glob(os.path.join(tmpdir, "*_test_output.txt"))
             self.assertEqual(1, len(output_files))
-            result_df = pandas.read_csv(output_files[0], sep="\t")
+            result_df = pandas.read_csv(output_files[0], sep="\t", dtype=str, keep_default_na=False)
             self.assertTrue(result_df.empty)
             self.assertEqual([SAMPLE_NAME_KEY, "field_a"], list(result_df.columns))
 
             # Fails file should have both rows
             fails_files = glob.glob(os.path.join(tmpdir, "*_test_output_fails.csv"))
             self.assertEqual(1, len(fails_files))
-            fails_df = pandas.read_csv(fails_files[0], sep=",")
+            fails_df = pandas.read_csv(fails_files[0], sep=",", dtype=str, keep_default_na=False)
             self.assertEqual(2, len(fails_df))
 
     # Tests for get_extended_metadata_from_df_and_yaml
@@ -3606,7 +3621,7 @@ class TestMetadataExtender(TestCase):
             # Verify main output file was created (internal cols removed by default)
             output_files = glob.glob(os.path.join(tmpdir, "*_test_output.txt"))
             self.assertEqual(1, len(output_files))
-            output_df = pandas.read_csv(output_files[0], sep="\t")
+            output_df = pandas.read_csv(output_files[0], sep="\t", dtype=str, keep_default_na=False)
             expected_output_df = pandas.DataFrame({
                 SAMPLE_NAME_KEY: ["sample1", "sample2"],
                 "body_product": ["UBERON:feces", "UBERON:feces"],
@@ -3679,7 +3694,7 @@ class TestMetadataExtender(TestCase):
             # Verify main output file excludes failure rows
             output_files = glob.glob(os.path.join(tmpdir, "*_test_output.txt"))
             self.assertEqual(1, len(output_files))
-            output_df = pandas.read_csv(output_files[0], sep="\t")
+            output_df = pandas.read_csv(output_files[0], sep="\t", dtype=str, keep_default_na=False)
             expected_output_df = pandas.DataFrame({
                 SAMPLE_NAME_KEY: ["sample1", "sample3"],
                 "body_product": ["UBERON:feces", "UBERON:feces"],
@@ -3694,7 +3709,7 @@ class TestMetadataExtender(TestCase):
             # Verify fails file contains the failed row
             fails_files = glob.glob(os.path.join(tmpdir, "*_test_output_fails.csv"))
             self.assertEqual(1, len(fails_files))
-            fails_df = pandas.read_csv(fails_files[0], sep=",")
+            fails_df = pandas.read_csv(fails_files[0], sep=",", dtype=str, keep_default_na=False)
             expected_fails_df = pandas.DataFrame({
                 SAMPLE_NAME_KEY: ["sample2"],
                 "body_product": ["not provided"],
@@ -3765,7 +3780,7 @@ class TestMetadataExtender(TestCase):
             validation_files = glob.glob(
                 os.path.join(tmpdir, "*_test_output_validation_errors.csv"))
             self.assertEqual(1, len(validation_files))
-            validation_df = pandas.read_csv(validation_files[0], sep=",")
+            validation_df = pandas.read_csv(validation_files[0], sep=",", dtype=str, keep_default_na=False)
             expected_validation_df = pandas.DataFrame({
                 "sample_name": ["sample1"],
                 "field_name": ["restricted_field"],
@@ -3806,7 +3821,7 @@ class TestMetadataExtender(TestCase):
             # Verify main output file includes internal columns
             output_files = glob.glob(os.path.join(tmpdir, "*_test_output.txt"))
             self.assertEqual(1, len(output_files))
-            output_df = pandas.read_csv(output_files[0], sep="\t", keep_default_na=False)
+            output_df = pandas.read_csv(output_files[0], sep="\t", dtype=str, keep_default_na=False)
             expected_output_df = pandas.DataFrame({
                 SAMPLE_NAME_KEY: ["sample1"],
                 "body_product": ["UBERON:feces"],
@@ -3829,6 +3844,7 @@ class TestMetadataExtender(TestCase):
 
     TEST_METADATA_CSV_FP = path.join(TEST_DIR, "data/test_metadata.csv")
     TEST_METADATA_TXT_FP = path.join(TEST_DIR, "data/test_metadata.txt")
+    TEST_METADATA_XLSX_FP = path.join(TEST_DIR, "data/test_metadata.xlsx")
     TEST_METADATA_WITH_ERRORS_FP = path.join(
         TEST_DIR, "data/test_metadata_with_errors.csv")
     TEST_STUDY_CONFIG_WITH_VALIDATION_FP = path.join(
@@ -3847,6 +3863,7 @@ class TestMetadataExtender(TestCase):
                 "body_product": ["UBERON:feces", "UBERON:feces"],
                 "body_site": ["gut", "gut"],
                 "description": ["human sample", "human sample"],
+                "dna_extracted": ["TRUE", "FALSE"],
                 "host_common_name": ["human", "human"],
                 QIITA_SAMPLE_TYPE: ["stool", "stool"],
                 SAMPLE_TYPE_KEY: ["stool", "stool"],
@@ -3861,12 +3878,13 @@ class TestMetadataExtender(TestCase):
             # Verify main output file was created (internal cols removed by default)
             output_files = glob.glob(os.path.join(tmpdir, "*_test_output.txt"))
             self.assertEqual(1, len(output_files))
-            output_df = pandas.read_csv(output_files[0], sep="\t")
+            output_df = pandas.read_csv(output_files[0], sep="\t", dtype=str, keep_default_na=False)
             expected_output_df = pandas.DataFrame({
                 SAMPLE_NAME_KEY: ["sample1", "sample2"],
                 "body_product": ["UBERON:feces", "UBERON:feces"],
                 "body_site": ["gut", "gut"],
                 "description": ["human sample", "human sample"],
+                "dna_extracted": ["TRUE", "FALSE"],
                 "host_common_name": ["human", "human"],
                 QIITA_SAMPLE_TYPE: ["stool", "stool"],
                 SAMPLE_TYPE_KEY: ["stool", "stool"],
@@ -3899,6 +3917,7 @@ class TestMetadataExtender(TestCase):
                 "body_product": ["UBERON:feces", "UBERON:feces"],
                 "body_site": ["gut", "gut"],
                 "description": ["human sample", "human sample"],
+                "dna_extracted": ["TRUE", "FALSE"],
                 "host_common_name": ["human", "human"],
                 QIITA_SAMPLE_TYPE: ["stool", "stool"],
                 SAMPLE_TYPE_KEY: ["stool", "stool"],
@@ -3913,12 +3932,13 @@ class TestMetadataExtender(TestCase):
             # Verify main output file was created
             output_files = glob.glob(os.path.join(tmpdir, "*_test_output.txt"))
             self.assertEqual(1, len(output_files))
-            output_df = pandas.read_csv(output_files[0], sep="\t")
+            output_df = pandas.read_csv(output_files[0], sep="\t", dtype=str, keep_default_na=False)
             expected_output_df = pandas.DataFrame({
                 SAMPLE_NAME_KEY: ["sample1", "sample2"],
                 "body_product": ["UBERON:feces", "UBERON:feces"],
                 "body_site": ["gut", "gut"],
                 "description": ["human sample", "human sample"],
+                "dna_extracted": ["TRUE", "FALSE"],
                 "host_common_name": ["human", "human"],
                 QIITA_SAMPLE_TYPE: ["stool", "stool"],
                 SAMPLE_TYPE_KEY: ["stool", "stool"],
@@ -3927,6 +3947,60 @@ class TestMetadataExtender(TestCase):
             })
             assert_frame_equal(expected_output_df, output_df)
 
+    def test_write_extended_metadata_xlsx_input(self):
+        """Test writing extended metadata from an Excel XLSX input file."""
+        with tempfile.TemporaryDirectory() as tmpdir:
+            result_df = write_extended_metadata(
+                self.TEST_METADATA_XLSX_FP, self.TEST_STUDY_CONFIG_FP,
+                tmpdir, "test_output", stds_fp=self.TEST_STDS_FP)
+
+            # Verify returned DataFrame
+            expected_result_df = pandas.DataFrame({
+                SAMPLE_NAME_KEY: ["sample1", "sample2"],
+                "body_product": ["UBERON:feces", "UBERON:feces"],
+                "body_site": ["gut", "gut"],
+                "description": ["human sample", "human sample"],
+                "dna_extracted": ["TRUE", "FALSE"],
+                "host_common_name": ["human", "human"],
+                QIITA_SAMPLE_TYPE: ["stool", "stool"],
+                SAMPLE_TYPE_KEY: ["stool", "stool"],
+                "study_custom_field": ["custom_value", "custom_value"],
+                "study_stool_field": ["stool_custom", "stool_custom"],
+                HOSTTYPE_SHORTHAND_KEY: ["human", "human"],
+                SAMPLETYPE_SHORTHAND_KEY: ["stool", "stool"],
+                QC_NOTE_KEY: ["", ""]
+            })
+            assert_frame_equal(expected_result_df, result_df)
+
+            # Verify main output file was created
+            output_files = glob.glob(os.path.join(tmpdir, "*_test_output.txt"))
+            self.assertEqual(1, len(output_files))
+            output_df = pandas.read_csv(output_files[0], sep="\t", dtype=str, keep_default_na=False)
+            expected_output_df = pandas.DataFrame({
+                SAMPLE_NAME_KEY: ["sample1", "sample2"],
+                "body_product": ["UBERON:feces", "UBERON:feces"],
+                "body_site": ["gut", "gut"],
+                "description": ["human sample", "human sample"],
+                "dna_extracted": ["TRUE", "FALSE"],
+                "host_common_name": ["human", "human"],
+                QIITA_SAMPLE_TYPE: ["stool", "stool"],
+                SAMPLE_TYPE_KEY: ["stool", "stool"],
+                "study_custom_field": ["custom_value", "custom_value"],
+                "study_stool_field": ["stool_custom", "stool_custom"]
+            })
+            assert_frame_equal(expected_output_df, output_df)
+
+            # Verify empty fails file was created
+            fails_files = glob.glob(os.path.join(tmpdir, "*_test_output_fails.csv"))
+            self.assertEqual(1, len(fails_files))
+            self.assertEqual(0, os.path.getsize(fails_files[0]))
+
+            # Verify empty validation errors file was created
+            validation_files = glob.glob(
+                os.path.join(tmpdir, "*_test_output_validation_errors.csv"))
+            self.assertEqual(1, len(validation_files))
+            self.assertEqual(0, os.path.getsize(validation_files[0]))
+
     def test_write_extended_metadata_with_validation_errors(self):
         """Test writing extended metadata when validation errors occur."""
         with tempfile.TemporaryDirectory() as tmpdir:
@@ -3941,6 +4015,7 @@ class TestMetadataExtender(TestCase):
                 "body_product": ["UBERON:feces", "UBERON:feces"],
                 "body_site": ["gut", "gut"],
                 "description": ["human sample", "human sample"],
+                "dna_extracted": ["TRUE", "FALSE"],
                 "host_common_name": ["human", "human"],
                 QIITA_SAMPLE_TYPE: ["stool", "stool"],
                 "restricted_field": ["invalid_value", "allowed_value"],
@@ -3954,12 +4029,13 @@ class TestMetadataExtender(TestCase):
             # Verify main output file was created
             output_files = glob.glob(os.path.join(tmpdir, "*_test_output.txt"))
             self.assertEqual(1, len(output_files))
-            output_df = pandas.read_csv(output_files[0], sep="\t")
+            output_df = pandas.read_csv(output_files[0], sep="\t", dtype=str, keep_default_na=False)
             expected_output_df = pandas.DataFrame({
                 SAMPLE_NAME_KEY: ["sample1", "sample2"],
                 "body_product": ["UBERON:feces", "UBERON:feces"],
                 "body_site": ["gut", "gut"],
                 "description": ["human sample", "human sample"],
+                "dna_extracted": ["TRUE", "FALSE"],
                 "host_common_name": ["human", "human"],
                 QIITA_SAMPLE_TYPE: ["stool", "stool"],
                 "restricted_field": ["invalid_value", "allowed_value"],
@@ -3971,7 +4047,7 @@ class TestMetadataExtender(TestCase):
             validation_files = glob.glob(
                 os.path.join(tmpdir, "*_test_output_validation_errors.csv"))
             self.assertEqual(1, len(validation_files))
-            validation_df = pandas.read_csv(validation_files[0], sep=",")
+            validation_df = pandas.read_csv(validation_files[0], sep=",", dtype=str, keep_default_na=False)
             expected_validation_df = pandas.DataFrame({
                 "sample_name": ["sample1"],
                 "field_name": ["restricted_field"],
@@ -4006,6 +4082,7 @@ class TestMetadataExtender(TestCase):
                 "body_product": ["UBERON:feces", "UBERON:feces"],
                 "body_site": ["gut", "gut"],
                 "description": ["human sample", "human sample"],
+                "dna_extracted": ["TRUE", "FALSE"],
                 "host_common_name": ["human", "human"],
                 QIITA_SAMPLE_TYPE: ["stool", "stool"],
                 SAMPLE_TYPE_KEY: ["stool", "stool"],
@@ -4020,12 +4097,13 @@ class TestMetadataExtender(TestCase):
             # Verify output file has .csv extension
             output_files = glob.glob(os.path.join(tmpdir, "*_test_output.csv"))
             self.assertEqual(1, len(output_files))
-            output_df = pandas.read_csv(output_files[0], sep=",")
+            output_df = pandas.read_csv(output_files[0], sep=",", dtype=str, keep_default_na=False)
             expected_output_df = pandas.DataFrame({
                 SAMPLE_NAME_KEY: ["sample1", "sample2"],
                 "body_product": ["UBERON:feces", "UBERON:feces"],
                 "body_site": ["gut", "gut"],
                 "description": ["human sample", "human sample"],
+                "dna_extracted": ["TRUE", "FALSE"],
                 "host_common_name": ["human", "human"],
                 QIITA_SAMPLE_TYPE: ["stool", "stool"],
                 SAMPLE_TYPE_KEY: ["stool", "stool"],
@@ -4048,6 +4126,7 @@ class TestMetadataExtender(TestCase):
                 "body_product": ["UBERON:feces", "UBERON:feces"],
                 "body_site": ["gut", "gut"],
                 "description": ["human sample", "human sample"],
+                "dna_extracted": ["TRUE", "FALSE"],
                 "host_common_name": ["human", "human"],
                 QIITA_SAMPLE_TYPE: ["stool", "stool"],
                 SAMPLE_TYPE_KEY: ["stool", "stool"],
@@ -4062,12 +4141,13 @@ class TestMetadataExtender(TestCase):
             # Verify main output file includes internal columns
             output_files = glob.glob(os.path.join(tmpdir, "*_test_output.txt"))
             self.assertEqual(1, len(output_files))
-            output_df = pandas.read_csv(output_files[0], sep="\t", keep_default_na=False)
+            output_df = pandas.read_csv(output_files[0], sep="\t", dtype=str, keep_default_na=False)
             expected_output_df = pandas.DataFrame({
                 SAMPLE_NAME_KEY: ["sample1", "sample2"],
                 "body_product": ["UBERON:feces", "UBERON:feces"],
                 "body_site": ["gut", "gut"],
                 "description": ["human sample", "human sample"],
+                "dna_extracted": ["TRUE", "FALSE"],
                 "host_common_name": ["human", "human"],
                 QIITA_SAMPLE_TYPE: ["stool", "stool"],
                 SAMPLE_TYPE_KEY: ["stool", "stool"],
@@ -4097,6 +4177,7 @@ class TestMetadataExtender(TestCase):
                 "body_product": ["UBERON:feces", "UBERON:feces"],
                 "body_site": ["gut", "gut"],
                 "description": ["human sample", "human sample"],
+                "dna_extracted": ["TRUE", "FALSE"],
                 "host_common_name": ["human", "human"],
                 QIITA_SAMPLE_TYPE: ["stool", "stool"],
                 SAMPLE_TYPE_KEY: ["stool", "stool"],
@@ -4111,12 +4192,13 @@ class TestMetadataExtender(TestCase):
             # Verify main output file was created
             output_files = glob.glob(os.path.join(tmpdir, "*_test_output.txt"))
             self.assertEqual(1, len(output_files))
-            output_df = pandas.read_csv(output_files[0], sep="\t")
+            output_df = pandas.read_csv(output_files[0], sep="\t", dtype=str, keep_default_na=False)
             expected_output_df = pandas.DataFrame({
                 SAMPLE_NAME_KEY: ["sample1", "sample2"],
                 "body_product": ["UBERON:feces", "UBERON:feces"],
                 "body_site": ["gut", "gut"],
                 "description": ["human sample", "human sample"],
+                "dna_extracted": ["TRUE", "FALSE"],
                 "host_common_name": ["human", "human"],
                 QIITA_SAMPLE_TYPE: ["stool", "stool"],
                 SAMPLE_TYPE_KEY: ["stool", "stool"],
@@ -4134,6 +4216,64 @@ class TestMetadataExtender(TestCase):
                 os.path.join(tmpdir, "*_test_output_validation_errors.csv"))
             self.assertEqual(0, len(validation_files))
 
+    def test_write_extended_metadata_preserves_string_booleans(self):
+        """Test that TRUE/FALSE string values are not converted to booleans.
+
+        This tests for a bug where loading a CSV without dtype=str causes
+        pandas to convert 'TRUE'/'FALSE' strings to boolean True/False,
+        which then fail validation against allowed string values.
+        """
+        with tempfile.TemporaryDirectory() as tmpdir:
+            # Create a CSV file with TRUE/FALSE string values
+            csv_content = (
+                "sample_name,hosttype_shorthand,sampletype_shorthand,dna_extracted\n"
+                "sample1,human,stool,TRUE\n"
+                "sample2,human,stool,FALSE\n"
+            )
+            csv_fp = path.join(tmpdir, "test_bool_strings.csv")
+            with open(csv_fp, "w") as f:
+                f.write(csv_content)
+
+            # Create a config that defines TRUE/FALSE as allowed string values
+            config_content = """
+default: "not provided"
+leave_requireds_blank: false
+overwrite_non_nans: false
+study_specific_metadata:
+  host_type_specific_metadata:
+    human:
+      default: "not provided"
+      leave_requireds_blank: false
+      overwrite_non_nans: false
+      sample_type_specific_metadata:
+        stool:
+          metadata_fields:
+            dna_extracted:
+              type: string
+              allowed:
+                - "TRUE"
+                - "FALSE"
+"""
+            config_fp = path.join(tmpdir, "test_bool_config.yml")
+            with open(config_fp, "w") as f:
+                f.write(config_content)
+
+            # Call write_extended_metadata
+            result_df = write_extended_metadata(
+                csv_fp, config_fp, tmpdir, "test_output",
+                stds_fp=self.TEST_STDS_FP)
+
+            # Verify the dna_extracted values are preserved as strings
+            self.assertEqual("TRUE", result_df.loc[0, "dna_extracted"])
+            self.assertEqual("FALSE", result_df.loc[1, "dna_extracted"])
+
+            # Verify no validation errors occurred
+            validation_files = glob.glob(
+                os.path.join(tmpdir, "*_test_output_validation_errors.csv"))
+            self.assertEqual(1, len(validation_files))
+            # The validation errors file should be empty (0 bytes)
+            self.assertEqual(0, os.path.getsize(validation_files[0]))
+
     # Integration tests
 
     TEST_PROJECT1_METADATA_FP = path.join(TEST_DIR, "data/test_project1_input_metadata.csv")
@@ -4142,6 +4282,7 @@ class TestMetadataExtender(TestCase):
         TEST_DIR, "data/test_project1_output_metadata.txt")
     TEST_PROJECT1_EXPECTED_FAILS_FP = path.join(
         TEST_DIR, "data/test_project1_output_fails.csv")
+
     def test_write_extended_metadata_from_df_project1_integration(self):
         """Integration test using project1 test data files."""
 
@@ -4153,12 +4294,8 @@ class TestMetadataExtender(TestCase):
             with open(path.join(debug_dir, f"UNMATCHED_2_{file_name}"), 'w') as debug_actual_file:
                 debug_actual_file.write(actual_content)
 
-
         # Load input metadata CSV
         input_df = pandas.read_csv(self.TEST_PROJECT1_METADATA_FP, dtype=str)
-        # for the columns "plating_notes" and "notes", fill NaN with empty string
-        input_df["plating_notes"] = input_df["plating_notes"].fillna("")
-        input_df["notes"] = input_df["notes"].fillna("")
 
         # Load study config
         study_config = _get_study_specific_config(self.TEST_PROJECT1_CONFIG_FP)
@@ -4204,6 +4341,58 @@ class TestMetadataExtender(TestCase):
             self.assertEqual(1, len(validation_files))
             self.assertEqual(0, os.path.getsize(validation_files[0]))
 
+    def test_write_extended_metadata_project1_integration(self):
+        """Integration test for write_extended_metadata using project1 test data files."""
+
+        def write_mismatched_debug_files(expected_content, actual_content, file_name):
+            """Write debug files to Desktop for unmatched content."""
+            debug_dir = path.join(path.expanduser("~"), "Desktop")
+            with open(path.join(debug_dir, f"UNMATCHED_1_{file_name}"), 'w') as debug_expected_file:
+                debug_expected_file.write(expected_content)
+            with open(path.join(debug_dir, f"UNMATCHED_2_{file_name}"), 'w') as debug_actual_file:
+                debug_actual_file.write(actual_content)
+
+        with tempfile.TemporaryDirectory() as tmpdir:
+            write_extended_metadata(
+                self.TEST_PROJECT1_METADATA_FP, self.TEST_PROJECT1_CONFIG_FP,
+                tmpdir, "test_output", remove_internals=True)
+
+            # Compare main output file directly to expected file
+            output_files = glob.glob(os.path.join(tmpdir, "*_test_output.txt"))
+            self.assertEqual(1, len(output_files))
+            with open(output_files[0], 'r') as actual_file:
+                actual_content = actual_file.read()
+            with open(self.TEST_PROJECT1_EXPECTED_OUTPUT_FP, 'r') as expected_file:
+                expected_content = expected_file.read()
+            try:
+                self.assertEqual(expected_content, actual_content)
+            except AssertionError:
+                write_mismatched_debug_files(
+                    expected_content, actual_content,
+                    "project1_output.txt")
+                raise
+
+            # Compare fails file directly to expected file
+            fails_files = glob.glob(os.path.join(tmpdir, "*_test_output_fails.csv"))
+            self.assertEqual(1, len(fails_files))
+            with open(fails_files[0], 'r') as actual_file:
+                actual_fails_content = actual_file.read()
+            with open(self.TEST_PROJECT1_EXPECTED_FAILS_FP, 'r') as expected_file:
+                expected_fails_content = expected_file.read()
+            try:
+                self.assertEqual(expected_fails_content, actual_fails_content)
+            except AssertionError:
+                write_mismatched_debug_files(
+                    expected_fails_content, actual_fails_content,
+                    "project1_fails.csv")
+                raise
+
+            # Verify validation errors file is empty
+            validation_files = glob.glob(
+                os.path.join(tmpdir, "*_test_output_validation_errors.csv"))
+            self.assertEqual(1, len(validation_files))
+            self.assertEqual(0, os.path.getsize(validation_files[0]))
+
     # Tests for _get_specified_column_name
 
     def test__get_specified_column_name_finds_column(self):