diff --git a/client/galaxy/scripts/components/RuleCollectionBuilder.vue b/client/galaxy/scripts/components/RuleCollectionBuilder.vue
index 1a5828750e97..ae1d471ae97e 100644
--- a/client/galaxy/scripts/components/RuleCollectionBuilder.vue
+++ b/client/galaxy/scripts/components/RuleCollectionBuilder.vue
@@ -59,6 +59,18 @@
                                 </select>
                             </label>
                         </rule-component>
+                        <rule-component rule-type="add_column_group_tag_value"
+                                        :display-rule-type="displayRuleType"
+                                        :builder="this">
+                            <label>
+                                {{ l("Value") }}
+                                <input type="text" v-model="addColumnGroupTagValueValue" />
+                            </label>
+                            <label>
+                                {{ l("Default") }}
+                                <input type="text" v-model="addColumnGroupTagValueDefault" />
+                            </label>
+                        </rule-component>
                         <rule-component rule-type="add_column_regex"
                                         :display-rule-type="displayRuleType"
                                         :builder="this">
@@ -286,6 +298,7 @@
                                     <div class="dropdown-menu" role="menu">
                                         <rule-target-component :builder="this" rule-type="add_column_basename" />
                                         <rule-target-component :builder="this" rule-type="add_column_metadata" v-if="metadataOptions"/>
+                                        <rule-target-component :builder="this" rule-type="add_column_group_tag_value" v-if="hasTagsMetadata"/>
                                         <rule-target-component :builder="this" rule-type="add_column_regex" />
                                         <rule-target-component :builder="this" rule-type="add_column_concatenate" />
                                         <rule-target-component :builder="this" rule-type="add_column_rownum" />
@@ -901,6 +914,8 @@ export default {
             addColumnRegexGroupCount: null,
             addColumnRegexType: "global",
             addColumnMetadataValue: 0,
+            addColumnGroupTagValueValue: "",
+            addColumnGroupTagValueDefault: "",
             addColumnConcatenateTarget0: 0,
             addColumnConcatenateTarget1: 0,
             addColumnRownumStart: 1,
@@ -1182,6 +1197,7 @@ export default {
                         metadataOptions["identifier" + index] = _l("Paired Identifier");
                     }
                 }
+                metadataOptions["tags"] = _l("Tags");
             } else if (this.elementsType == "ftp") {
                 metadataOptions["path"] = _l("Path");
             } else if (this.elementsType == "library_datasets") {
@@ -1194,6 +1210,10 @@ export default {
             }
             return metadataOptions;
         },
+        hasTagsMetadata() {
+            // TODO: allow for dataset, library_datasets also - here and just above in metadataOptions.
+            return this.elementsType == "collection_contents";
+        },
         collectionType() {
             let identifierColumns = [];
             if (this.mappingAsDict.list_identifiers) {
@@ -1680,9 +1700,9 @@ export default {
                 if (collectionTypeLevelSepIndex === -1) {
                     // Flat collection at this depth.
                     // sources are the elements
-                    // TOOD: right thing is probably this: data.push([]);
                     data.push([]);
-                    sources.push({ identifiers: identifiers, dataset: elementObject });
+                    const source = { identifiers: identifiers, dataset: elementObject, tags: elementObject.tags };
+                    sources.push(source);
                 } else {
                     const restCollectionType = collectionType.slice(collectionTypeLevelSepIndex + 1);
                     let elementObj = this.populateElementsFromCollectionDescription(
diff --git a/client/galaxy/scripts/mvc/rules/rule-definitions.js b/client/galaxy/scripts/mvc/rules/rule-definitions.js
index d2af5a83a965..2db7bd6f5405 100644
--- a/client/galaxy/scripts/mvc/rules/rule-definitions.js
+++ b/client/galaxy/scripts/mvc/rules/rule-definitions.js
@@ -167,13 +167,21 @@ const RULES = {
         apply: (rule, data, sources, columns) => {
             const ruleValue = rule.value;
             let newRow;
-            if (ruleValue.startsWith("identifier")) {
+            if (ruleValue.indexOf("identifier") == 0) {
                 const identifierIndex = parseInt(ruleValue.substring("identifier".length));
                 newRow = (row, index) => {
                     const newRow = row.slice();
                     newRow.push(sources[index]["identifiers"][identifierIndex]);
                     return newRow;
                 };
+            } else if (ruleValue == "tags") {
+                newRow = (row, index) => {
+                    const newRow = row.slice();
+                    const tags = sources[index]["tags"];
+                    tags.sort();
+                    newRow.push(tags.join(","));
+                    return newRow;
+                };
             } else if (ruleValue == "hid" || ruleValue == "name" || ruleValue == "path") {
                 newRow = (row, index) => {
                     const newRow = row.slice();
@@ -188,6 +196,47 @@ const RULES = {
             return { data, columns };
         }
     },
+    add_column_group_tag_value: {
+        title: _l("Add Column from Group Tag Value"),
+        display: (rule, colHeaders) => {
+            return `Add column for value of group tag ${rule.value}.`;
+        },
+        init: (component, rule) => {
+            if (!rule) {
+                component.addColumnGroupTagValueValue = null;
+                component.addColumnGroupTagValueDefault = '';
+            } else {
+                component.addColumnGroupTagValueValue = rule.value;
+                component.addColumnGroupTagValueDefault = rule.default_value;
+            }
+        },
+        save: (component, rule) => {
+            rule.value = component.addColumnGroupTagValueValue;
+            rule.default_value = component.addColumnGroupTagValueDefault;
+        },
+        apply: (rule, data, sources, columns) => {
+            const ruleValue = rule.value;
+            const groupTagPrefix = `group:${ruleValue}:`;
+            const newRow = (row, index) => {
+                const newRow = row.slice();
+                const tags = sources[index]["tags"];
+                tags.sort();
+                let groupTagValue = rule.default_value;
+                for (let index in tags) {
+                    const tag = tags[index];
+                    if ( tag.indexOf(groupTagPrefix) == 0 ) {
+                        groupTagValue = tag.substr(groupTagPrefix.length);
+                        break;
+                    }
+                }
+                newRow.push(groupTagValue);
+                return newRow;
+            };
+            data = data.map(newRow);
+            columns.push(NEW_COLUMN);
+            return { data, columns };
+        }
+    },
     add_column_regex: {
         title: _l("Using a Regular Expression"),
         display: (rule, colHeaders) => {
diff --git a/lib/galaxy/managers/collections.py b/lib/galaxy/managers/collections.py
index 42a5e03be99c..5aa631e8fa67 100644
--- a/lib/galaxy/managers/collections.py
+++ b/lib/galaxy/managers/collections.py
@@ -492,7 +492,8 @@ def __init_rule_data(self, elements, collection_type_description, parent_identif
             identifiers = parent_identifiers + [element.element_identifier]
             if not element.is_collection:
                 data.append([])
-                sources.append({"identifiers": identifiers, "dataset": element_object})
+                source = {"identifiers": identifiers, "dataset": element_object, "tags": element_object.make_tag_string_list()}
+                sources.append(source)
             else:
                 child_collection_type_description = collection_type_description.child_collection_type_description()
                 element_data, element_sources = self.__init_rule_data(
diff --git a/lib/galaxy/tools/__init__.py b/lib/galaxy/tools/__init__.py
index f10cb7f98466..317ccab75e1e 100755
--- a/lib/galaxy/tools/__init__.py
+++ b/lib/galaxy/tools/__init__.py
@@ -2707,7 +2707,6 @@ class ApplyRulesTool(DatabaseOperationTool):
     tool_type = 'apply_rules'
 
     def produce_outputs(self, trans, out_data, output_collections, incoming, history, **kwds):
-        log.info(incoming)
         hdca = incoming["input"]
         rule_set = RuleSet(incoming["rules"])
         copied_datasets = []
diff --git a/lib/galaxy/util/rules_dsl.py b/lib/galaxy/util/rules_dsl.py
index 2172532e01ef..0aa51c3f140c 100644
--- a/lib/galaxy/util/rules_dsl.py
+++ b/lib/galaxy/util/rules_dsl.py
@@ -75,11 +75,50 @@ def validate_rule(self, rule):
 
     def apply(self, rule, data, sources):
         rule_value = rule["value"]
-        identifier_index = int(rule_value[len("identifier"):])
+        if rule_value.startswith("identifier"):
+            identifier_index = int(rule_value[len("identifier"):])
+
+            new_rows = []
+            for index, row in enumerate(data):
+                new_rows.append(row + [sources[index]["identifiers"][identifier_index]])
+
+        elif rule_value == "tags":
+
+            def sorted_tags(index):
+                tags = sorted(sources[index]["tags"])
+                return [",".join(tags)]
+
+            new_rows = []
+            for index, row in enumerate(data):
+                new_rows.append(row + sorted_tags(index))
+
+        return new_rows, sources
+
+
+class AddColumnGroupTagValueRuleDefinition(BaseRuleDefinition):
+    rule_type = "add_column_group_tag_value"
+
+    def validate_rule(self, rule):
+        _ensure_rule_contains_keys(rule, {"value": six.string_types})
+
+    def apply(self, rule, data, sources):
+        rule_value = rule["value"]
+        tag_prefix = "group:%s:" % rule_value
 
         new_rows = []
         for index, row in enumerate(data):
-            new_rows.append(row + [sources[index]["identifiers"][identifier_index]])
+            group_tag_value = None
+            source = sources[index]
+            tags = source["tags"]
+            for tag in sorted(tags):
+                if tag.startswith(tag_prefix):
+                    group_tag_value = tag[len(tag_prefix):]
+                    break
+
+            if group_tag_value is None:
+                group_tag_value = rule.get("default_value", "")
+
+            new_rows.append(row + [group_tag_value])
 
         return new_rows, sources
 
@@ -296,7 +335,8 @@ def apply(self, rule, data, sources):
         target_column = rule["target_column"]
 
         def _filter(index):
-            return not invert if len(data[target_column]) == 0 else invert
+            non_empty = len(data[index][target_column]) != 0
+            return not invert if non_empty else invert
 
         return _filter_index(_filter, data), _filter_index(_filter, sources)
 
@@ -524,6 +564,7 @@ def display(self):
 
 RULES_DEFINITION_CLASSES = [
     AddColumnMetadataRuleDefinition,
+    AddColumnGroupTagValueRuleDefinition,
     AddColumnConcatenateRuleDefinition,
     AddColumnBasenameRuleDefinition,
     AddColumnRegexRuleDefinition,
diff --git a/test/api/test_tools.py b/test/api/test_tools.py
index 51599151b7e6..20ce7e2404dd 100644
--- a/test/api/test_tools.py
+++ b/test/api/test_tools.py
@@ -353,6 +353,9 @@ def test_apply_rules_2(self):
     def test_apply_rules_3(self):
         self._apply_rules_and_check(rules_test_data.EXAMPLE_3)
 
+    def test_apply_rules_4(self):
+        self._apply_rules_and_check(rules_test_data.EXAMPLE_4)
+
     @skip_without_tool("multi_select")
     def test_multi_select_as_list(self):
         with self.dataset_populator.test_history() as history_id:
diff --git a/test/base/data/rules_dsl_spec.yml b/test/base/data/rules_dsl_spec.yml
index 8f0943b21674..276290cf5088 100644
--- a/test/base/data/rules_dsl_spec.yml
+++ b/test/base/data/rules_dsl_spec.yml
@@ -325,6 +325,7 @@
   final:
     data: [["20", "dog"], ["30", "cat"]]
     sources: [4, 5]
+
 - rules:
     - type: add_filter_compare
       target_column: 0
@@ -336,6 +337,7 @@
   final:
     data: [["13", "rat"], ["20", "dog"], ["30", "cat"]]
     sources: [3, 4, 5]
+
 - rules:
     - type: sort
       numeric: false
@@ -346,6 +348,7 @@
   final:
     data: [["bark", "dog"], ["meow", "cat"], ["moo", "cow"]]
     sources: [3, 2, 1]
+
 - rules:
     - type: sort
       numeric: false
@@ -356,6 +359,7 @@
   final:
     data: [["meow", "cat"], ["moo", "cow"], ["bark", "dog"]]
     sources: [2, 1, 3]
+
 - rules:
     - type: sort
       numeric: false
@@ -366,6 +370,7 @@
   final:
     data: [["bark", "Dog"], ["meow", "cat"], ["moo", "cow"]]
     sources: [3, 2, 1]
+
 - rules:
     - type: swap_columns
       target_column_0: 0
@@ -376,6 +381,7 @@
   final:
     data: [["cow", "moo"], ["cat", "meow"], ["Dog", "bark"]]
     sources: [1, 2, 3]
+
 - rules:
     - type: split_columns
       target_columns_0: [0]
@@ -386,28 +392,84 @@
   final:
     data: [["moo", "A"], ["cow", "A"], ["meow", "B"], ["cat", "B"], ["bark", "C"], ["Dog", "C"]]
     sources: [1, 1, 2, 2, 3, 3]
+
+- rules:
+    - type: add_column_metadata
+      value: identifier0
+  initial:
+    data: [["moo"], ["meow"], ["bark"]]
+    sources: [{"identifiers": ["cow"]}, {"identifiers": ["cat"]}, {"identifiers": ["dog"]}]
+  final:
+    data: [["moo", "cow"], ["meow", "cat"], ["bark", "dog"]]
+
+- rules:
+    - type: add_column_metadata
+      value: tags
+  initial:
+    data: [["moo"], ["meow"], ["bark"]]
+    sources: [{"identifiers": ["cow"], "tags": ["farm"]}, {"identifiers": ["cat"], "tags": ["house"]}, {"identifiers": ["dog"], "tags": ["house", "firestation"]}]
+  final:
+    data: [["moo", "farm"], ["meow", "house"], ["bark", "firestation,house"]]
+
+- rules:
+    - type: add_column_group_tag_value
+      value: where
+      default_value: ''
+  initial:
+    data: [["moo"], ["meow"], ["bark"]]
+    sources: [{"identifiers": ["cow"], "tags": ["group:where:farm"]}, {"identifiers": ["cat"], "tags": ["group:where:house"]}, {"identifiers": ["dog"], "tags": ["group:where:house"]}]
+  final:
+    data: [["moo", "farm"], ["meow", "house"], ["bark", "house"]]
+
+- doc: Test add_column_group_tag_value default value.
+  rules:
+    - type: add_column_group_tag_value
+      value: where
+      default_value: 'barn'
+  initial:
+    data: [["moo"], ["meow"], ["bark"]]
+    sources: [{"identifiers": ["cow"], "tags": []}, {"identifiers": ["cat"], "tags": ["group:where:house"]}, {"identifiers": ["dog"], "tags": ["group:where:firestation"]}]
+  final:
+    data: [["moo", "barn"], ["meow", "house"], ["bark", "firestation"]]
+
+- doc: Test add_column_group_tag_value sorts and grabs first tag value if multiple present.
+  rules:
+    - type: add_column_group_tag_value
+      value: where
+      default_value: 'barn'
+  initial:
+    data: [["moo"], ["meow"], ["bark"]]
+    sources: [{"identifiers": ["cow"], "tags": []}, {"identifiers": ["cat"], "tags": ["group:where:house", "group:where:kittenpile"]}, {"identifiers": ["dog"], "tags": ["group:where:house", "group:where:firestation"]}]
+  final:
+    data: [["moo", "barn"], ["meow", "house"], ["bark", "firestation"]]
+
 - rules:
     - type: invalid_rule_type
   error: true
+
 - rules:
     - type: add_filter_compare
       target_column: 0
       value: 13
       compare_type: invalid_compare_type
   error: true
+
 - rules:
     - type: add_column_concatenate
       target_column: 0
   error: true
+
 - rules:
     - type: add_column_basename
       target_column_0: 0
   error: true
+
 - rules:
     - type: add_column_regex
       target_column: 0
       regex: '(o)+'
   error: true
+
 - rules:
     - type: add_column_regex
       target_column: 0
diff --git a/test/base/populators.py b/test/base/populators.py
index ace65d6cac94..8651c96c2009 100644
--- a/test/base/populators.py
+++ b/test/base/populators.py
@@ -917,13 +917,20 @@ def read_test_data(test_dict):
             elements_data = value.get("elements", [])
             elements = []
             for element_data in elements_data:
-                identifier = element_data["identifier"]
+                # Adapt differences between test_data dict and fetch API description.
+                if "name" not in element_data:
+                    identifier = element_data["identifier"]
+                    element_data["name"] = identifier
                 input_type = element_data.get("type", "raw")
+                content = None
                 if input_type == "File":
                     content = read_test_data(element_data)
                 else:
                     content = element_data["content"]
-                elements.append((identifier, content))
+                if content is not None:
+                    element_data["src"] = "pasted"
+                    element_data["paste_content"] = content
+                elements.append(element_data)
             # TODO: make this collection_type
             collection_type = value["type"]
             new_collection_kwds = {}
diff --git a/test/base/rules_test_data.py b/test/base/rules_test_data.py
index 3fb3eb2bb39e..b7da95b23e3d 100644
--- a/test/base/rules_test_data.py
+++ b/test/base/rules_test_data.py
@@ -30,6 +30,21 @@ def check_example_3(hdca, dataset_populator):
     assert first_element["element_identifier"] == "test0forward"
 
 
+def check_example_4(hdca, dataset_populator):
+    assert hdca["collection_type"] == "list:list"
+    assert hdca["element_count"] == 2
+    first_collection_level = hdca["elements"][0]
+    assert first_collection_level["element_identifier"] == "single", hdca
+    assert first_collection_level["element_type"] == "dataset_collection"
+    second_collection_level = first_collection_level["object"]
+    assert "elements" in second_collection_level, hdca
+    assert len(second_collection_level["elements"]) == 1, hdca
+    i1_element = second_collection_level["elements"][0]
+    assert "object" in i1_element, hdca
+    assert "element_identifier" in i1_element
+    assert i1_element["element_identifier"] == "i1", hdca
+
+
 EXAMPLE_1 = {
     "rules": {
         "rules": [
@@ -130,3 +145,48 @@ def check_example_3(hdca, dataset_populator):
     "check": check_example_3,
     "output_hid": 7,
 }
+
+# Nesting with group tags.
+EXAMPLE_4 = {
+    "rules": {
+        "rules": [
+            {
+                "type": "add_column_metadata",
+                "value": "identifier0",
+            },
+            {
+                "type": "add_column_group_tag_value",
+                "value": "type",
+                "default_value": "unused"
+            }
+        ],
+        "mapping": [
+            {
+                "type": "list_identifiers",
+                "columns": [1, 0],
+            }
+        ],
+    },
+    "test_data": {
+        "type": "list",
+        "elements": [
+            {
+                "identifier": "i1",
+                "content": "0",
+                "tags": ["random", "group:type:single"]
+            },
+            {
+                "identifier": "i2",
+                "content": "1",
+                "tags": ["random", "group:type:paired"]
+            },
+            {
+                "identifier": "i3",
+                "content": "2",
+                "tags": ["random", "group:type:paired"]
+            },
+        ]
+    },
+    "check": check_example_4,
+    "output_hid": 8,
+}
diff --git a/test/galaxy_selenium/navigates_galaxy.py b/test/galaxy_selenium/navigates_galaxy.py
index 8a8d4ba345ac..b613455456ef 100644
--- a/test/galaxy_selenium/navigates_galaxy.py
+++ b/test/galaxy_selenium/navigates_galaxy.py
@@ -1023,6 +1023,8 @@ def tool_open(self, tool_id, outer=False):
             tool_link = self.components.tool_panel.outer_tool_link(tool_id=tool_id)
         else:
             tool_link = self.components.tool_panel.tool_link(tool_id=tool_id)
+        tool_element = tool_link.wait_for_present()
+        self.driver.execute_script("arguments[0].scrollIntoView(true);", tool_element)
         tool_link.wait_for_and_click()
 
     def tool_parameter_div(self, expanded_parameter_id):
diff --git a/test/galaxy_selenium/smart_components.py b/test/galaxy_selenium/smart_components.py
index 01286cf978b6..9423355721b2 100644
--- a/test/galaxy_selenium/smart_components.py
+++ b/test/galaxy_selenium/smart_components.py
@@ -78,6 +78,9 @@ def wait_for_absent_or_hidden(self, **kwds):
     def wait_for_absent(self, **kwds):
         self._has_driver.wait_for_absent(self._target, **kwds)
 
+    def wait_for_present(self, **kwds):
+        return self._has_driver.wait_for_present(self._target, **kwds)
+
     def assert_absent(self, **kwds):
         self._has_driver.assert_absent(self._target, **kwds)
 
diff --git a/test/selenium_tests/test_tool_form.py b/test/selenium_tests/test_tool_form.py
index 6300574a0044..5c731f19d259 100644
--- a/test/selenium_tests/test_tool_form.py
+++ b/test/selenium_tests/test_tool_form.py
@@ -178,6 +178,11 @@ def test_run_apply_rules_3(self):
         self._apply_rules_and_check(rules_test_data.EXAMPLE_3)
         self.screenshot("tool_apply_rules_example_3_final")
 
+    @selenium_test
+    def test_run_apply_rules_4(self):
+        self._apply_rules_and_check(rules_test_data.EXAMPLE_4)
+        self.screenshot("tool_apply_rules_example_4_final")
+
     @selenium_test
     @managed_history
     def test_run_apply_rules_tutorial(self):
diff --git a/test/unit/test_rule_utils.py b/test/unit/test_rule_utils.py
index 249ec2341f52..51f990d387bc 100644
--- a/test/unit/test_rule_utils.py
+++ b/test/unit/test_rule_utils.py
@@ -11,11 +11,22 @@ def test_rules():
             initial = test_case["initial"]
             final_data, final_sources = rule_set.apply(initial["data"], initial["sources"])
             expected_final = test_case["final"]
-            for final_row, expected_final_row in zip(final_data, expected_final["data"]):
+            expected_final_data = expected_final["data"]
+            msg = "Incorrect number of rows, %s != %s" % (final_data, expected_final_data)
+            assert len(expected_final_data) == len(final_data), msg
+            for final_row, expected_final_row in zip(final_data, expected_final_data):
                 msg = "%s != %s" % (final_row, expected_final_row)
                 assert len(final_row) == len(expected_final_row), msg
                 for final_val, expected_final_val in zip(final_row, expected_final_row):
                     assert final_val == expected_final_val, msg
+            expected_final_sources = expected_final.get("sources", None)
+            if expected_final_sources:
+                msg = "Incorrect number of sources, %s != %s" % (expected_final_sources, final_sources)
+                assert len(expected_final_sources) == len(final_sources), msg
+                for final_source, expected_final_source in zip(final_sources, expected_final_sources):
+                    msg = "%s != %s" % (final_source, expected_final_source)
+                    assert final_source == expected_final_source, msg
+
         elif "error" in test_case:
             assert rule_set.has_errors, "rule [%s] does not contain errors" % test_case
         else: