diff --git a/openpype/hosts/nuke/plugins/publish/extract_render_local.py b/openpype/hosts/nuke/plugins/publish/extract_render_local.py
index 50a5d014831..057bca11ac1 100644
--- a/openpype/hosts/nuke/plugins/publish/extract_render_local.py
+++ b/openpype/hosts/nuke/plugins/publish/extract_render_local.py
@@ -7,7 +7,7 @@
 
 class NukeRenderLocal(openpype.api.Extractor):
     # TODO: rewrite docstring to nuke
-    """Render the current Fusion composition locally.
+    """Render the current Nuke composition locally.
 
     Extract the result of savers by starting a comp render
     This will run the local render of Fusion.
diff --git a/openpype/hosts/nuke/plugins/publish/precollect_instances.py b/openpype/hosts/nuke/plugins/publish/precollect_instances.py
index 1a8fa3e6ad9..8bf7280ceaa 100644
--- a/openpype/hosts/nuke/plugins/publish/precollect_instances.py
+++ b/openpype/hosts/nuke/plugins/publish/precollect_instances.py
@@ -151,15 +151,11 @@ def process(self, context):
                 "resolutionWidth": resolution_width,
                 "resolutionHeight": resolution_height,
                 "pixelAspect": pixel_aspect,
-                "review": review
+                "review": review,
+                "representations": []
 
             })
             self.log.info("collected instance: {}".format(instance.data))
             instances.append(instance)
 
-        # create instances in context data if not are created yet
-        if not context.data.get("instances"):
-            context.data["instances"] = list()
-
-        context.data["instances"].extend(instances)
         self.log.debug("context: {}".format(context))
diff --git a/openpype/hosts/nuke/plugins/publish/precollect_workfile.py b/openpype/hosts/nuke/plugins/publish/precollect_workfile.py
index a2d1c80628f..7349a8f4245 100644
--- a/openpype/hosts/nuke/plugins/publish/precollect_workfile.py
+++ b/openpype/hosts/nuke/plugins/publish/precollect_workfile.py
@@ -17,7 +17,7 @@ class CollectWorkfile(pyblish.api.ContextPlugin):
     label = "Pre-collect Workfile"
     hosts = ['nuke']
 
-    def process(self, context):
+    def process(self, context):  # sourcery skip: avoid-builtin-shadow
         root = nuke.root()
 
         current_file = os.path.normpath(nuke.root().name())
@@ -74,6 +74,14 @@ def process(self, context):
         }
         context.data.update(script_data)
 
+        # creating representation
+        representation = {
+            'name': 'nk',
+            'ext': 'nk',
+            'files': base_name,
+            "stagingDir": staging_dir,
+        }
+
         # creating instance data
         instance.data.update({
             "subset": subset,
@@ -82,26 +90,10 @@ def process(self, context):
             "publish": root.knob('publish').value(),
             "family": family,
             "families": [family],
-            "representations": list()
+            "representations": [representation]
         })
 
         # adding basic script data
         instance.data.update(script_data)
 
-        # creating representation
-        representation = {
-            'name': 'nk',
-            'ext': 'nk',
-            'files': base_name,
-            "stagingDir": staging_dir,
-        }
-
-        instance.data["representations"].append(representation)
-
         self.log.info('Publishing script version')
-
-        # create instances in context data if not are created yet
-        if not context.data.get("instances"):
-            context.data["instances"] = list()
-
-        context.data["instances"].append(instance)
diff --git a/openpype/hosts/nuke/plugins/publish/precollect_writes.py b/openpype/hosts/nuke/plugins/publish/precollect_writes.py
index 8669f4f485f..e050fc8c52c 100644
--- a/openpype/hosts/nuke/plugins/publish/precollect_writes.py
+++ b/openpype/hosts/nuke/plugins/publish/precollect_writes.py
@@ -175,6 +175,11 @@ def process(self, instance):
                 "frameEndHandle": last_frame,
             })
 
+            # make sure rendered sequence on farm will
+            # be used for exctract review
+            if not instance.data["review"]:
+                instance.data["useSequenceForReview"] = False
+
         # * Add audio to instance if exists.
         # Find latest versions document
         version_doc = pype.get_latest_version(
diff --git a/openpype/modules/deadline/plugins/publish/submit_publish_job.py b/openpype/modules/deadline/plugins/publish/submit_publish_job.py
index 39a2aa27615..0583c25b57b 100644
--- a/openpype/modules/deadline/plugins/publish/submit_publish_job.py
+++ b/openpype/modules/deadline/plugins/publish/submit_publish_job.py
@@ -641,6 +641,10 @@ def _get_representations(self, instance, exp_files):
 
     def _solve_families(self, instance, preview=False):
         families = instance.get("families")
+
+        # test also instance data review attribute
+        preview = preview or instance.get("review")
+
         # if we have one representation with preview tag
         # flag whole instance for review and for ftrack
         if preview:
@@ -750,6 +754,10 @@ def process(self, instance):
                 "family": "prerender",
                 "families": []})
 
+        # also include review attribute if available
+        if "review" in data:
+            instance_skeleton_data["review"] = data["review"]
+
         # skip locking version if we are creating v01
         instance_version = instance.data.get("version")  # take this if exists
         if instance_version != 1:
diff --git a/openpype/settings/defaults/project_settings/deadline.json b/openpype/settings/defaults/project_settings/deadline.json
index f0b2a7e5557..6b0d7586c40 100644
--- a/openpype/settings/defaults/project_settings/deadline.json
+++ b/openpype/settings/defaults/project_settings/deadline.json
@@ -83,9 +83,6 @@
                 "maya": [
                     ".*([Bb]eauty).*"
                 ],
-                "nuke": [
-                    ".*"
-                ],
                 "aftereffects": [
                     ".*"
                 ],