Connectivity tags

murthylab · Jul 18, 2023 · 27cf901 · 27cf901
1 parent 48661cb
commit 27cf901
Show file tree

Hide file tree

Showing 6 changed files with 48 additions and 26 deletions.
diff --git a/src/configuration.py b/src/configuration.py
@@ -48,7 +48,10 @@
     "coordinates": "Marked neuron coordinates. "
     "FlyWire Supervoxel IDs and position coordinates for cells in the dataset. One cell might have zero or more "
     "marked coordinates and supervoxel IDs.",
-    "neuropil_synapse_table": "In-out synapse & partner counts by neuropil. "
+    "connectivity_tags": "Descriptors for neurons connectivity characteristics, derived from network analysis. Each "
+    "neuron can have 0 or more tags, separated by commas in the second column. "
+    'See <a href="" data-toggle="modal" data-target="#connectivityTagsModal">details</a>.',
+    "neuropil_synapse_table": "In-out synapse & partner counts by neuropil."
     "For every cell and neuropil (region), contains the number of input and output synapses, as well as the "
     "number of input and output partners the cell has in that neuropil. One row for each cell with the first "
     "column containing the FlyWire Root ID of the cell, followed by a column for every combination of neuropil X "

diff --git a/src/data/downloadable_files_metadata.json b/src/data/downloadable_files_metadata.json
@@ -7,8 +7,9 @@
       "morphology_clusters.csv.gz": 736744,
       "connectivity_clusters.csv.gz": 522912,
       "connections.csv.gz": 26124953,
-      "labels.csv.gz": 3519519,
+      "labels.csv.gz": 3533768,
       "coordinates.csv.gz": 4584155,
+      "connectivity_tags.csv.gz": 653687,
       "neuropil_synapse_table.csv.gz": 4490002
     },
     "571": {
@@ -17,8 +18,9 @@
       "cell_stats.csv.gz": 2257920,
       "morphology_clusters.csv.gz": 209413,
       "connections.csv.gz": 32220716,
-      "labels.csv.gz": 2924127,
+      "labels.csv.gz": 3049771,
       "coordinates.csv.gz": 3945113,
+      "connectivity_tags.csv.gz": 616461,
       "neuropil_synapse_table.csv.gz": 2866446
     }
   },
@@ -82,17 +84,17 @@
         "col 5 - nt_type": "6 unique values"
       },
       "labels.csv.gz": {
-        "# rows": "123,985 (+ header)",
+        "# rows": "124,394 (+ header)",
         "# columns": "9",
-        "col 1 - root_id": "90,112 unique values",
-        "col 2 - label": "6,601 unique values",
-        "col 3 - user_id": "132 unique values",
-        "col 4 - position": "102,955 unique values",
-        "col 5 - supervoxel_id": "102,224 unique values",
+        "col 1 - root_id": "90,394 unique values",
+        "col 2 - label": "6,612 unique values",
+        "col 3 - user_id": "133 unique values",
+        "col 4 - position": "103,364 unique values",
+        "col 5 - supervoxel_id": "102,633 unique values",
         "col 6 - label_id": "all rows contain unique values",
-        "col 7 - date_created": "37,271 unique values",
-        "col 8 - user_name": "129 unique values in 123,335 rows, empty in 650 rows",
-        "col 9 - user_affiliation": "52 unique values in 123,272 rows, empty in 713 rows"
+        "col 7 - date_created": "37,443 unique values",
+        "col 8 - user_name": "130 unique values in 123,744 rows, empty in 650 rows",
+        "col 9 - user_affiliation": "52 unique values in 123,681 rows, empty in 713 rows"
       },
       "coordinates.csv.gz": {
         "# rows": "207,700 (+ header)",
@@ -101,6 +103,12 @@
         "col 2 - position": "196,123 unique values",
         "col 3 - supervoxel_id": "193,412 unique values"
       },
+      "connectivity_tags.csv.gz": {
+        "# rows": "117,334 (+ header)",
+        "# columns": "2",
+        "col 1 - root_id": "all rows contain unique values",
+        "col 2 - connectivity_tag": "69 unique values"
+      },
       "neuropil_synapse_table.csv.gz": {
         "# rows": "124,891 (+ header)",
         "# columns": "321",
@@ -480,17 +488,17 @@
         "col 5 - nt_type": "6 unique values"
       },
       "labels.csv.gz": {
-        "# rows": "110,661 (+ header)",
+        "# rows": "114,356 (+ header)",
         "# columns": "9",
-        "col 1 - root_id": "79,296 unique values",
-        "col 2 - label": "6,221 unique values",
-        "col 3 - user_id": "129 unique values",
-        "col 4 - position": "90,152 unique values",
-        "col 5 - supervoxel_id": "89,501 unique values",
+        "col 1 - root_id": "82,182 unique values",
+        "col 2 - label": "6,325 unique values",
+        "col 3 - user_id": "130 unique values",
+        "col 4 - position": "93,844 unique values",
+        "col 5 - supervoxel_id": "93,191 unique values",
         "col 6 - label_id": "all rows contain unique values",
-        "col 7 - date_created": "4,496 unique values",
-        "col 8 - user_name": "126 unique values in 110,020 rows, empty in 641 rows",
-        "col 9 - user_affiliation": "50 unique values in 109,957 rows, empty in 704 rows"
+        "col 7 - date_created": "5,688 unique values",
+        "col 8 - user_name": "127 unique values in 113,715 rows, empty in 641 rows",
+        "col 9 - user_affiliation": "50 unique values in 113,652 rows, empty in 704 rows"
       },
       "coordinates.csv.gz": {
         "# rows": "179,104 (+ header)",
@@ -499,6 +507,12 @@
         "col 2 - position": "169,473 unique values",
         "col 3 - supervoxel_id": "167,584 unique values"
       },
+      "connectivity_tags.csv.gz": {
+        "# rows": "111,607 (+ header)",
+        "# columns": "2",
+        "col 1 - root_id": "all rows contain unique values",
+        "col 2 - connectivity_tag": "27 unique values"
+      },
       "neuropil_synapse_table.csv.gz": {
         "# rows": "117,605 (+ header)",
         "# columns": "301",

diff --git a/src/service/cell_details.py b/src/service/cell_details.py
@@ -28,7 +28,6 @@
     display,
 )
 from src.utils.graph_algos import reachable_node_counts
-from src.utils.markers import extract_markers
 
 
 @lru_cache
@@ -102,7 +101,7 @@ def cached_cell_details(
         "Connectivity Tags<br><small>"
         '<a href="" data-toggle="modal" data-target="#connectivityTagsModal">'
         'info & credits <i class="fa-solid fa-up-right-from-square"></i></a></small>': concat_labels(
-            extract_markers(neuron_data=nd, markers_type="connectivity_label")
+            nd["connectivity_tag"]
         ),
         "Community Labels<br><small>"
         f'<a href="{url_for("app.labeling_log", root_id=root_id)}" target="_blank">'
@@ -155,7 +154,7 @@ def insert_further_analysis_links(key, icon, search_endpoint):
                 "app.search", filter_string=f"{OP_DOWNSTREAM} {root_id}"
             ),
         )
-        if "connectivity_label:reciprocal" in nd["marker"]:
+        if "reciprocal" in nd["connectivity_tag"]:
             up, dn = neuron_db.connections_up_down(root_id)
             reciprocal_count = len(set(up).intersection(dn))
             insert_related_cell_links(

diff --git a/src/templates/download.html b/src/templates/download.html
@@ -72,5 +72,6 @@
 </div>
 
 {% include "cell_annotations_modal.html" %}
+{% include "connectivity_tags_modal.html" %}
 
 {% endblock content %}
diff --git a/src/utils/label_cleaning.py b/src/utils/label_cleaning.py
@@ -176,7 +176,11 @@ def remove_corrected(labels_latest_to_oldest):
 
 # labels we want to exclude (non-informative wrt all other corrections)
 def blacklisted(lbl):
-    blacklisted_labels = {"not a neuron", "not a neuron; glia", "correction - not optic lobe"}
+    blacklisted_labels = {
+        "not a neuron",
+        "not a neuron; glia",
+        "correction - not optic lobe",
+    }
     return lbl.lower() in blacklisted_labels or can_be_flywire_root_id(lbl)
 
 

diff --git a/tests/unit/test_neuron_data.py b/tests/unit/test_neuron_data.py
@@ -561,7 +561,8 @@ def test_no_garbage_labels(self):
                     "part of comprehensive neck connective tracing",
                 ]:
                     self.assertFalse(
-                        garbage.lower() in lbl.lower(), f"{lbl} contains {garbage}, all labels: {labels}"
+                        garbage.lower() in lbl.lower(),
+                        f"{lbl} contains {garbage}, all labels: {labels}",
                     )
 
     def test_thumbnails(self):