From 6af5bce18af8670ac9c8cc91cb85a923f1d97a79 Mon Sep 17 00:00:00 2001
From: Matin Nuhamunada <matinnu@biosustain.dtu.dk>
Date: Mon, 19 Feb 2024 15:33:42 +0000
Subject: [PATCH] fix: add empty handlers if no knownclusterblast hits are
 found

---
 .../bgcflow/data/get_antismash_overview.py    | 47 +++++++++++--------
 workflow/notebook/antismash.py.ipynb          | 34 ++++++++++++++
 2 files changed, 62 insertions(+), 19 deletions(-)

diff --git a/workflow/bgcflow/bgcflow/data/get_antismash_overview.py b/workflow/bgcflow/bgcflow/data/get_antismash_overview.py
index cfbb37bf..3f6d41eb 100644
--- a/workflow/bgcflow/bgcflow/data/get_antismash_overview.py
+++ b/workflow/bgcflow/bgcflow/data/get_antismash_overview.py
@@ -76,25 +76,34 @@ def get_antismash_overview(json_path, outfile, genome_id=False, n_hits=1):
             assert n_hits > 0
 
             output_hits = []
-
-            for n, hits in enumerate(knownclusterblast["ranking"]):
-                if n + 1 <= (n_hits):
-                    most_similar_mibig_id = hits[0]["accession"]
-                    most_similar_mibig_description = hits[0]["description"]
-                    most_similar_mibig_clustertype = hits[0]["cluster_type"]
-                    n_genes_in_target = len(hits[0]["tags"])
-                    n_genes_hits = hits[1]["hits"]
-                    hit_similarity = n_genes_hits / n_genes_in_target
-                    output_hits.append(
-                        {
-                            "most_similar_known_cluster_id": most_similar_mibig_id,
-                            "most_similar_known_cluster_description": most_similar_mibig_description,
-                            "most_similar_known_cluster_type": most_similar_mibig_clustertype,
-                            "similarity": hit_similarity,
-                        }
-                    )
-                else:
-                    pass
+            logging.info(f'Getting most similar known clusters from KnownClusterBlast')
+            if knownclusterblast["total_hits"] == 0:
+                logging.debug("No knowncluster blast hits found. Returning empty values.")
+                output_hits.append(
+                    {
+                        "most_similar_known_cluster_id": None,
+                        "most_similar_known_cluster_description": None,
+                        "most_similar_known_cluster_type": None,
+                        "similarity": None,
+                    }
+                )
+            else:
+                for n, hits in enumerate(knownclusterblast["ranking"]):
+                    if n + 1 <= (n_hits):
+                        most_similar_mibig_id = hits[0]["accession"]
+                        most_similar_mibig_description = hits[0]["description"]
+                        most_similar_mibig_clustertype = hits[0]["cluster_type"]
+                        n_genes_in_target = len(hits[0]["tags"])
+                        n_genes_hits = hits[1]["hits"]
+                        hit_similarity = n_genes_hits / n_genes_in_target
+                        output_hits.append(
+                            {
+                                "most_similar_known_cluster_id": most_similar_mibig_id,
+                                "most_similar_known_cluster_description": most_similar_mibig_description,
+                                "most_similar_known_cluster_type": most_similar_mibig_clustertype,
+                                "similarity": hit_similarity,
+                            }
+                        )
 
             bgc_id = f"{record['id']}.region{str(c+1).zfill(3)}"
             output_cluster = {
diff --git a/workflow/notebook/antismash.py.ipynb b/workflow/notebook/antismash.py.ipynb
index afc07885..3f2c8d63 100644
--- a/workflow/notebook/antismash.py.ipynb
+++ b/workflow/notebook/antismash.py.ipynb
@@ -173,6 +173,40 @@
     "display(HTML(DT(df, columnDefs=[{\"className\": \"dt-center\", \"targets\": \"_all\"}],)))"
    ]
   },
+  {
+   "cell_type": "markdown",
+   "id": "fa933ce3",
+   "metadata": {},
+   "source": [
+    "## Regions Table\n",
+    "Click on the BGC ids to go to the region's antiSMASH result.\n",
+    "\n",
+    "[Download Table]({{ project().file_server() }}/tables/df_regions_antismash_{{project().dependency_version()}}.csv){:target=\"_blank\" .md-button}"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "0fdd5327",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "regions_table = report_dir / f\"tables/df_regions_antismash_{antismash_version}.csv\"\n",
+    "df_regions_table = pd.read_csv(regions_table)\n",
+    "\n",
+    "server_path = \"<a href='{{ project().file_server() }}/antismash/{{project().dependency_version()}}/\"\n",
+    "for i in df_regions_table.index:\n",
+    "    gid = df_regions_table.loc[i, \"genome_id\"]\n",
+    "    r, c = str(df_regions_table.loc[i, \"region\"]).split(\".\")\n",
+    "    region_id = f\"#r{r}c{c}\"\n",
+    "    bgc_id = df_regions_table.loc[i, \"bgc_id\"]\n",
+    "    df_regions_table.loc[i, \"Genome ID\"] = server_path + f\"{gid}/index.html' target='_blank''>{gid}</a>\"\n",
+    "    df_regions_table.loc[i, \"BGC ID\"] = server_path + f\"{gid}/index.html{region_id}' target='_blank''>{bgc_id}</a>\"\n",
+    "\n",
+    "display(HTML(DT(df_regions_table.loc[:, [\"BGC ID\", \"Genome ID\", \"region\", \"contig_edge\", \"product\", \"most_similar_known_cluster_id\", \n",
+    "                                         \"most_similar_known_cluster_description\", \"most_similar_known_cluster_type\", \"similarity\"]], columnDefs=[{\"className\": \"dt-center\", \"targets\": \"_all\"}],)))"
+   ]
+  },
   {
    "cell_type": "markdown",
    "id": "a9111659-b673-4842-8ef3-d0fa9b52aa8c",