From 09e3c75c2fca6278fb0885132d1a0029abd64999 Mon Sep 17 00:00:00 2001
From: valentynbez <walentyn.bzsz@gmail.com>
Date: Wed, 5 Jun 2024 22:27:20 +0200
Subject: [PATCH] fix: empty results against PDB

---
 mDeepFRI/mmseqs.py   | 11 ++++++++---
 mDeepFRI/pipeline.py |  9 +++++++--
 2 files changed, 15 insertions(+), 5 deletions(-)

diff --git a/mDeepFRI/mmseqs.py b/mDeepFRI/mmseqs.py
index 8b8d066..4c15452 100644
--- a/mDeepFRI/mmseqs.py
+++ b/mDeepFRI/mmseqs.py
@@ -341,9 +341,14 @@ def from_best_matches(cls, filepath: str):
                                    delimiter="\t",
                                    encoding="utf-8",
                                    names=True)
-
-        query_file = np.unique(result_arr["query_file"])[0]
-        database = np.unique(result_arr["database_file"])[0]
+        try:
+            query_file = np.unique(result_arr["query_file"])[0]
+        except IndexError:
+            query_file = None
+        try:
+            database = np.unique(result_arr["database_file"])[0]
+        except IndexError:
+            database = None
 
         return cls(result_arr, query_file, database)
 
diff --git a/mDeepFRI/pipeline.py b/mDeepFRI/pipeline.py
index 7097f3e..1879fe0 100644
--- a/mDeepFRI/pipeline.py
+++ b/mDeepFRI/pipeline.py
@@ -102,8 +102,13 @@ def hierarchical_database_search(query_file: str,
         # store the location of the result for the next step
         db.mmseqs_result = mmseqs_results_path
 
-        # logging
-        unique_hits = np.unique(best_matches["query"])
+        # catch error if no matches to database
+        # a case from phage proteins
+        try:
+            unique_hits = np.unique(best_matches["query"])
+        except IndexError:
+            unique_hits = np.array([])
+
         if "pdb100" in db.name:
             pdb_hits = unique_hits
         else: