Added BFVD as a foldseek database (#344)

* Added BFVD as structure database * Minor modification * Added version file * Distinguished BFMD and BFVD * Changed url to the db * Add biorxiv id, enable taxdb flag and reorder --------- Co-authored-by: Milot Mirdita <[email protected]>
steineggerlab · Sep 10, 2024 · 52029c0 · 52029c0
1 parent be9fc33
commit 52029c0
Show file tree

Hide file tree

Showing 2 changed files with 17 additions and 1 deletion.
diff --git a/data/structdatabases.sh b/data/structdatabases.sh
@@ -182,6 +182,14 @@ case "${SELECTION}" in
         tar xvfz "${TMP_PATH}/${MODEL}" -C "${OUTDB}"
         INPUT_TYPE="MODEL_WEIGHTS"
     ;;
+    "BFVD")
+        if notExists "${TMP_PATH}/bfvd.tar.gz"; then
+            downloadFile "https://bfvd.steineggerlab.workers.dev/bfvd_foldseekdb.tar.gz" "${TMP_PATH}/bfvd.tar.gz"
+            downloadFile "https://bfvd.steineggerlab.workers.dev/bfvd.version" "${TMP_PATH}/version"
+        fi
+        tar xvfz "${TMP_PATH}/bfvd.tar.gz" -C "${TMP_PATH}"
+        push_back "${TMP_PATH}/bfvd"
+        INPUT_TYPE="FOLDSEEK_DB"
 esac
 
 if notExists "${OUTDB}.dbtype"; then

diff --git a/src/FoldseekBase.cpp b/src/FoldseekBase.cpp
@@ -435,11 +435,19 @@ std::vector<DatabaseDownload> externalDownloads = {
         {
                 "BFMD",
                 "BFMD Big fantastic multimer database (combined multimers from large prediction projects).",
-                "Kim et al. Rapid and Sensitive Protein Complex Alignment with Foldseek-Multimer. bioRxiv (2024)",
+                "Kim W et al. Rapid and Sensitive Protein Complex Alignment with Foldseek-Multimer. bioRxiv (2024)",
                 "https://foldseek.steineggerlab.workers.dev/bfmd.version",
                 true, Parameters::DBTYPE_AMINO_ACIDS, structdatabases_sh, structdatabases_sh_len,
                 {}
         },
+        {
+                "BFVD",
+                "BFVD Big fantastic virus database (predicted viral protein structures based on the viral sequence representatives of the UniRef30 clusters",
+                "Kim R et al. BFVD - a large repository of predicted viral protein structures. bioRxiv, 2024.09.08.611582v1 (2024)",
+                "https://bfvd.steineggerlab.workers.dev",
+                true, Parameters::DBTYPE_AMINO_ACIDS, structdatabases_sh, structdatabases_sh_len,
+                {}
+        },
         {
                 "ProstT5",
                 "Protein language model to predict 3Di directly from sequence.",