Add RNACentral to DB downloader

Author: milot-mirdita

data/workflow/databases.sh

@@ -285,6 +285,14 @@ case "${SELECTION}" in
       push_back "${TMP_PATH}/silva.fasta.gz"
       INPUT_TYPE="FASTA_LIST"
     ;;
+    "RNAcentral")
+        if notExists "${TMP_PATH}/rnacentral.fasta.gz"; then
+            downloadFile "https://ftp.ebi.ac.uk/pub/databases/RNAcentral/releases/${RNACENTRAL_REL}/release_notes.txt" "${TMP_PATH}/version"
+            downloadFile "https://ftp.ebi.ac.uk/pub/databases/RNAcentral/releases/${RNACENTRAL_REL}/sequences/rnacentral_species_specific_ids.fasta.gz" "${TMP_PATH}/rnacentral.fasta.gz"
+        fi
+        push_back "${TMP_PATH}/rnacentral.fasta.gz"
+        INPUT_TYPE="FASTA_LIST"
+    ;;
     "Kalamari")
         if notExists "${TMP_PATH}/kalamari.tsv"; then
             printf "3.7 %s\n" "$(date "+%s")" > "${TMP_PATH}/version"

src/workflow/Databases.cpp

@@ -157,6 +157,13 @@ std::vector<DatabaseDownload> downloads = {{
                                                    "https://www.arb-silva.de",
                                                    true, Parameters::DBTYPE_NUCLEOTIDES, databases_sh, databases_sh_len,
                                                    { { "SILVA_REL", "138" } }
+                                           }, {
+                                                   "RNAcentral",
+                                                   "RNAcentral is a comprehensive database of non-coding RNA sequences that aggregates data from multiple expert databases.",
+                                                   "RNAcentral Consortium: RNAcentral 2021: secondary structure integration, improved sequence search and new member databases. Nucleic Acids Res 49(D1), D212-D220 (2021)",
+                                                   "https://rnacentral.org",
+                                                   false, Parameters::DBTYPE_NUCLEOTIDES, databases_sh, databases_sh_len,
+                                                   { { "RNACENTRAL_REL", "26.0" } }
                                            }, {
                                                    "Resfinder",
                                                    "ResFinder is a database that captures antimicrobial resistance genes from whole-genome data sets.",