Merge commit '97d4c6cfb57bb7f0994015580579f31a18aaf9c5'

steineggerlab · Aug 2, 2024 · 04876ca · 04876ca
2 parents bc212bc + 97d4c6c
commit 04876ca
Show file tree

Hide file tree

Showing 20 changed files with 293 additions and 54 deletions.
diff --git a/lib/mmseqs/CMakeLists.txt b/lib/mmseqs/CMakeLists.txt
@@ -117,6 +117,7 @@ if (NATIVE_ARCH AND (MMSEQS_ARCH STREQUAL ""))
         set(MMSEQS_ARCH "-march=native")
     endif ()
 endif ()
+set(MMSEQS_ARCH ${MMSEQS_ARCH} CACHE INTERNAL "")
 
 if (NOT (MMSEQS_ARCH STREQUAL ""))
     set(MMSEQS_CXX_FLAGS "${MMSEQS_CXX_FLAGS} ${MMSEQS_ARCH}")

diff --git a/lib/mmseqs/azure-pipelines.yml b/lib/mmseqs/azure-pipelines.yml
@@ -40,7 +40,7 @@ jobs:
   - job: build_macos
     displayName: macOS
     pool:
-      vmImage: 'macos-11'
+      vmImage: 'macos-12'
     steps:
       - checkout: self
         submodules: true

diff --git a/lib/mmseqs/data/workflow/createtaxdb.sh b/lib/mmseqs/data/workflow/createtaxdb.sh
@@ -40,13 +40,22 @@ downloadFile() {
         ARIA)
             FILENAME=$(basename "${OUTPUT}")
             DIR=$(dirname "${OUTPUT}")
-            aria2c --max-connection-per-server="$ARIA_NUM_CONN" --allow-overwrite=true -o "$FILENAME" -d "$DIR" "$URL" && return 0
+            if aria2c -c --max-connection-per-server="$ARIA_NUM_CONN" --allow-overwrite=true -o "${FILENAME}.aria2" -d "$DIR" "$URL"; then
+                mv -f -- "${OUTPUT}.aria2" "${OUTPUT}"
+                return 0
+            fi
             ;;
         CURL)
-            curl -o "$OUTPUT" "$URL" && return 0
+            if curl -L -C - -o "${OUTPUT}.curl" "$URL"; then
+                mv -f -- "${OUTPUT}.curl" "${OUTPUT}"
+                return 0
+            fi
             ;;
         WGET)
-            wget -O "$OUTPUT" "$URL" && return 0
+            if wget -O "${OUTPUT}.wget" -c "$URL"; then
+                mv -f -- "${OUTPUT}.wget" "${OUTPUT}"
+                return 0
+            fi
             ;;
         esac
     done
@@ -59,7 +68,7 @@ if { [ "${DBMODE}" = "1" ] && notExists "${TAXDBNAME}_taxonomy"; } || { [ "${DBM
         # Download NCBI taxon information
         if notExists "${TMP_PATH}/ncbi_download.complete"; then
             echo "Download taxdump.tar.gz"
-            downloadFile "https://ftp.ncbi.nih.gov/pub/taxonomy/taxdump.tar.gz" "${TMP_PATH}/taxdump.tar.gz"
+            downloadFile "https://ftp.ncbi.nlm.nih.gov/pub/taxonomy/taxdump.tar.gz" "${TMP_PATH}/taxdump.tar.gz"
             tar -C "${TMP_PATH}" -xzf "${TMP_PATH}/taxdump.tar.gz" names.dmp nodes.dmp merged.dmp delnodes.dmp
             touch "${TMP_PATH}/ncbi_download.complete"
             rm -f "${TMP_PATH}/taxdump.tar.gz"

diff --git a/lib/mmseqs/data/workflow/databases.sh b/lib/mmseqs/data/workflow/databases.sh
@@ -40,13 +40,22 @@ downloadFile() {
         ARIA)
             FILENAME=$(basename "${OUTPUT}")
             DIR=$(dirname "${OUTPUT}")
-            aria2c --max-connection-per-server="$ARIA_NUM_CONN" --allow-overwrite=true -o "$FILENAME" -d "$DIR" "$URL" && return 0
+            if aria2c -c --max-connection-per-server="$ARIA_NUM_CONN" --allow-overwrite=true -o "${FILENAME}.aria2" -d "$DIR" "$URL"; then
+                mv -f -- "${OUTPUT}.aria2" "${OUTPUT}"
+                return 0
+            fi
             ;;
         CURL)
-            curl -L -o "$OUTPUT" "$URL" && return 0
+            if curl -L -C - -o "${OUTPUT}.curl" "$URL"; then
+                mv -f -- "${OUTPUT}.curl" "${OUTPUT}"
+                return 0
+            fi
             ;;
         WGET)
-            wget -O "$OUTPUT" "$URL" && return 0
+            if wget -O "${OUTPUT}.wget" -c "$URL"; then
+                mv -f -- "${OUTPUT}.wget" "${OUTPUT}"
+                return 0
+            fi
             ;;
         esac
     done
@@ -118,9 +127,9 @@ case "${SELECTION}" in
         if notExists "${TMP_PATH}/nr.gz"; then
             date "+%s" > "${TMP_PATH}/version"
             downloadFile "https://ftp.ncbi.nlm.nih.gov/blast/db/FASTA/nr.gz" "${TMP_PATH}/nr.gz"
-            downloadFile "https://ftp.ncbi.nih.gov/pub/taxonomy/accession2taxid/prot.accession2taxid.gz" "${TMP_PATH}/prot.accession2taxid.gz"
+            downloadFile "https://ftp.ncbi.nlm.nih.gov/pub/taxonomy/accession2taxid/prot.accession2taxid.gz" "${TMP_PATH}/prot.accession2taxid.gz"
             gunzip "${TMP_PATH}/prot.accession2taxid.gz"
-            downloadFile "https://ftp.ncbi.nih.gov/pub/taxonomy/accession2taxid/pdb.accession2taxid.gz" "${TMP_PATH}/pdb.accession2taxid.gz"
+            downloadFile "https://ftp.ncbi.nlm.nih.gov/pub/taxonomy/accession2taxid/pdb.accession2taxid.gz" "${TMP_PATH}/pdb.accession2taxid.gz"
             gunzip "${TMP_PATH}/pdb.accession2taxid.gz"
         fi
         push_back "${TMP_PATH}/nr.gz"
@@ -212,8 +221,8 @@ case "${SELECTION}" in
     ;;
     "CDD")
         if notExists "${TMP_PATH}/msa.msa.gz"; then
-            downloadFile "https://ftp.ncbi.nih.gov/pub/mmdb/cdd/cdd.info" "${TMP_PATH}/version"
-            downloadFile "https://ftp.ncbi.nih.gov/pub/mmdb/cdd/fasta.tar.gz" "${TMP_PATH}/msa.tar.gz"
+            downloadFile "https://ftp.ncbi.nlm.nih.gov/pub/mmdb/cdd/cdd.info" "${TMP_PATH}/version"
+            downloadFile "https://ftp.ncbi.nlm.nih.gov/pub/mmdb/cdd/fasta.tar.gz" "${TMP_PATH}/msa.tar.gz"
         fi
         INPUT_TYPE="FASTA_MSA"
         SED_FIX_LOOKUP='s|\.FASTA||g'

diff --git a/lib/mmseqs/data/workflow/taxpercontig.sh b/lib/mmseqs/data/workflow/taxpercontig.sh
@@ -45,7 +45,11 @@ if [ -n "${ORF_FILTER}" ]; then
     fi
 
     if notExists "${TMP_PATH}/orfs_aln.list"; then
-        awk '$3 > 1 { print $1 }' "${TMP_PATH}/orfs_aln.index" > "${TMP_PATH}/orfs_aln.list"
+        # shellcheck disable=SC2086
+        "$MMSEQS" recoverlongestorf "${ORFS_DB}" "${TMP_PATH}/orfs_aln" "${TMP_PATH}/orfs_aln_recovered.list" ${THREADS_PAR} \
+            || fail "recoverlongestorf died"
+        awk '$3 > 1 { print $1 }' "${TMP_PATH}/orfs_aln.index" > "${TMP_PATH}/orfs_aln_remain.list"
+        cat  "${TMP_PATH}/orfs_aln_recovered.list" "${TMP_PATH}/orfs_aln_remain.list" > "${TMP_PATH}/orfs_aln.list"
     fi
 
     if notExists "${TMP_PATH}/orfs_filter.dbtype"; then

diff --git a/lib/mmseqs/data/workflow/tsv2exprofiledb.sh b/lib/mmseqs/data/workflow/tsv2exprofiledb.sh
@@ -17,25 +17,37 @@ OUT="$2"
 [ -d "${OUT}.tsv" ] && echo "${OUT} is a directory!" && exit 1;
 
 if notExists "${OUT}_h.dbtype"; then
-  "$MMSEQS" tsv2db "${IN}_h.tsv" "${OUT}_h" --output-dbtype 12 ${VERBOSITY}
+  MMSEQS_FORCE_MERGE=1 "$MMSEQS" tsv2db "${IN}_h.tsv" "${OUT}_h" --output-dbtype 12 ${VERBOSITY}
 fi
 
 if notExists "${OUT}.dbtype"; then
-  "$MMSEQS" tsv2db "${IN}.tsv" "${OUT}_tmp" --output-dbtype 0 ${VERBOSITY}
-  MMSEQS_FOCE_MERGE=1 "$MMSEQS" compress "${OUT}_tmp" "${OUT}" ${VERBOSITY}
-  "$MMSEQS" rmdb "${OUT}_tmp" ${VERBOSITY}
+  if [ -n "${COMPRESSED}" ]; then
+    "$MMSEQS" tsv2db "${IN}.tsv" "${OUT}_tmp" --output-dbtype 0 ${VERBOSITY}
+    MMSEQS_FORCE_MERGE=1 "$MMSEQS" compress "${OUT}_tmp" "${OUT}" ${VERBOSITY}
+    "$MMSEQS" rmdb "${OUT}_tmp" ${VERBOSITY}
+  else
+    MMSEQS_FORCE_MERGE=1 "$MMSEQS" tsv2db "${IN}.tsv" "${OUT}" --output-dbtype 0 ${VERBOSITY}
+  fi
 fi
 
 if notExists "${OUT}_seq.dbtype"; then
-  "$MMSEQS" tsv2db "${IN}_seq.tsv" "${OUT}_seq_tmp" --output-dbtype 0 ${VERBOSITY}
-  MMSEQS_FOCE_MERGE=1 "$MMSEQS" compress "${OUT}_seq_tmp" "${OUT}_seq" ${VERBOSITY}
-  "$MMSEQS" rmdb "${OUT}_seq_tmp" ${VERBOSITY}
+  if [ -n "${COMPRESSED}" ]; then
+    "$MMSEQS" tsv2db "${IN}_seq.tsv" "${OUT}_seq_tmp" --output-dbtype 0 ${VERBOSITY}
+    MMSEQS_FORCE_MERGE=1 "$MMSEQS" compress "${OUT}_seq_tmp" "${OUT}_seq" ${VERBOSITY}
+    "$MMSEQS" rmdb "${OUT}_seq_tmp" ${VERBOSITY}
+  else
+    "$MMSEQS" tsv2db "${IN}_seq.tsv" "${OUT}_seq" --output-dbtype 0 ${VERBOSITY}
+  fi
 fi
 
 if notExists "${OUT}_aln.dbtype"; then
-  "$MMSEQS" tsv2db "${IN}_aln.tsv" "${OUT}_aln_tmp" --output-dbtype 5 ${VERBOSITY}
-  MMSEQS_FOCE_MERGE=1 "$MMSEQS" compress "${OUT}_aln_tmp" "${OUT}_aln" ${VERBOSITY}
-  "$MMSEQS" rmdb "${OUT}_aln_tmp" ${VERBOSITY}
+  if [ -n "${COMPRESSED}" ]; then
+    "$MMSEQS" tsv2db "${IN}_aln.tsv" "${OUT}_aln_tmp" --output-dbtype 5 ${VERBOSITY}
+    MMSEQS_FORCE_MERGE=1 "$MMSEQS" compress "${OUT}_aln_tmp" "${OUT}_aln" ${VERBOSITY}
+    "$MMSEQS" rmdb "${OUT}_aln_tmp" ${VERBOSITY}
+  else
+    MMSEQS_FORCE_MERGE=1 "$MMSEQS" tsv2db "${IN}_aln.tsv" "${OUT}_aln" --output-dbtype 5 ${VERBOSITY}
+  fi
 fi
 
 if notExists "${OUT}_seq_h.dbtype"; then

diff --git a/lib/mmseqs/src/CMakeLists.txt b/lib/mmseqs/src/CMakeLists.txt
@@ -131,9 +131,9 @@ else ()
 endif ()
 
 target_link_libraries(mmseqs-framework tinyexpr ${ZSTD_LIBRARIES} microtar)
-if (CYGWIN)
-    target_link_libraries(mmseqs-framework nedmalloc)
-endif ()
+# if (CYGWIN)
+#     target_link_libraries(mmseqs-framework nedmalloc)
+# endif ()
 
 if (EMSCRIPTEN)
     target_compile_definitions(mmseqs-framework PUBLIC -DHAVE_ZLIB=1 -DHAVE_BZLIB=1)
@@ -222,6 +222,7 @@ if (OPENMP_FOUND)
     if (NOT "${CMAKE_CXX_COMPILER_ID}" STREQUAL "GNU")
         target_link_libraries(mmseqs-framework ${OpenMP_CXX_LIBRARIES})
     endif()
+    target_include_directories(mmseqs-framework PUBLIC ${OpenMP_CXX_INCLUDE_DIRS})
     append_target_property(mmseqs-framework COMPILE_FLAGS ${OpenMP_CXX_FLAGS})
     append_target_property(mmseqs-framework LINK_FLAGS ${OpenMP_CXX_FLAGS})
 elseif (REQUIRE_OPENMP)

diff --git a/lib/mmseqs/src/CommandDeclarations.h b/lib/mmseqs/src/CommandDeclarations.h
@@ -98,6 +98,7 @@ extern int ungappedprefilter(int argc, const char **argv, const Command& command
 extern int gappedprefilter(int argc, const char **argv, const Command& command);
 extern int unpackdb(int argc, const char **argv, const Command& command);
 extern int rbh(int argc, const char **argv, const Command& command);
+extern int recoverlongestorf(int argc, const char **argv, const Command& command);
 extern int result2flat(int argc, const char **argv, const Command& command);
 extern int result2msa(int argc, const char **argv, const Command& command);
 extern int result2dnamsa(int argc, const char **argv, const Command& command);

diff --git a/lib/mmseqs/src/MMseqsBase.cpp b/lib/mmseqs/src/MMseqsBase.cpp
@@ -907,6 +907,14 @@ std::vector<Command> baseCommands = {
                 "Eli Levy Karin <eli.levy.karin@gmail.com> ",
                 "<i:contigsSequenceDB> <i:extractedOrfsHeadersDB> <o:orfsAlignedToContigDB>",
                 CITATION_MMSEQS2, {{"",DbType::ACCESS_MODE_INPUT, DbType::NEED_DATA, NULL}}},
+        {"recoverlongestorf",    recoverlongestorf,    &par.onlythreads,          COMMAND_EXPERT,
+                "Recover longest ORF for taxonomy annotation after elimination",
+                NULL,
+                "Sung-eun Jang",
+                "<i:orfDB> <i:resultDB> <o:tsvFile>",
+                CITATION_MMSEQS2, {{"orfDB", DbType::ACCESS_MODE_INPUT, DbType::NEED_DATA, &DbValidator::sequenceDb},
+                                                           {"resultDB", DbType::ACCESS_MODE_INPUT, DbType::NEED_DATA, &DbValidator::resultDb},
+                                                           {"tsvFile", DbType::ACCESS_MODE_OUTPUT, DbType::NEED_DATA, &DbValidator::flatfile}}},
         {"reverseseq",          reverseseq,            &par.reverseseq,           COMMAND_SEQUENCE,
                 "Reverse (without complement) sequences",
                 NULL,
@@ -1142,7 +1150,7 @@ std::vector<Command> baseCommands = {
                 "<i:hhsuiteHHMDB> <o:profileDB>",
                 CITATION_MMSEQS2,{{"",DbType::ACCESS_MODE_INPUT, DbType::NEED_DATA, NULL}}},
 
-        {"tsv2exprofiledb",      tsv2exprofiledb,      &par.onlyverbosity,        COMMAND_PROFILE_PROFILE,
+        {"tsv2exprofiledb",      tsv2exprofiledb,      &par.verbandcompression,   COMMAND_PROFILE_PROFILE,
                 "Create a expandable profile db from TSV files",
                 NULL,
                 "Milot Mirdita <milot@mirdita.de>",

diff --git a/lib/mmseqs/src/alignment/Alignment.cpp b/lib/mmseqs/src/alignment/Alignment.cpp
@@ -419,7 +419,7 @@ void Alignment::run(const std::string &outDB, const std::string &outDBIndex, con
 
                         // recompute alignment boundaries (without changing evalue)
                         const bool isIdentity = (queryDbKey == swResults[result].dbKey && (includeIdentity || sameQTDB)) ? true : false;
-                        Matcher::result_t res = realigner->getSWResult(&dbSeq, INT_MAX, false, realignCov, covThr, FLT_MAX, realignSwMode, seqIdMode, isIdentity);
+                        Matcher::result_t res = realigner->getSWResult(&dbSeq, INT_MAX, false, covMode, realignCov, FLT_MAX, realignSwMode, seqIdMode, isIdentity);
 
                         const bool covOK = Util::hasCoverage(realignCov, covMode, res.qcov, res.dbcov);
                         if (covOK == true || isIdentity) {

diff --git a/lib/mmseqs/src/commons/MultiParam.h b/lib/mmseqs/src/commons/MultiParam.h
@@ -23,21 +23,21 @@ class NuclAA {
     T first;
     T second;
 
-    NuclAA<T>(const NuclAA<T> &value) {
+    NuclAA(const NuclAA<T> &value) {
         this->first = value.first;
         this->second = value.second;
     }
 
     static const T max;
 
-    NuclAA<T>() {}
+    NuclAA() {}
 
-    NuclAA<T>(T first) {
+    NuclAA(T first) {
         this->first = first;
         this->second = first;
     }
 
-    NuclAA<T>(T first, T second) {
+    NuclAA(T first, T second) {
         this->first = first;
         this->second = second;
     }
@@ -88,21 +88,21 @@ class SeqProf {
     T first;
     T second;
 
-    SeqProf<T>(const SeqProf<T> &value) {
+    SeqProf(const SeqProf<T> &value) {
         this->first = value.first;
         this->second = value.second;
     }
 
     static const T max;
 
-    SeqProf<T>() {}
+    SeqProf() {}
 
-    SeqProf<T>(T first) {
+    SeqProf(T first) {
         this->first = first;
         this->second = first;
     }
 
-    SeqProf<T>(T first, T second) {
+    SeqProf(T first, T second) {
         this->first = first;
         this->second = second;
     }

diff --git a/lib/mmseqs/src/commons/Parameters.cpp b/lib/mmseqs/src/commons/Parameters.cpp
@@ -283,7 +283,7 @@ Parameters::Parameters():
         // taxonomyreport
         PARAM_REPORT_MODE(PARAM_REPORT_MODE_ID, "--report-mode", "Report mode", "Taxonomy report mode 0: Kraken 1: Krona", typeid(int), (void *) &reportMode, "^[0-1]{1}$"),
         // createtaxdb
-        PARAM_NCBI_TAX_DUMP(PARAM_NCBI_TAX_DUMP_ID, "--ncbi-tax-dump", "NCBI tax dump directory", "NCBI tax dump directory. The tax dump can be downloaded here \"ftp://ftp.ncbi.nih.gov/pub/taxonomy/taxdump.tar.gz\"", typeid(std::string), (void *) &ncbiTaxDump, ""),
+        PARAM_NCBI_TAX_DUMP(PARAM_NCBI_TAX_DUMP_ID, "--ncbi-tax-dump", "NCBI tax dump directory", "NCBI tax dump directory. The tax dump can be downloaded here \"ftp://ftp.ncbi.nlm.nih.gov/pub/taxonomy/taxdump.tar.gz\"", typeid(std::string), (void *) &ncbiTaxDump, ""),
         PARAM_TAX_MAPPING_FILE(PARAM_TAX_MAPPING_FILE_ID, "--tax-mapping-file", "Taxonomy mapping file", "File to map sequence identifier to taxonomical identifier", typeid(std::string), (void *) &taxMappingFile, ""),
         PARAM_TAX_MAPPING_MODE(PARAM_TAX_MAPPING_MODE_ID, "--tax-mapping-mode", "Taxonomy mapping mode", "Map taxonomy based on sequence database 0: .lookup file 1: .source file", typeid(int), (void *) &taxMappingMode, "^[0-1]{1}$"),
         PARAM_TAX_DB_MODE(PARAM_TAX_DB_MODE_ID, "--tax-db-mode", "Taxonomy db mode", "Create taxonomy database as: 0: .dmp flat files (human readable) 1: binary dump (faster readin)", typeid(int), (void *) &taxDbMode, "^[0-1]{1}$"),
@@ -1870,7 +1870,7 @@ void Parameters::parseParameters(int argc, const char *pargv[], const Command &c
                     EXIT(EXIT_FAILURE);
                 }
                 filenames.emplace_back(posix);
-                delete posix;
+                delete[] posix;
             }
 #else
             filenames.emplace_back(pargv[argIdx]);

diff --git a/lib/mmseqs/src/prefiltering/CacheFriendlyOperations.cpp b/lib/mmseqs/src/prefiltering/CacheFriendlyOperations.cpp
@@ -27,7 +27,7 @@ CacheFriendlyOperations<BINSIZE>::CacheFriendlyOperations(size_t maxElement, siz
 }
 
 template<unsigned int BINSIZE>
-CacheFriendlyOperations<BINSIZE>::~CacheFriendlyOperations<BINSIZE>(){
+CacheFriendlyOperations<BINSIZE>::~CacheFriendlyOperations(){
     delete[] duplicateBitArray;
     delete[] binDataFrame;
     delete[] tmpElementBuffer;

diff --git a/lib/mmseqs/src/util/CMakeLists.txt b/lib/mmseqs/src/util/CMakeLists.txt
@@ -47,6 +47,7 @@ set(util_source_files
         util/profile2pssm.cpp
         util/profile2neff.cpp
         util/profile2seq.cpp
+        util/recoverlongestorf.cpp
         util/result2dnamsa.cpp
         util/result2flat.cpp
         util/result2msa.cpp