Merge branch 'develop'

current_version.txt

@@ -1,3 +1,3 @@
 VERSION_MAJOR 1
 VERSION_MINOR 5
-VERSION_PATCH 1
+VERSION_PATCH 2

doc/source/alevin.rst

@@ -82,9 +82,9 @@ values here can speed up the run substantially. In our testing we found that usu
 
 This is an optional argument, where user can explicitly specify the whitelist CB to use for cell detection and CB sequence correction. If not given, alevin generates its own set of putative CBs.
 
-.. note:: Not 10x 724k whitelist
+.. note:: Not 10x 737k whitelist
 
-   This flag does not use the biologically known whitelist provided by 10x, instead it's per experiment level whitelist file e.g. the file generated by cellranger with the name `barcodes.tsv`.
+   This flag does not use the technologically defined whitelisted cellular barcodes provided by 10x, instead it's a per experiment level list of subsampled cellular barcodes that need to quantified for consistency with other tools for example an input would be a file generated by cellranger with the name `barcodes.tsv` (uncompressed).
 
 """"""""""""
 ``--noQuant``

doc/source/conf.py

@@ -57,7 +57,7 @@
 # The short X.Y version.
 version = '1.5'
 # The full version, including alpha/beta/rc tags.
-release = '1.5.1'
+release = '1.5.2'
 
 # The language for content autogenerated by Sphinx. Refer to documentation
 # for a list of supported languages.

docker/Dockerfile

@@ -6,7 +6,7 @@ MAINTAINER salmon.maintainer@gmail.com
 
 ENV PACKAGES git gcc make g++ libboost-all-dev liblzma-dev libbz2-dev \
     ca-certificates zlib1g-dev libcurl4-openssl-dev curl unzip autoconf apt-transport-https ca-certificates gnupg software-properties-common wget
-ENV SALMON_VERSION 1.5.1
+ENV SALMON_VERSION 1.5.2
 
 # salmon binary will be installed in /home/salmon/bin/salmon
 

docker/build_test.sh

@@ -1,3 +1,3 @@
 #! /bin/bash
-SALMON_VERSION=1.5.1
+SALMON_VERSION=1.5.2
 docker build --no-cache -t combinelab/salmon:${SALMON_VERSION} -t combinelab/salmon:latest .

include/GZipWriter.hpp

@@ -41,6 +41,9 @@ class GZipWriter {
   template <typename ExpT>
   bool writeMeta(const SalmonOpts& opts, const ExpT& experiment, const MappingStatistics& mstats);
 
+  template <typename SCExpT>
+  bool writeMetaFryMode(const SalmonOpts& opts, const SCExpT& experiment, const MappingStatistics& mstats);
+
   template <typename ProtocolT>
   bool writeMetaAlevin(const AlevinOpts<ProtocolT>& opts,
                        boost::filesystem::path aux_dir);

include/SalmonConfig.hpp

@@ -9,13 +9,13 @@
     the Free Software Foundation, either version 3 of the License, or
     (at your option) any later version.
 
-    Sailfish is distributed in the hope that it will be useful,
+    Salmon is distributed in the hope that it will be useful,
     but WITHOUT ANY WARRANTY; without even the implied warranty of
     MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
     GNU General Public License for more details.
 
     You should have received a copy of the GNU General Public License
-    along with Sailfish.  If not, see <http://www.gnu.org/licenses/>.
+    along with Salmon.  If not, see <http://www.gnu.org/licenses/>.
 <HEADER
 **/
 
@@ -27,8 +27,8 @@
 namespace salmon {
 constexpr char majorVersion[] = "1";
 constexpr char minorVersion[] = "5";
-constexpr char patchVersion[] = "1";
-constexpr char version[] = "1.5.1";
+constexpr char patchVersion[] = "2";
+constexpr char version[] = "1.5.2";
 constexpr uint32_t indexVersion = 5;
 constexpr char requiredQuasiIndexVersion[] = "p7";
 } // namespace salmon

include/cuckoohash_map.hh

@@ -705,7 +705,9 @@ private:
   // true if the key is small and simple, which means using partial keys for
   // lookup would probably slow us down
   static constexpr bool is_simple() {
-    return std::is_pod<key_type>::value && sizeof(key_type) <= 8;
+    return std::is_standard_layout<key_type>::value &&
+           std::is_trivial<key_type>::value &&
+           sizeof(key_type) <= 8;
   }
 
   // Whether or not the data is nothrow-move-constructible.

scripts/fetchPufferfish.sh

@@ -22,11 +22,11 @@ if [ -d ${INSTALL_DIR}/src/pufferfish ] ; then
     rm -fr ${INSTALL_DIR}/src/pufferfish
 fi
 
-SVER=salmon-v1.5.1
+SVER=salmon-v1.5.2
 #SVER=develop
 #SVER=sketch-mode
 
-EXPECTED_SHA256=468e0c23a32d81524f7acadc8326efb155628970c15fd6cb843d26a61478bfde
+EXPECTED_SHA256=86c7ff465d40b8184dca7f6afee693ad1db63be5bf63242161ea39d3507d6d25
 
 mkdir -p ${EXTERNAL_DIR}
 curl -k -L https://github.com/COMBINE-lab/pufferfish/archive/${SVER}.zip -o ${EXTERNAL_DIR}/pufferfish.zip

src/Alevin.cpp

@@ -890,14 +890,24 @@ void initiatePipeline(AlevinOpts<ProtocolT>& aopt,
 
   if (aopt.just_align) {
     // if we are just aligning 
+
+    // write out the cmd_info.json to make sure we have that
+    boost::filesystem::path outputDirectory = vm["output"].as<std::string>();
+    bool isWriteOk = aut::writeCmdInfo(outputDirectory / "cmd_info.json", orderedOptions);
+    if(!isWriteOk){
+      fmt::print(stderr, "Writing cmd_info.json in output directory failed.\nExiting now.");
+      exit(1);
+    }
+
+    // do the actual mapping
     auto rc = alevin_sc_align(aopt, sopt, orderedOptions);
+
     if (rc == 0) {
       aopt.jointLog->info("sc-align successful.");
-      std::exit(0);
     } else {
       aopt.jointLog->error("sc-align exited with return code {}", rc);
-      std::exit(rc);
     }
+    std::exit(rc);
   }
 
   /*
@@ -946,7 +956,7 @@ void initiatePipeline(AlevinOpts<ProtocolT>& aopt,
     // Write out information about the command / run
     bool isWriteOk = aut::writeCmdInfo(cmdInfoPath / "cmd_info.json", orderedOptions);
     if(!isWriteOk){
-      fmt::print(stderr, "writing in output directory failed\n Exiting Now");
+      fmt::print(stderr, "Writing cmd_info.json in output directory failed.\nExiting now.");
       exit(1);
     }
   }
@@ -1010,7 +1020,6 @@ salmon-based processing of single-cell RNA-seq data.
     green[3] = '0' + static_cast<char>(fmt::GREEN);
     red[3] = '0' + static_cast<char>(fmt::RED);
 
-
     bool noTgMap {false};
     bool dropseq = vm["dropseq"].as<bool>();
     bool indrop = vm["indrop"].as<bool>();

src/GZipWriter.cpp

@@ -891,6 +891,84 @@ bool GZipWriter::writeMeta(const SalmonOpts& opts, const ExpT& experiment, const
   return true;
 }
 
+/**
+ * Write the ``main'' metadata to file when executing in alevin-fry mode.  Currently this 
+ * writes a stripped down version of meta_info.json:
+ *   -- A json file with information about the run
+ */
+template <typename ExpT>
+bool GZipWriter::writeMetaFryMode(const SalmonOpts& opts, const ExpT& experiment, const MappingStatistics& mstats) {
+
+  namespace bfs = boost::filesystem;
+  using salmon::utils::DuplicateTargetStatus;
+
+  bfs::path auxDir = path_ / opts.auxDir;
+  bool auxSuccess = boost::filesystem::create_directories(auxDir);
+  auto numBootstraps = 0;
+  auto numSamples = 0;
+
+  bfs::path info = auxDir / "meta_info.json";
+  {
+    std::ofstream os(info.string());
+    cereal::JSONOutputArchive oa(os);
+
+    std::string sampType = "none";
+    auto& transcripts = experiment.transcripts();
+    oa(cereal::make_nvp("salmon_version", std::string(salmon::version)));
+    oa(cereal::make_nvp("samp_type", sampType));
+
+    std::string optType = "rad_mode";
+    oa(cereal::make_nvp("opt_type", optType));
+
+    std::vector<std::string> errors;
+    oa(cereal::make_nvp("quant_errors", errors));
+
+    auto libStrings = getLibTypeStrings(experiment);
+    oa(cereal::make_nvp("num_libraries", libStrings.size()));
+    oa(cereal::make_nvp("library_types", libStrings));
+
+    auto has_dups = experiment.index_retains_duplicates();
+    switch(has_dups) {
+      case DuplicateTargetStatus::RETAINED_DUPLICATES:
+        oa(cereal::make_nvp("keep_duplicates", true));
+        break;
+      case DuplicateTargetStatus::REMOVED_DUPLICATES:
+        oa(cereal::make_nvp("keep_duplicates", false));
+        break;
+      case DuplicateTargetStatus::UNKNOWN:
+      default:
+        break;
+    }
+
+    auto numValidTargets = transcripts.size();
+    auto numDecoys = experiment.getNumDecoys();
+    oa(cereal::make_nvp("num_valid_targets", numValidTargets));
+    oa(cereal::make_nvp("num_decoy_targets", numDecoys));
+
+    oa(cereal::make_nvp("length_classes", experiment.getLengthQuantiles()));
+    oa(cereal::make_nvp("index_seq_hash", experiment.getIndexSeqHash256()));
+    oa(cereal::make_nvp("index_name_hash", experiment.getIndexNameHash256()));
+    oa(cereal::make_nvp("index_seq_hash512", experiment.getIndexSeqHash512()));
+    oa(cereal::make_nvp("index_name_hash512", experiment.getIndexNameHash512()));
+    oa(cereal::make_nvp("index_decoy_seq_hash", experiment.getIndexDecoySeqHash256()));
+    oa(cereal::make_nvp("index_decoy_name_hash", experiment.getIndexDecoyNameHash256()));
+    oa(cereal::make_nvp("num_bootstraps", numSamples));
+    oa(cereal::make_nvp("num_processed", experiment.numObservedFragments()));
+    oa(cereal::make_nvp("num_mapped", experiment.numMappedFragments()));
+    //oa(cereal::make_nvp("num_decoy_fragments", mstats.numDecoyFragments.load()));
+    //oa(cereal::make_nvp("num_dovetail_fragments", mstats.numDovetails.load()));
+    oa(cereal::make_nvp("num_fragments_filtered_vm", mstats.numFragmentsFiltered.load()));
+    oa(cereal::make_nvp("num_alignments_below_threshold_for_mapped_fragments_vm",
+                        mstats.numMappingsFiltered.load()));
+    //oa(cereal::make_nvp("percent_mapped",
+    //                    experiment.effectiveMappingRate() * 100.0));
+    oa(cereal::make_nvp("call", std::string("quant")));
+    oa(cereal::make_nvp("start_time", opts.runStartTime));
+    oa(cereal::make_nvp("end_time", opts.runStopTime));
+  }
+  return true;
+}
+
 bool GZipWriter::writeAbundances(
                                  std::vector<double>& alphas,
                                  std::vector<Transcript>& transcripts) {
@@ -1730,6 +1808,11 @@ GZipWriter::writeMeta<SCExpT>(const SalmonOpts& opts,
                               const SCExpT& experiment,
                               const MappingStatistics& mstats);
 
+template bool
+GZipWriter::writeMetaFryMode<SCExpT>(const SalmonOpts& opts,
+                              const SCExpT& experiment,
+                              const MappingStatistics& mstats);
+
 template bool
 GZipWriter::writeMeta<BulkAlignLibT<UnpairedRead,AlignmentModel>>(const SalmonOpts& opts, const BulkAlignLibT<UnpairedRead,AlignmentModel>& experiment,
                                                    const MappingStatistics& mstats);

src/SalmonAlevin.cpp

@@ -2227,6 +2227,7 @@ void do_sc_align(ReadExperimentT& experiment,
   }
 
 
+
   BasicBinWriter bw;
   //  RADHeader
   RADHeader rh;
@@ -2667,8 +2668,8 @@ int alevin_sc_align(AlevinOpts<ProtocolT>& aopt,
   using std::string;
   namespace bfs = boost::filesystem;
   namespace po = boost::program_options;
-  try{
-    //auto fileLog = sopt.fileLog;
+
+  try {
     auto jointLog = aopt.jointLog;
     auto indexDirectory = sopt.indexDirectory;
     auto outputDirectory = sopt.outputDirectory;
@@ -2748,6 +2749,11 @@ int alevin_sc_align(AlevinOpts<ProtocolT>& aopt,
 
     do_sc_align<QuasiAlignment>(experiment, sopt,
                                 mstats, sopt.numThreads, aopt);
+
+    // write meta-information about the run
+    GZipWriter gzw(outputDirectory, jointLog);
+    sopt.runStopTime = salmon::utils::getCurrentTimeAsString();
+    gzw.writeMetaFryMode(sopt, experiment, mstats);
  } catch (po::error& e) {
     std::cerr << "Exception : [" << e.what() << "]. Exiting.\n";
     std::exit(1);
@@ -2761,7 +2767,6 @@ int alevin_sc_align(AlevinOpts<ProtocolT>& aopt,
               << " alevin --help\nExiting.\n";
     std::exit(1);
   }
-
   return 0;
 }