CMakeLists.txt

 cmake_minimum_required(VERSION 3.2)
-project(spoa LANGUAGES CXX VERSION 1.1.3)
+project(spoa LANGUAGES CXX VERSION 1.1.5)
 
 include(GNUInstallDirs)
 

README.md

@@ -4,7 +4,7 @@
 [![Build status for c++/clang++](https://travis-ci.org/rvaser/spoa.svg?branch=master)](https://travis-ci.org/rvaser/spoa)
 [![Published in Genome Research](https://img.shields.io/badge/published%20in-Genome%20Research-blue.svg)](https://doi.org/10.1101/gr.214270.116)
 
-Spoa (SIMD POA) is a c++ implementation of the partial order alignment (POA) algorithm (as described in 10.1093/bioinformatics/18.3.452) which is used to generate consensus sequences (as described in 10.1093/bioinformatics/btg109). It supports three alignment modes: local (Smith-Waterman), global (Needleman-Wunsch) and semi-global alignment (overlap). It supports Intel SSE4.1+ and AVX2 (marginally faster due to high latency shifts) vectorization.
+Spoa (SIMD POA) is a c++ implementation of the partial order alignment (POA) algorithm (as described in 10.1093/bioinformatics/18.3.452) which is used to generate consensus sequences (as described in 10.1093/bioinformatics/btg109). It supports three alignment modes: local (Smith-Waterman), global (Needleman-Wunsch) and semi-global alignment (overlap). It supports Intel SSE4.1+ and AVX2 vectorization (marginally faster due to high latency shifts).
 
 ## Dependencies
 
@@ -39,6 +39,8 @@ Optionally, you can run `sudo make install` to install spoa library (and executa
 
 ***Note***: if you omitted `--recursive` from `git clone`, run `git submodule init` and `git submodule update` before proceeding with compilation.
 
+To build unit tests add `-Dspoa_build_tests=ON` while running `cmake`. After installation, an executable named `spoa_test` will be created in `build/bin`.
+
 ## Usage
 
 Usage of spoa is as following:
@@ -46,7 +48,8 @@ Usage of spoa is as following:
     spoa [options ...] <sequences>
 
         <sequences>
-            input file in FASTA/FASTQ format containing sequences
+            input file in FASTA/FASTQ format (can be compressed with gzip)
+            containing sequences
 
         options:
             -m, --match <int>
@@ -70,6 +73,8 @@ Usage of spoa is as following:
                     0 - consensus
                     1 - multiple sequence alignment
                     2 - 0 & 1
+            -d, --dot <file>
+                output file for the final POA graph in DOT format
             --version
                 prints the version number
             -h, --help

debian/changelog

+spoa (1.1.5-1) UNRELEASED; urgency=medium
+
+  * New upstream version
+  * debhelper 12
+  * Standards-Version: 4.3.0
+  * Upstream changed ABI and we stick here to the broken way to set
+    soversion = version but upstream was informed to invent better soversions
+    https://github.com/rvaser/spoa/issues/14
+
+ -- Andreas Tille <tille@debian.org>  Mon, 28 Jan 2019 20:04:58 +0100
+
 spoa (1.1.3-4) unstable; urgency=medium
 
   * Set architecture to amd64 since the code is not really portable

debian/compat

-11
+12

debian/control

@@ -3,14 +3,14 @@ Maintainer: Debian Med Packaging Team <debian-med-packaging@lists.alioth.debian.
 Uploaders: Andreas Tille <tille@debian.org>
 Section: science
 Priority: optional
-Build-Depends: debhelper (>= 11~),
+Build-Depends: debhelper (>= 12~),
                cmake,
                d-shlibs,
                rename,
                libbioparser-dev,
                libgtest-dev,
                zlib1g-dev
-Standards-Version: 4.2.1
+Standards-Version: 4.3.0
 Vcs-Browser: https://salsa.debian.org/med-team/spoa
 Vcs-Git: https://salsa.debian.org/med-team/spoa.git
 Homepage: https://github.com/rvaser/spoa
@@ -27,7 +27,7 @@ Description: SIMD partial order alignment tool
  (Smith-Waterman), global (Needleman-Wunsch) and semi-global alignment
  (overlap).
 
-Package: libspoa1.1.3
+Package: libspoa1.1.5
 Architecture: amd64
 Section: libs
 Depends: ${shlibs:Depends},
@@ -47,7 +47,7 @@ Architecture: amd64
 Section: libdevel
 Depends: ${shlibs:Depends},
          ${misc:Depends},
-         libspoa1.1.3 (= ${binary:Version})
+         libspoa1.1.5 (= ${binary:Version})
 Description: SIMD partial order alignment library (development files)
  Spoa (SIMD POA) is a c++ implementation of the partial order alignment
  (POA) algorithm (as described in 10.1093/bioinformatics/18.3.452) which

debian/libspoa1.1.3.symbols → debian/libspoa1.1.5.symbols

-libspoa.so.1.1.3 libspoa1.1.3 #MINVER#
+libspoa.so.1.1.5 libspoa1.1.5 #MINVER#
  _ZN4spoa11createGraphEv@Base 1.1.3
  _ZN4spoa15AlignmentEngine25align_sequence_with_graphERKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEERKSt10unique_ptrINS_5GraphESt14default_deleteISA_EE@Base 1.1.3
  _ZN4spoa15AlignmentEngineC1ENS_13AlignmentTypeEaaa@Base 1.1.3
  _ZN4spoa15AlignmentEngineC2ENS_13AlignmentTypeEaaa@Base 1.1.3
- (optional)_ZN4spoa19SimdAlignmentEngine10initializeINS_14InstructionSetIiEEEEvPKcRKSt10unique_ptrINS_5GraphESt14default_deleteIS7_EEjjj@Base 1.1.3
- (optional)_ZN4spoa19SimdAlignmentEngine10initializeINS_14InstructionSetIsEEEEvPKcRKSt10unique_ptrINS_5GraphESt14default_deleteIS7_EEjjj@Base 1.1.3
  _ZN4spoa19SimdAlignmentEngine25align_sequence_with_graphEPKcjRKSt10unique_ptrINS_5GraphESt14default_deleteIS4_EE@Base 1.1.3
- (optional)_ZN4spoa19SimdAlignmentEngine5alignINS_14InstructionSetIiEEEESt6vectorISt4pairIiiESaIS6_EEPKcjRKSt10unique_ptrINS_5GraphESt14default_deleteISC_EE@Base 1.1.3
- (optional)_ZN4spoa19SimdAlignmentEngine5alignINS_14InstructionSetIsEEEESt6vectorISt4pairIiiESaIS6_EEPKcjRKSt10unique_ptrINS_5GraphESt14default_deleteISC_EE@Base 1.1.3
  _ZN4spoa19SimdAlignmentEngine7reallocEjjj@Base 1.1.3
  _ZN4spoa19SimdAlignmentEngine8preallocEjj@Base 1.1.3
  _ZN4spoa19SimdAlignmentEngineC1ENS_13AlignmentTypeEaaa@Base 1.1.3
@@ -60,12 +56,13 @@ libspoa.so.1.1.3 libspoa1.1.3 #MINVER#
  _ZN4spoa5GraphD2Ev@Base 1.1.3
  _ZNK4spoa4Node8coverageEv@Base 1.1.3
  _ZNK4spoa4Node9successorERjj@Base 1.1.3
- _ZNK4spoa5Graph14print_graphvizEv@Base 1.1.3
  _ZNK4spoa5Graph16update_alignmentERSt6vectorISt4pairIiiESaIS3_EERKS1_IiSaIiEE@Base 1.1.3
  _ZNK4spoa5Graph22extract_subgraph_nodesERSt6vectorIbSaIbEEjj@Base 1.1.3
  _ZNK4spoa5Graph23is_topologically_sortedEv@Base 1.1.3
  _ZNK4spoa5Graph38initialize_multiple_sequence_alignmentERSt6vectorIjSaIjEE@Base 1.1.3
  _ZNK4spoa5Graph8subgraphEjjRSt6vectorIiSaIiEE@Base 1.1.3
+ _ZNK4spoa5Graph9print_dotERKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE@Base 1.1.5
+ _ZNKSt5ctypeIcE8do_widenEc@Base 1.1.5
  _ZNSt10_HashtableIjjSaIjENSt8__detail9_IdentityESt8equal_toIjESt4hashIjENS1_18_Mod_range_hashingENS1_20_Default_ranged_hashENS1_20_Prime_rehash_policyENS1_17_Hashtable_traitsILb0ELb1ELb1EEEE21_M_insert_unique_nodeEmmPNS1_10_Hash_nodeIjLb0EEEm@Base 1.1.3
  _ZNSt11_Deque_baseIjSaIjEE17_M_initialize_mapEm@Base 1.1.3
  _ZNSt11_Deque_baseIjSaIjEED1Ev@Base 1.1.3
@@ -85,17 +82,12 @@ libspoa.so.1.1.3 libspoa1.1.3 #MINVER#
  _ZNSt6vectorINSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEESaIS5_EE17_M_realloc_insertIJRS5_EEEvN9__gnu_cxx17__normal_iteratorIPS5_S7_EEDpOT_@Base 1.1.3
  _ZNSt6vectorISt10shared_ptrIN4spoa4EdgeEESaIS3_EE17_M_realloc_insertIJRS3_EEEvN9__gnu_cxx17__normal_iteratorIPS3_S5_EEDpOT_@Base 1.1.3
  _ZNSt6vectorISt10unique_ptrIN4spoa4NodeESt14default_deleteIS2_EESaIS5_EE17_M_realloc_insertIJS5_EEEvN9__gnu_cxx17__normal_iteratorIPS5_S7_EEDpOT_@Base 1.1.3
- (optional)_ZNSt6vectorISt4pairIiiESaIS1_EE12emplace_backIJRKjiEEEvDpOT_@Base 1.1.3
- (optional)_ZNSt6vectorISt4pairIiiESaIS1_EE12emplace_backIJiRiEEEvDpOT_@Base 1.1.3
- (optional)_ZNSt6vectorISt4pairIiiESaIS1_EE12emplace_backIJjiEEEvDpOT_@Base 1.1.3
  _ZNSt6vectorISt4pairIiiESaIS1_EE17_M_realloc_insertIJjjEEEvN9__gnu_cxx17__normal_iteratorIPS1_S3_EEDpOT_@Base 1.1.3
  _ZNSt6vectorIbSaIbEE14_M_fill_insertESt13_Bit_iteratormb@Base 1.1.3
  _ZNSt6vectorIiSaIiEE14_M_fill_insertEN9__gnu_cxx17__normal_iteratorIPiS1_EEmRKi@Base 1.1.3
  _ZNSt6vectorIjSaIjEE12emplace_backIJRKiEEEvDpOT_@Base 1.1.3
  _ZNSt6vectorIjSaIjEE12emplace_backIJRiEEEvDpOT_@Base 1.1.3
  _ZNSt6vectorIjSaIjEE12emplace_backIJRjEEEvDpOT_@Base 1.1.3
- (optional)_ZNSt6vectorIjSaIjEE12emplace_backIJiEEEvDpOT_@Base 1.1.3
- (optional)_ZNSt6vectorIjSaIjEE12emplace_backIJjEEEvDpOT_@Base 1.1.3
  _ZNSt6vectorIjSaIjEE14_M_fill_insertEN9__gnu_cxx17__normal_iteratorIPjS1_EEmRKj@Base 1.1.3
  _ZNSt6vectorIjSaIjEE17_M_realloc_insertIJRKjEEEvN9__gnu_cxx17__normal_iteratorIPjS1_EEDpOT_@Base 1.1.3
  _ZTIN4spoa15AlignmentEngineE@Base 1.1.3

debian/patches/fix_soversion.patch

+Note: Patch is not activated to not derive to much from upstream
+Description: Invent SOVERSION different from upstream version
+Author: Andreas Tille <tille@debian.org>
+Last-Update: Mon, 28 Jan 2019 20:04:58 +0100
+Bug-Upstream: https://github.com/rvaser/spoa/issues/14
+
+--- a/CMakeLists.txt
++++ b/CMakeLists.txt
+@@ -41,7 +41,7 @@ target_include_directories(spoa_static P
+ set_target_properties(spoa
+     PROPERTIES
+     VERSION ${spoa_VERSION}
+-    SOVERSION ${spoa_VERSION})
++    SOVERSION 2)
+ 
+ install(TARGETS spoa DESTINATION ${CMAKE_INSTALL_LIBDIR})
+ install(TARGETS spoa_static DESTINATION ${CMAKE_INSTALL_LIBDIR})

debian/patches/series

 use_debian_packaged_libs.patch
 shared_and_static.patch
 no_march.patch
+# fix_soversion.patch

include/spoa/graph.hpp

@@ -78,7 +78,7 @@ public:
     void update_alignment(Alignment& alignment,
         const std::vector<int32_t>& subgraph_to_graph_mapping) const;
 
-    void print_graphviz() const;
+    void print_dot(const std::string& path) const;
 
     friend std::unique_ptr<Graph> createGraph();
 private:

src/graph.cpp

@@ -8,6 +8,7 @@
 #include <assert.h>
 #include <algorithm>
 #include <stack>
+#include <fstream>
 
 #include "spoa/graph.hpp"
 
@@ -679,7 +680,13 @@ void Graph::update_alignment(Alignment& alignment,
     }
 }
 
-void Graph::print_graphviz() const {
+void Graph::print_dot(const std::string& path) const {
+
+    if (path.empty()) {
+        return;
+    }
+
+    std::ofstream out(path);
 
     std::vector<int32_t> in_consensus(nodes_.size(), -1);
     int32_t rank = 0;
@@ -687,30 +694,34 @@ void Graph::print_graphviz() const {
         in_consensus[id] = rank++;
     }
 
-    printf("digraph %u {\n", num_sequences_);
-    printf("    graph [rankdir=LR]\n");
+    out << "digraph " << num_sequences_ << " {" << std::endl;
+    out << "    graph [rankdir=LR]" << std::endl;
     for (uint32_t i = 0; i < nodes_.size(); ++i) {
-        printf("    %u [label = \"%u - %c\"", i, i, decoder_[nodes_[i]->code_]);
+        out << "    " << i << " [label = \"" << i << " - ";
+        out << static_cast<char>(decoder_[nodes_[i]->code_]) << "\"";
         if (in_consensus[i] != -1) {
-            printf(", style=filled, fillcolor=goldenrod1");
+            out << ", style=filled, fillcolor=goldenrod1";
         }
-        printf("]\n");
+        out << "]" << std::endl;
 
         for (const auto& edge: nodes_[i]->out_edges_) {
-            printf("    %u -> %u [label = \"%lu\"", i, edge->end_node_id_,
-                edge->total_weight_);
+            out << "    " << i << " -> " << edge->end_node_id_;
+            out << " [label = \"" << edge->total_weight_ << "\"";
             if (in_consensus[i] + 1 == in_consensus[edge->end_node_id_]) {
-                printf(", color=goldenrod1");
+                out << ", color=goldenrod1";
             }
-            printf("]\n");
+            out << "]" << std::endl;
         }
         for (const auto& aid: nodes_[i]->aligned_nodes_ids_) {
             if (aid > i) {
-                printf("    %u -> %u [style = dotted, arrowhead = none]\n", i, aid);
+                out << "    " << i << " -> " << aid;
+                out << " [style = dotted, arrowhead = none]" << std::endl;
             }
         }
     }
-    printf("}\n");
+    out << "}" << std::endl;
+
+    out.close();
 }
 
 }

src/main.cpp

@@ -7,7 +7,7 @@
 #include "spoa/spoa.hpp"
 #include "bioparser/bioparser.hpp"
 
-static const char* version = "v1.1.3";
+static const char* version = "v1.1.5";
 
 static struct option options[] = {
     {"match", required_argument, 0, 'm'},
@@ -15,6 +15,7 @@ static struct option options[] = {
     {"gap", required_argument, 0, 'g'},
     {"algorithm", required_argument, 0, 'l'},
     {"result", required_argument, 0, 'r'},
+    {"dot", required_argument, 0, 'd'},
     {"version", no_argument, 0, 'v'},
     {"help", no_argument, 0, 'h'},
     {0, 0, 0, 0}
@@ -31,8 +32,10 @@ int main(int argc, char** argv) {
     uint8_t algorithm = 0;
     uint8_t result = 0;
 
+    std::string dot_path = "";
+
     char opt;
-    while ((opt = getopt_long(argc, argv, "m:x:g:l:r:h", options, nullptr)) != -1) {
+    while ((opt = getopt_long(argc, argv, "m:x:g:l:r:d:h", options, nullptr)) != -1) {
         switch (opt) {
             case 'm':
                 match = atoi(optarg);
@@ -49,6 +52,9 @@ int main(int argc, char** argv) {
             case 'r':
                 result = atoi(optarg);
                 break;
+            case 'd':
+                dot_path = optarg;
+                break;
             case 'v':
                 printf("%s\n", version);
                 exit(0);
@@ -66,22 +72,30 @@ int main(int argc, char** argv) {
         exit(1);
     }
 
-    std::string input_path = argv[optind];
-    auto extension = input_path.substr(std::min(input_path.rfind('.'),
-        input_path.size()));
+    std::string sequences_path = argv[optind];
+
+    auto is_suffix = [](const std::string& src, const std::string& suffix) -> bool {
+        if (src.size() < suffix.size()) {
+            return false;
+        }
+        return src.compare(src.size() - suffix.size(), suffix.size(), suffix) == 0;
+    };
 
     std::unique_ptr<bioparser::Parser<spoa::Sequence>> sparser = nullptr;
 
-    if (extension == ".fasta" || extension == ".fa") {
+    if (is_suffix(sequences_path, ".fasta") || is_suffix(sequences_path, ".fa") ||
+        is_suffix(sequences_path, ".fasta.gz") || is_suffix(sequences_path, ".fa.gz")) {
         sparser = bioparser::createParser<bioparser::FastaParser, spoa::Sequence>(
-            input_path);
-    } else if (extension == ".fastq" || extension == ".fq") {
+            sequences_path);
+    } else if (is_suffix(sequences_path, ".fastq") || is_suffix(sequences_path, ".fq") ||
+        is_suffix(sequences_path, ".fastq.gz") || is_suffix(sequences_path, ".fq.gz")) {
         sparser = bioparser::createParser<bioparser::FastqParser, spoa::Sequence>(
-            input_path);
+            sequences_path);
     } else {
         fprintf(stderr, "[spoa::] error: "
             "file %s has unsupported format extension (valid extensions: "
-            ".fasta, .fa, .fastq, .fq)!\n", input_path.c_str());
+            ".fasta, .fasta.gz, .fa, .fa.gz, .fastq, .fastq.gz, .fq, .fq.gz)!\n",
+            sequences_path.c_str());
         exit(1);
     }
 
@@ -120,6 +134,8 @@ int main(int argc, char** argv) {
         }
     }
 
+    graph->print_dot(dot_path);
+
     return 0;
 }
 
@@ -128,7 +144,8 @@ void help() {
         "usage: spoa [options ...] <sequences>\n"
         "\n"
         "    <sequences>\n"
-        "        input file in FASTA/FASTQ format containing sequences\n"
+        "        input file in FASTA/FASTQ format (can be compressed with gzip)\n"
+        "        containing sequences\n"
         "\n"
         "    options:\n"
         "        -m, --match <int>\n"
@@ -152,6 +169,8 @@ void help() {
         "                0 - consensus\n"
         "                1 - multiple sequence alignment\n"
         "                2 - 0 & 1\n"
+        "        -d, --dot <file>\n"
+        "            output file for the final POA graph in DOT format\n"
         "        --version\n"
         "            prints the version number\n"
         "        -h, --help\n"