Merge branch 'pin-requests' into read-lr-giraffe

Makefile

@@ -215,6 +215,9 @@ ifeq ($(shell uname -s),Darwin)
     # We don't actually do any static linking on Mac, so we leave this empty.
     START_STATIC =
     END_STATIC =
+
+    # We need to use special flags to let us rename libraries
+    LD_RENAMEABLE_FLAGS = -Wl,-headerpad -Wl,-headerpad_max_install_names
 else
     # We are not running on OS X
     $(info OS is Linux)
@@ -252,7 +255,8 @@ else
     # Note that END_STATIC is only safe to use in a mostly-dynamic build, and has to appear or we will try to statically link secret trailing libraries.
     END_STATIC = -Wl,-Bdynamic
 
-
+    # We don't need any flags because we don't need to rename libraries with install_name_tool
+    LD_RENAMEABLE_FLAGS =
 endif
 
 # Set the C++ standard we are using
@@ -656,7 +660,7 @@ ifeq ($(shell uname -s),Darwin)
 endif
 
 $(LIB_DIR)/libdeflate.a: $(LIBDEFLATE_DIR)/*.h $(LIBDEFLATE_DIR)/lib/*.h $(LIBDEFLATE_DIR)/lib/*/*.h $(LIBDEFLATE_DIR)/lib/*.c $(LIBDEFLATE_DIR)/lib/*/*.c
-	+. ./source_me.sh && cd $(LIBDEFLATE_DIR) && V=1 $(MAKE) $(FILTER) && cp libdeflate.a $(CWD)/$(LIB_DIR) && cp libdeflate.h $(CWD)/$(INC_DIR)
+	+. ./source_me.sh && cd $(LIBDEFLATE_DIR) && V=1 LDFLAGS="$(LDFLAGS) $(LD_RENAMEABLE_FLAGS)" $(MAKE) $(FILTER) && cp libdeflate.a $(CWD)/$(LIB_DIR) && cp libdeflate.h $(CWD)/$(INC_DIR)
 
 # We build htslib after libdeflate so it can use libdeflate.
 # We need to do some wizardry to get it to pick up the right build and target system types on modern autotools.

ontology/vg.html

@@ -688,7 +688,7 @@ <h2>
 <td><a href="http://www.w3.org/2000/01/rdf-schema#comment">
 					rdfs:comment</a></td>
 <td>
-						"A step along a path in the variant graph. A series of steps along a path represent an assembled sequence that was originally inserted into the the variant graph. A step points to a :Node or the reverse complement of a node and has a rank (step number)."
+						"A step along a path in the variant graph. A series of steps along a path represent an assembled sequence that was originally inserted into the variant graph. A step points to a :Node or the reverse complement of a node and has a rank (step number)."
 						<sup><a href="http://www.w3.org/1999/02/22-rdf-syntax-ns#http://www.w3.org/2001/XMLSchema#string">
 				xsd:string</a></sup>
 </td>

ontology/vg.ttl

@@ -31,7 +31,7 @@
 .
 :Step
   rdf:type owl:Class ;
-  rdfs:comment "A step along a path in the variant graph. A series of steps along a path represent an assembled sequence that was originally inserted into the the variant graph. A step points to a :Node or the reverse complement of a node and has a rank (step number)."^^xsd:string ;
+  rdfs:comment "A step along a path in the variant graph. A series of steps along a path represent an assembled sequence that was originally inserted into the variant graph. A step points to a :Node or the reverse complement of a node and has a rank (step number)."^^xsd:string ;
   rdfs:label "Step"^^xsd:string ;
   rdfs:subClassOf owl:Thing ;
 .

src/algorithms/k_widest_paths.cpp

@@ -246,7 +246,7 @@ vector<pair<double, vector<handle_t>>> yens_k_widest_paths(const HandleGraph* g,
                 forgotten_nodes.insert(g->flip(prev_path[j]));
             }
 
-            // find our path from the the spur_node to the sink
+            // find our path from the spur_node to the sink
             pair<double, vector<handle_t>> spur_path_v = widest_dijkstra(g, spur_node, sink, node_weight_callback, edge_weight_callback,
                                                                          [&](handle_t h) {return forgotten_nodes.count(h);},
                                                                          [&](edge_t e) {return forgotten_edges.count(e);},

src/cactus.cpp

@@ -148,7 +148,7 @@ void getReachableBridges(stCactusEdgeEnd *edgeEnd1, stList *bridgeEnds) {
 }
 
 /**
- * Finds an arbitrary pair of telomeres in a Cactus graph, which are are either
+ * Finds an arbitrary pair of telomeres in a Cactus graph, which are either
  * a pair of bridge edge ends or a pair of chain edge ends, oriented such that
  * they form a pair of boundaries.
  *

src/cluster.cpp

@@ -735,7 +735,7 @@ void MEMClusterer::HitGraph::prune_low_scoring_edges(vector<vector<size_t>>& com
         }
         cerr << strm.str();
 #endif
-        // the the original component
+        // the original component
         components[component_idx] = move(new_components[0]);
         // add the remaining to the end
         for (size_t i = 1; i < new_components.size(); i++) {

src/gbwt_extender.cpp

@@ -2284,7 +2284,14 @@ WFAAlignment WFAExtender::connect(std::string sequence, pos_t from, pos_t to) co
 }
 
 WFAAlignment WFAExtender::suffix(const std::string& sequence, pos_t from) const {
-    return this->connect(sequence, from, pos_t(0, false, 0));
+    WFAAlignment result = this->connect(sequence, from, pos_t(0, false, 0));
+
+    if (!result.edits.empty() && result.length == sequence.length() && (result.edits.back().first == WFAAlignment::match || result.edits.back().first == WFAAlignment::mismatch)) {
+        // The alignment used all of the sequence and has a match/mismatch at the appropriate end
+        result.score += this->aligner->full_length_bonus; 
+    }
+
+    return result;
 }
 
 WFAAlignment WFAExtender::prefix(const std::string& sequence, pos_t to) const {
@@ -2297,6 +2304,10 @@ WFAAlignment WFAExtender::prefix(const std::string& sequence, pos_t to) const {
     WFAAlignment result = this->connect(reverse_complement(sequence), to, pos_t(0, false, 0));
     result.flip(*(this->graph), sequence);
 
+    if (!result.edits.empty() && result.length == sequence.length() && (result.edits.front().first == WFAAlignment::match || result.edits.front().first == WFAAlignment::mismatch)) {
+        result.score += this->aligner->full_length_bonus; 
+    }
+
     return result;
 }
 

src/gbwt_extender.hpp

@@ -412,9 +412,11 @@ class WFAExtender {
      * entire sequence with an acceptable score, returns the highest-scoring
      * partial alignment, which may be empty.
      *
+     * Applies the full-length bonus if the result ends with a match or mismatch.
+     * TODO: Use the full-length bonus to determine the optimal alignment.
+     *
      * NOTE: This creates a suffix of the full alignment by aligning a
      * prefix of the sequence.
-     * TODO: Should we use full-length bonuses?
      */
     WFAAlignment suffix(const std::string& sequence, pos_t from) const;
 
@@ -424,9 +426,11 @@ class WFAExtender {
      * sequence with an acceptable score, returns the highest-scoring partial
      * alignment, which may be empty.
      *
+     * Applies the full-length bonus if the result begins with a match or mismatch.
+     * TODO: Use the full-length bonus to determine the optimal alignment.
+     *
      * NOTE: This creates a prefix of the full alignment by aligning a suffix
      * of the sequence.
-     * TODO: Should we use full-length bonuses?
      */
     WFAAlignment prefix(const std::string& sequence, pos_t to) const;
 

src/graph_caller.cpp

@@ -2302,7 +2302,7 @@ string NestedFlowCaller::flatten_alt_allele(const string& nested_allele, int all
                     // todo: passing in a single ploidy simplisitic, would need to derive from the calls when
                     // reucrising
                     // in practice, the results will nearly the same but still needs fixing
-                    // we try to get the the allele from the genotype if possible, but fallback on the fallback_allele
+                    // we try to get the allele from the genotype if possible, but fallback on the fallback_allele
                     int trav_allele = fallback_allele >= 0 ? fallback_allele : record.genotype_by_ploidy[ploidy-1].first[allele];
                     const SnarlTraversal& traversal = record.travs[trav_allele];
                     string nested_snarl_allele = trav_string(graph, traversal);

src/index_registry.cpp

@@ -780,7 +780,7 @@ IndexRegistry VGIndexes::get_vg_index_registry() {
             }
         }
 
-        // we'll partition sequences that have the same samples (chunking the the VCFs
+        // we'll partition sequences that have the same samples (chunking the VCFs
         // ultimately requires that we do this)
         map<set<string>, vector<string>> sample_set_contigs;
         for (int i = 0; i < vcf_samples.size(); ++i) {

src/splicing.cpp

@@ -1633,7 +1633,7 @@ multipath_alignment_t&& fuse_spliced_alignments(const Alignment& alignment,
                 }
             }
             else if (!have_right_linker) {
-                // the splice segment on the right side is empty, make connection the the left side
+                // the splice segment on the right side is empty, make connection the left side
                 auto connection = left_mp_aln.mutable_subpath(right_subpaths_begin - 1)->add_connection();
                 connection->set_next(left_mp_aln.subpath_size());
                 connection->set_score(splice_score);

src/subcommand/chunk_main.cpp

@@ -715,11 +715,9 @@ int main_chunk(int argc, char** argv) {
         unique_ptr<MutablePathMutableHandleGraph> subgraph;
         map<string, int> trace_thread_frequencies;
         if (!component_ids.empty()) {
-            cout << "here?" << endl;
             subgraph = vg::io::new_output_graph<MutablePathMutableHandleGraph>(output_format);
             chunker.extract_component(component_ids[i], *subgraph, false);
             output_regions[i] = region;
-            cout << "nope" << endl;
         }
         else if (id_range == false) {
             subgraph = vg::io::new_output_graph<MutablePathMutableHandleGraph>(output_format);

src/transcriptome.hpp

@@ -281,7 +281,7 @@ class Transcriptome {
         /// Parse gtf/gff3 attribute value.
         string parse_attribute_value(const string & attribute, const string & name) const;
 
-        /// Returns the the mean node length of the graph
+        /// Returns the mean node length of the graph
         float mean_node_length() const;
 
         /// Adds the exon coordinates to a transcript.