Merge pull request #313 from OpenTreeOfLife/figures-to-html-file

add figures to the HTML file

Author: kcranston

doc/method/Makefile

@@ -5,13 +5,10 @@
 REF_HOME=../..
 JYTHON=$(REF_HOME)/bin/jython
 
-all: draft.html supplementary/taxonomy_stats.csv
+all: out/draft.html supplementary/taxonomy_stats.csv
 
 # use --html4tags flag?
 
-draft.html: out/draft.html
-	ln -sf $< $@
-
 out/draft.html: out/draft.md lib/Markdown.pl
 	@mkdir -p out
 	bin/markdown $< >$@.new

doc/method/introduction.md

@@ -41,6 +41,8 @@ The Open Tree of Life project consists of a set of tools for
 The outcome is one or more summary trees combining phylogenetic and
 taxonomic knowledge. Figure 1 illustrates an overview of the process of combining phylogeny and taxonomy, while the details are described in a separate publication [ref Redelings & Holder 2017].
 
+<img src="../figures/fig1.jpeg" width="512" height="384"/>
+
 Although Open Tree is primarily a phylogenetics effort, it requires a
 reference taxonomy that can support each of these functions.
 

doc/method/method-details.md

@@ -379,6 +379,10 @@ Figure 3 illustrates each of these six cases.
    incompatible with the workspace if the nodes that its aligned
    children align with do not all have the same parent.
 
+<img src="../figures/fig3a.jpeg" width="512" height="384"/>
+<img src="../figures/fig3b.jpeg" width="512" height="384"/>
+
+
 ## Final patches
 
 After all source taxonomies are aligned and merged, we apply general

doc/method/method-intro.md

@@ -100,7 +100,7 @@ hundreds of thousands of similar simple grafting events.
 
 The other merge method is an _insertion_, where the unaligned
 node has descendants that are in S. This always
-occurs when S' has greater resolution than S. For example, see Figure 2b, where GBIF provides greater resolution than NCBI.
+occurs when S' has greater resolution than S. For example, see Figure 2b, where WoRMS provides greater resolution than NCBI.
 
 The vast majority of alignment and merge situations are simple, similar to the
 examples shown in Figure 2. However, even a small fraction of special cases can add up to
@@ -113,6 +113,8 @@ by trial and error - finding cases that fail and then adding / modifying
 heuristics to address the underlying cause, or adding an _ad hoc_ adjustment for
 cases that are rare or overly complex.
 
+<img src="../figures/fig2.jpeg" width="512" height="384"/>
+
 The following sections describe the source taxonomies, and then detail the
 taxonomy combination method.
 

doc/method/results.md

@@ -15,10 +15,12 @@ The methods and results presented here are for version 3.0 of the Open Tree Taxo
 
 ## Results of assembly procedure
 
-As OTT is assembled, the alignment procedure processes every source node, either choosing an alignment target for it in the workspace based on the results of the heuristics, or leaving it unaligned. Figure 4 illustrates the results of the alignment phase. The presence of a single candidate node does not automatically align the two nodes - we still apply the heuristics to ensure a match (and occasionally reject the single candidate).  
+As OTT is assembled, the alignment procedure processes every source node, either choosing an alignment target for it in the workspace based on the results of the heuristics, or leaving it unaligned. Figure 4 illustrates the action of the alignment phase. The presence of a single candidate node does not automatically align the two nodes - we still apply the heuristics to ensure a match (and occasionally reject the single candidate).  
 
 We counted the frequency of success for each heuristic, i.e. the number of times that a particular heuristic was the one that accepted the winning candidate from among two or more candidates. Table X shows these results. Separation (do not align taxa in disjoint separation taxa; used first), Lineage (align taxa with shared lineage; used midway through) and Same-name-string (prefer candidates who primary name-string matches; used last) were by far the most frequent.
 
+<img src="../figures/fig4.jpeg" width="512" height="384"/>
+
 ## Results of merge procedure
 
 After assembly, the next step in the method is to merge the unaligned nodes into the workspace taxonomy. Of the 3,774,509 unaligned nodes, the vast majority (92%) are inserted into the workspace. Grafting accounts for 7% of the merge operations, and less than 1% are either absorptions or remain unmerged due to ambiguities.