Fix Table 3 reference

This build is based on
7b5b0e6.

This commit was created by the following Travis CI build and job:
https://travis-ci.org/dhimmel/rephetio-manuscript/builds/282958938
https://travis-ci.org/dhimmel/rephetio-manuscript/jobs/282958939

[ci skip]

The full commit message that triggered this build is copied below:

Fix Table 3 reference

manuscript.html

@@ -13,7 +13,7 @@
   <meta name="author" content="Ari Green">
   <meta name="author" content="Pouya Khankhanian">
   <meta name="author" content="Sergio E. Baranzini">
-  <meta name="dcterms.date" content="2017-09-22">
+  <meta name="dcterms.date" content="2017-10-03">
   <meta name="keywords" content="Rephetio, Hetionet, repurposing, hetnets, systems pharmacology, networks, machine learning, data science">
   <title>Systematic integration of biomedical knowledge prioritizes drugs for repurposing</title>
   <style type="text/css">code{white-space: pre;}</style>
@@ -30,7 +30,7 @@
 <header>
 <h1 class="title">Systematic integration of biomedical knowledge prioritizes drugs for repurposing</h1>
 </header>
-<p><small><em> This manuscript was automatically generated from <a href="https://github.com/dhimmel/rephetio-manuscript/tree/5358af54ea83dcabb3eed3805ba87455b00f5bf8">dhimmel/rephetio-manuscript@5358af5</a> on September 22, 2017. </em></small></p>
+<p><small><em> This manuscript was automatically generated from <a href="https://github.com/dhimmel/rephetio-manuscript/tree/7b5b0e67fb9ab159c84c230dd5c07001941d82af">dhimmel/rephetio-manuscript@7b5b0e6</a> on October 3, 2017. </em></small></p>
 <h2 id="authors">Authors</h2>
 <ul>
 <li><p><strong>Daniel S. Himmelstein</strong><br> <img src="images/orcid.svg" alt="ORCID icon" height="13" /> <a href="https://orcid.org/0000-0002-3012-7446">0000-0002-3012-7446</a> · <img src="images/github.svg" alt="GitHub icon" height="13" /> <a href="https://github.com/dhimmel">dhimmel</a> · <img src="images/twitter.svg" alt="Twitter icon" height="13" /> <a href="https://twitter.com/dhimmel">dhimmel</a><br> <small> Program in Biological &amp; Medical Informatics, University of California, San Francisco; Department of Systems Pharmacology &amp; Translational Therapeutics, University of Pennsylvania </small></p></li>
@@ -358,7 +358,7 @@ <h3 id="systematic-mechanisms-of-efficacy">Systematic mechanisms of efficacy</h3
 <figure>
 <img src="https://github.com/dhimmel/learn/raw/ef5f7a6b76b6a01499d65b95e3d7ca93ac5aba57/prediction/figure/features.png" alt="Figure 2: Performance by type and model coefficients. A) The performance of the DWPCs for 1,206 metapaths, organized by their composing metaedges. The larger dots represent metapaths that were significantly affected by permutation (false discovery rate &lt; 5%). Metaedges are ordered by their best performing metapath. Since a metapath’s performance is limited by its least informative metaedge, the best performing metapath for a metaedge provides a lower bound on the pharmacologic utility of a given domain of information. B) Barplot of the model coefficients. Features were standardized prior to model fitting to make the coefficients comparable [39]." id="fig:features" /><figcaption>Figure 2: <strong>Performance by type and model coefficients.</strong> A) The performance of the DWPCs for <a href="http://het.io/repurpose/metapaths.html" title="Project Rephetio Metapath Browser">1,206 metapaths</a>, organized by their composing metaedges. The larger dots represent metapaths that were significantly affected by permutation (false discovery rate &lt; 5%). Metaedges are ordered by their best performing metapath. Since a metapath’s performance is limited by its least informative metaedge, the best performing metapath for a metaedge provides a lower bound on the pharmacologic utility of a given domain of information. B) Barplot of the model coefficients. Features were standardized prior to model fitting to make the coefficients comparable <span class="citation" data-cites="uHJFShcv">[<a href="#ref-uHJFShcv">39</a>]</span>.</figcaption>
 </figure>
-<p>Overall, 709 of the 1,206 metapaths exhibited a statistically significant Δ AUROC at a false discovery rate cutoff of 5%. These 709 metapaths included all 24 metaedges, suggesting that each type of relationship we integrated provided at least some therapeutic utility. However, not all metaedges were equally present in significant metapaths: 259 significant metapaths included a <em>Compound–binds–Gene</em> metaedge, whereas only 4 included a <em>Gene–participates–Cellular Component</em> metaedge. <a href="#metapath_table">Table 3</a> lists the predictiveness of several metapaths of interest. Refer to the <a href="#discussion">Discussion</a> for our interpretation of these findings.</p>
+<p>Overall, 709 of the 1,206 metapaths exhibited a statistically significant Δ AUROC at a false discovery rate cutoff of 5%. These 709 metapaths included all 24 metaedges, suggesting that each type of relationship we integrated provided at least some therapeutic utility. However, not all metaedges were equally present in significant metapaths: 259 significant metapaths included a <em>Compound–binds–Gene</em> metaedge, whereas only 4 included a <em>Gene–participates–Cellular Component</em> metaedge. Table <a href="#tbl:metapaths">3</a> lists the predictiveness of several metapaths of interest. Refer to the <a href="#discussion">Discussion</a> for our interpretation of these findings.</p>
 <a name="tbl:metapaths"></a>
 <table>
 <caption>Table 3: <strong>The predictiveness of select metapaths.</strong> A small selection of interesting or influential metapaths is provided (<a href="http://het.io/repurpose/metapaths.html" title="Project Rephetio Metapath Browser">complete table online</a>). Len. refers to number of metaedges composing the metapath. Δ AUROC and −log10(<em>p</em>) assess the performance of a metapath’s DWPC in discriminating treatments from non-treatments (in the all-features stage as <a href="#machine-learning-approach">described in Methods</a>). <em>p</em> assesses whether permutation affected AUROC. For reference, <em>p</em> = 0.05 corresponds to −log10(<em>p</em>) = 1.30. Note that several metapaths shown here provided little evidence that Δ AUROC ≠ 0 underscoring their poor ability to predict whether a compound treated a disease. Coef. reports a metapath’s logistic regression coefficient as seen in Figure <a href="#fig:features">2</a>B. Metapaths removed in feature selection have missing coefficients whereas metapaths given zero-weight by the elastic net have coef. = 0.0. </caption>

manuscript.md

@@ -9,7 +9,7 @@ author-meta:
 - Ari Green
 - "Pouya\_Khankhanian"
 - "Sergio\_E.\_Baranzini"
-date-meta: '2017-09-22'
+date-meta: '2017-10-03'
 keywords:
 - Rephetio
 - Hetionet
@@ -27,8 +27,8 @@ title: Systematic integration of biomedical knowledge prioritizes drugs for repu
 
 <small><em>
 This manuscript was automatically generated
-from [dhimmel/rephetio-manuscript@5358af5](https://github.com/dhimmel/rephetio-manuscript/tree/5358af54ea83dcabb3eed3805ba87455b00f5bf8)
-on September 22, 2017.
+from [dhimmel/rephetio-manuscript@7b5b0e6](https://github.com/dhimmel/rephetio-manuscript/tree/7b5b0e67fb9ab159c84c230dd5c07001941d82af)
+on October  3, 2017.
 </em></small>
 
 ## Authors
@@ -323,7 +323,7 @@ Features were standardized prior to model fitting to make the coefficients compa
 Overall, 709 of the 1,206 metapaths exhibited a statistically significant Δ AUROC at a false discovery rate cutoff of 5%.
 These 709 metapaths included all 24 metaedges, suggesting that each type of relationship we integrated provided at least some therapeutic utility.
 However, not all metaedges were equally present in significant metapaths: 259 significant metapaths included a _Compound–binds–Gene_ metaedge, whereas only 4 included a _Gene–participates–Cellular Component_ metaedge.
-[Table 3](#metapath_table) lists the predictiveness of several metapaths of interest.
+Table @tbl:metapaths lists the predictiveness of several metapaths of interest.
 Refer to the [Discussion](#discussion) for our interpretation of these findings.
 
 Abbrev. | Len. | Δ AUROC | −log₁₀(*p*) | Coef. | Metapath

variables.json

@@ -1,5 +1,5 @@
 {
-  "date": "September 22, 2017",
+  "date": "October  3, 2017",
   "authors": [
     {
       "github": "dhimmel",
@@ -61,7 +61,7 @@
   ],
   "ci_source": {
     "repo_slug": "dhimmel/rephetio-manuscript",
-    "commit": "5358af54ea83dcabb3eed3805ba87455b00f5bf8"
+    "commit": "7b5b0e67fb9ab159c84c230dd5c07001941d82af"
   },
   "manuscript_stats": {
     "reference_counts": {