Fix more relative links

blab · Jan 11, 2017 · caac31f · caac31f
1 parent 25d1faf
commit caac31f
Show file tree

Hide file tree

Showing 2 changed files with 2 additions and 2 deletions.
diff --git a/recombination/r_squared/README.md b/recombination/r_squared/README.md
@@ -1,3 +1,3 @@
-![](https://github.com/blab/siv-cst/blob/master/figures/png/FigS1.png)
+![](../../figures/png/FigS1.png)
 ##Extensive divergence makes sitewise measures of genetic linkage ineffective  
 For pairs of biallelic sites (ignoring rare variants), R^2 was used to estimate how strongly the allele in one site predicts the allele in the second site, with values of 0 indicating no linkage and 1 indicating perfect linkage. The mean value of R^2 was 0.044, indicating very low levels of linkage overall.
diff --git a/recombination/rspr/README.md b/recombination/rspr/README.md
@@ -1,4 +1,4 @@
-![](https://github.com/blab/siv-cst/blob/master/figures/png/FigS2.png)
+![](../../figures/png/FigS2.png)
 
 ## Figure S2: No evidence of linkage between nonadjacent segments of the SIV genome.
 The alignment used for GARD analyses (LANL compendium with HIV overrepresentation reduced) was split along the breakpoints identified by GARD to yield the 12 genomic segments, and a maximum likelihood tree was constructed for each. The number of steps required to turn one tree topology into another was assessed for each pair of trees with the Rooted Subtree-Prune-and-Regraft (rSPR) package. Segment pairs with similar topologies have lower scores than segments with less similar topologies.