altman plot path update

pages/altman_bland_analysis_of_methods.md

@@ -23,7 +23,7 @@ For more similar papers see the series of BMJ statistical notes by Altman & Blan
 ## The fianl comparison results
 ### Scores per sample and Bland-Altman plot
 
-<img src="{{ site.baseurl }}{% link assets/images/altman_bland_analysis_of_methods.png %}" width="100%">
+<img src="{{ "assets/images/altman_bland_analysis_of_methods.png" | relative_url }}" width="100%">
 **Figure 1.** Scores per sample and Bland-Altman plot as produced by the provided R code.
 
 ### Correlation test result and repeatability coefficients

pages/altman_bland_correlation.md

@@ -95,7 +95,7 @@ Table II. Analysis of variance for the data in table I (as shown in Altman & Bla
 Incorrectly calculating the correlation coefficient 
 by ignoring the fact that we have 47 observations on only 8 subjects, would produce -0.07, P=0.7. 
 
-<img src="{{ site.baseurl }}{% link assets/images/altman_bland_repeated_observations.png %}" width="100%">
+<img src="{{ "assets/images/altman_bland_repeated_observations.png" | relative_url }}" width="100%">
 **Figure 1.** Recreation of "(Y) pH against (X) PaCO2 for eight subjects, with parallel lines fitted for each subject" as used in 
 bland1995statistics.
 Interestingly, replotting this data shows that their figure was not fully accurate (forgivable before the days of Rstudio in 1995, and not important for this example).

pages/altman_bland_roc_curve.md

@@ -26,7 +26,7 @@ Altman and Bland mention that this method was developed in the 1950s for evaluat
 An aside for history buffs, from [wikipedia](https://en.wikipedia.org/wiki/Receiver_operating_characteristic):
 > The ROC curve was first used during World War II for the analysis of radar signals before it was employed in signal detection theory.[56] Following the attack on Pearl Harbor in 1941, the United States army began new research to increase the prediction of correctly detected Japanese aircraft from their radar signals. For these purposes they measured the ability of a radar receiver operator to make these important distinctions, which was called the Receiver Operating Characteristic.[57]
 
-<img src="{{ site.baseurl }}{% link assets/images/receiver_operating_characteristic_plots.png %}" width="100%">
+<img src="{{ "assets/images/receiver_operating_characteristic_plots.png" | relative_url }}" width="100%">
 
 The example shown in Figure 1 uses graft versus host disease, with an index measurement whose definition is not important.
 The _Yes_ indicate _cases_ and _No_ indicate _controls_ in our terminology, respectively. 
@@ -81,7 +81,7 @@ $$\frac{\text{No. cases above cutoff} + \text{No. controls below cutoff }}{ \tex
 ## Example ROC cuvre
 To implement this method, I include here an example in R code.
 
-<img src="{{ site.baseurl }}{% link assets/images/receiver_operating_characteristic_curve.png %}" width="100%">
+<img src="{{ "assets/images/receiver_operating_characteristic_curve.png" | relative_url }}" width="100%">
 
 ``` R
 # Modified example from https://stackoverflow.com/questions/31138751/roc-curve-from-training-data-in-caret

pages/lawlessgenomics_to_docs.sh

@@ -11,6 +11,8 @@ do
 
 # Use sed to transform the file content as required and save it to the new location
 sed -e 's/layout: topic/layout: default/' \
+-e 's/{{ site.baseurl }}{% link \(.*\) %}"/{{ "assets\/\1"/' \
+-e 's/" width="100%">/" | relative_url }}" width="100%">/' \
 -e '1a\
 nav_order: 5' \
 -e '1a\