From 18f1035d3e1e3cfe92842e40380c7f70061a1df4 Mon Sep 17 00:00:00 2001
From: yeredh <yered.h@gmail.com>
Date: Sat, 18 Apr 2015 23:01:04 -0400
Subject: [PATCH] minor typos
---
course4/normalization.Rmd | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/course4/normalization.Rmd b/course4/normalization.Rmd
index 0013cd85..4a6a8925 100644
--- a/course4/normalization.Rmd
+++ b/course4/normalization.Rmd
@@ -226,7 +226,7 @@ mypar(1,1)
maplot(log2(y1),log2(y2),ylim=c(-1,1),curve.add=FALSE)
```
-For this type of data, the variance depends on the mean. We seek a transfromation that stabilizies the variance of the estimates of $\theta$ after we subctract the additive background estimate and divide by the estimate of the gain.
+For this type of data, the variance depends on the mean. We seek a transformation that stabilizies the variance of the estimates of $\theta$ after we subctract the additive background estimate and divide by the estimate of the gain.
```{r}
ny1=(y1-b1)/A1
@@ -247,7 +247,7 @@ $$
\text{arsinh}(y) = \log\left(y + \sqrt{y^2+1} \right)
$$
-The `vsn` library implements this apprach. It estimates $\beta$ and $A$ by assuming that most genes don't change, i.e. $\theta$ does not depend on $i$.
+The `vsn` library implements this approach. It estimates $\beta$ and $A$ by assuming that most genes don't change, i.e. $\theta$ does not depend on $i$.
```{r}
library(vsn)