lecture slides

ocbe-uio · Jan 16, 2024 · 0932d52 · 0932d52
1 parent c296a27
commit 0932d52
Show file tree

Hide file tree

Showing 13 changed files with 705 additions and 600 deletions.
diff --git a/dev/lablecture_day3_pca.key b/dev/lablecture_day3_pca.key
diff --git a/docs/lab/lab_day3_pca.html b/docs/lab/lab_day3_pca.html
diff --git a/docs/lab/lab_day4_clustering_files/figure-html/clust-nci60-cutree-1.png b/docs/lab/lab_day4_clustering_files/figure-html/clust-nci60-cutree-1.png
diff --git a/docs/lab/lab_day4_clustering_files/figure-html/pca-nci-keeppc-1.png b/docs/lab/lab_day4_clustering_files/figure-html/pca-nci-keeppc-1.png
diff --git a/docs/lab/lab_day4_clustering_files/figure-html/pca-nci-plot2-1.png b/docs/lab/lab_day4_clustering_files/figure-html/pca-nci-plot2-1.png
diff --git a/docs/lab/lab_day4_clustering_files/figure-html/pca-nci-plotpc-1.png b/docs/lab/lab_day4_clustering_files/figure-html/pca-nci-plotpc-1.png
diff --git a/docs/lab/overview.html b/docs/lab/overview.html
@@ -233,9 +233,15 @@ <h2 class="anchored" data-anchor-id="lab-notes-and-r-scripts">Lab notes and R sc
 <td style="text-align: center;">Day 3</td>
 <td style="text-align: center;"><a href="../lab/lab_day3_pca.html">Day 3: Principal Component Analysis</a></td>
 <td style="text-align: center;"><a href="https://github.com/ocbe-uio/course_med3007/blob/main/lab/code/MED3007_Lab2_pca.R">Code</a></td>
-<td style="text-align: center;"></td>
+<td style="text-align: center;"><a href="presentation/Lab_pca.pdf">Slides</a></td>
 </tr>
 <tr class="odd">
+<td style="text-align: center;"></td>
+<td style="text-align: center;"></td>
+<td style="text-align: center;"></td>
+<td style="text-align: center;"></td>
+</tr>
+<tr class="even">
 <td style="text-align: center;">Day 4</td>
 <td style="text-align: center;"><a href="../lab/lab_day4_clustering.html">Day 4: Clustering</a></td>
 <td style="text-align: center;"><a href="https://github.com/ocbe-uio/course_med3007/blob/main/lab/code/MED3007_Lab3_clustering.R">Code</a></td>
@@ -571,10 +577,11 @@ <h2 class="anchored" data-anchor-id="lab-notes-and-r-scripts">Lab notes and R sc
 <span id="cb1-35"><a href="#cb1-35" aria-hidden="true" tabindex="-1"></a>| Day 1   | <span class="co">[</span><span class="ot">Day 1: Introduction to R</span><span class="co">](lab_day1_intro.qmd)</span>  |  <span class="co">[</span><span class="ot">Code</span><span class="co">](https://github.com/ocbe-uio/course_med3007/blob/main/lab/code/MED3007_Lab0_intro_r.R)</span>  |<span class="co">[</span><span class="ot">Slides</span><span class="co">](presentation/Lab_IntroR.pdf)</span> |</span>
 <span id="cb1-36"><a href="#cb1-36" aria-hidden="true" tabindex="-1"></a>|    | <span class="co">[</span><span class="ot">(supplement): Navigate RStudio and workspace</span><span class="co">](lab_day1s_rstudio.qmd)</span>  |  | |</span>
 <span id="cb1-37"><a href="#cb1-37" aria-hidden="true" tabindex="-1"></a>| Day 2  | <span class="co">[</span><span class="ot">Day 2: Multiple testing</span><span class="co">](lab_day2_testing.qmd)</span>  |   <span class="co">[</span><span class="ot">Code</span><span class="co">](https://github.com/ocbe-uio/course_med3007/blob/main/lab/code/MED3007_Lab1.R)</span>, <span class="co">[</span><span class="ot">Code (solution)</span><span class="co">](https://github.com/ocbe-uio/course_med3007/blob/main/lab/code/MED3007_Lab1_exercise_solution.R)</span> | |</span>
-<span id="cb1-38"><a href="#cb1-38" aria-hidden="true" tabindex="-1"></a>| Day 3  | <span class="co">[</span><span class="ot">Day 3: Principal Component Analysis</span><span class="co">](lab_day3_pca.qmd)</span> |<span class="co">[</span><span class="ot">Code</span><span class="co">](https://github.com/ocbe-uio/course_med3007/blob/main/lab/code/MED3007_Lab2_pca.R)</span> | |</span>
-<span id="cb1-39"><a href="#cb1-39" aria-hidden="true" tabindex="-1"></a>| Day 4  | <span class="co">[</span><span class="ot">Day 4: Clustering</span><span class="co">](lab_day4_clustering.qmd)</span> |<span class="co">[</span><span class="ot">Code</span><span class="co">](https://github.com/ocbe-uio/course_med3007/blob/main/lab/code/MED3007_Lab3_clustering.R)</span> | |</span>
-<span id="cb1-40"><a href="#cb1-40" aria-hidden="true" tabindex="-1"></a></span>
-<span id="cb1-41"><a href="#cb1-41" aria-hidden="true" tabindex="-1"></a></span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
+<span id="cb1-38"><a href="#cb1-38" aria-hidden="true" tabindex="-1"></a>| Day 3  | <span class="co">[</span><span class="ot">Day 3: Principal Component Analysis</span><span class="co">](lab_day3_pca.qmd)</span> |<span class="co">[</span><span class="ot">Code</span><span class="co">](https://github.com/ocbe-uio/course_med3007/blob/main/lab/code/MED3007_Lab2_pca.R)</span> | <span class="co">[</span><span class="ot">Slides</span><span class="co">](presentation/Lab_pca.pdf)</span></span>
+<span id="cb1-39"><a href="#cb1-39" aria-hidden="true" tabindex="-1"></a> |</span>
+<span id="cb1-40"><a href="#cb1-40" aria-hidden="true" tabindex="-1"></a>| Day 4  | <span class="co">[</span><span class="ot">Day 4: Clustering</span><span class="co">](lab_day4_clustering.qmd)</span> |<span class="co">[</span><span class="ot">Code</span><span class="co">](https://github.com/ocbe-uio/course_med3007/blob/main/lab/code/MED3007_Lab3_clustering.R)</span> | |</span>
+<span id="cb1-41"><a href="#cb1-41" aria-hidden="true" tabindex="-1"></a></span>
+<span id="cb1-42"><a href="#cb1-42" aria-hidden="true" tabindex="-1"></a></span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
 </div></div></div></div></div>
 </div> <!-- /content -->
 

diff --git a/docs/lab/presentation/Lab_pca.pdf b/docs/lab/presentation/Lab_pca.pdf
diff --git a/docs/search.json b/docs/search.json
diff --git a/lab/code/MED3007_Lab1_exercise_solution.R b/lab/code/MED3007_Lab1_exercise_solution.R
@@ -1,3 +1,9 @@
+###-------------------------###
+### SOLUTIONS TO EXERCISES  ###
+###-------------------------###
+
+
+
 ###--------------###
 ### 2. EXERCISE  ###
 ###--------------###
@@ -19,13 +25,17 @@ png('./Rplots/ex2_histogram_vitaminD.png', height = 500, width = 500)
 hist(Testfil_Rcourse$vitD_v1, prob = TRUE, xlab = 'vitamin D', main = 'histogram of vitamin D')
 dev.off() # for closing the plot file
 
-png('./Rplots/ex2_boxplot_vitaminD.png', height = 500, width = 500)
+png('./ex2_boxplot_vitaminD.png', height = 500, width = 500)
 boxplot(vitD_v1 ~ gender, Testfil_Rcourse)
 dev.off() # for closing the plot file
 
 
-vit1 <- Testfil_Rcourse[which(data$gender==0),'vitD_v1']
-vit2 <- Testfil_Rcourse[which(data$gender==1),'vitD_v1']
+vit1 <- Testfil_Rcourse[which(Testfil_Rcourse$gender==0),'vitD_v1']
+vit2 <- Testfil_Rcourse[which(Testfil_Rcourse$gender==1),'vitD_v1']
+# alternative syntax:
+vit1 <- Testfil_Rcourse$vitD_v1[which(Testfil_Rcourse$gender==0)]
+vit2 <- Testfil_Rcourse$vitD_v1[which(Testfil_Rcourse$gender==1)]
+
 
 
 qqnorm(vit1, main = 'vitamin D - females')
@@ -37,7 +47,7 @@ qqline(vit2)
 shapiro.test(vit1)
 shapiro.test(vit2) # data is normally distb.
 
-t.test(vitD_v1 ~ gender, data) # there is a difference in vitD between groups (gender)
+t.test(vitD_v1 ~ gender, Testfil_Rcourse) # there is a difference in vitD between groups (gender)
 
 
 
@@ -55,7 +65,8 @@ t.test(vitD_v1 ~ gender, data) # there is a difference in vitD between groups (g
 
 ## 1. Load in the data using read.csv(). You will need to select header=F.
 
-exp.data <-  read.csv("./data/Ch10Ex11.csv",header=FALSE)
+exp.data <-  read.csv("./data/Ch10Ex11.csv",header=F)
+dim(exp.data)
 # I want each row to represent a sample, and each column a gene
 exp.data <- t(exp.data) 
 dim(exp.data)
@@ -83,6 +94,9 @@ abline(h=0)
 
 # let us first apply the t-test and get an idea of the significance in the data
 # How many significant p-values (without any correction)
+pval.ttest <- apply(exp.data, 2, 
+                    function(x){t.test(x[which(groups==1)], x[which(groups==2)])$p.value})
+
 sum(pval.ttest < alpha)
 
 # how many expected false positives?
@@ -101,11 +115,15 @@ sum(pval.fdr  < alpha)
 
 # Get the significant genes (here the genes are numbered from 1 to 1000)
 sign.genes <- which(pval.fdr < alpha)
+# or 
+sign.genes <- which(pval.fwer < alpha)
 
 # Plot them
 plot(genes.gr.means[1,], xlab = 'genes', main = 'significant genes in blue', 
      ylim = range(genes.gr.means), ylab = 'mean expr')
 points(genes.gr.means[2,], col=2)
-points(sign.genes, genes.gr.means[1,sign.genes], col=4, pch=4)
-points(sign.genes,genes.gr.means[2,sign.genes], col=4, pch=4)
+points(sign.genes, genes.gr.means[1,sign.genes], col=4, pch=4, lwd=2)
+points(sign.genes,genes.gr.means[2,sign.genes], col=4, pch=4, lwd=2)
 abline(h=0)
+
+
diff --git a/lab/lab_day3_pca.qmd b/lab/lab_day3_pca.qmd
@@ -11,7 +11,7 @@ Download datasets [here](https://github.com/ocbe-uio/course_med3007/tree/main/la
 
 R script: [Code](https://github.com/ocbe-uio/course_med3007/blob/main/lab/code/MED3007_Lab2_pca.R)
 
-Lab Lecture 
+Lab Lecture: [Slides](presentation/Lab_pca.pdf)
 
 
 
@@ -187,6 +187,29 @@ scores_food <- pc_food$x
 round(scores_food, digits = 2)
 ```
 
+(Optional) You can double check that the score is indeed the product of the original data (after scaling) with loading. Take Albania for example (first row of the data). Its score for PC1 is 2.94. Let us try to reproduce this number from formula.
+
+```{r}
+#| label: pca-food-albania
+#| warning: false
+#| echo: true
+
+food[1,] 
+# scale the data
+food_s <- scale(food)
+round(food_s[1,], digits = 2) # after centering and scaling
+# loading for PC1 is the first column
+round(loading_food[,1], digits = 2)
+# let us muliple element by element, then sum together
+prod_by_element <- as.numeric(food_s[1,]) * loading_food[, 1]
+round(prod_by_element, digits = 2)
+# sum all elements together
+sum(prod_by_element)
+```
+
+
+
+
 Another important result from the `prcomp()` command is the explained variance (PVE, proportion of variance explained). All PCs explain 100% of the variance; while the first PC explains the largest proportion (46%), second PC explains the second largest proportion (18.2%) and so on. 
 
 You can find the variance (squared standard deviation, `sdev`) using the following commands. You can compare with the results from `summary()`, see if they correspond.

diff --git a/lab/overview.qmd b/lab/overview.qmd
@@ -35,7 +35,8 @@ day 4 (clustering)
 | Day 1   | [Day 1: Introduction to R](lab_day1_intro.qmd)  |  [Code](https://github.com/ocbe-uio/course_med3007/blob/main/lab/code/MED3007_Lab0_intro_r.R)  |[Slides](presentation/Lab_IntroR.pdf) |
 |    | [(supplement): Navigate RStudio and workspace](lab_day1s_rstudio.qmd)  |  | |
 | Day 2  | [Day 2: Multiple testing](lab_day2_testing.qmd)  |   [Code](https://github.com/ocbe-uio/course_med3007/blob/main/lab/code/MED3007_Lab1.R), [Code (solution)](https://github.com/ocbe-uio/course_med3007/blob/main/lab/code/MED3007_Lab1_exercise_solution.R) | |
-| Day 3  | [Day 3: Principal Component Analysis](lab_day3_pca.qmd) |[Code](https://github.com/ocbe-uio/course_med3007/blob/main/lab/code/MED3007_Lab2_pca.R) | |
+| Day 3  | [Day 3: Principal Component Analysis](lab_day3_pca.qmd) |[Code](https://github.com/ocbe-uio/course_med3007/blob/main/lab/code/MED3007_Lab2_pca.R) | [Slides](presentation/Lab_pca.pdf)
+ |
 | Day 4  | [Day 4: Clustering](lab_day4_clustering.qmd) |[Code](https://github.com/ocbe-uio/course_med3007/blob/main/lab/code/MED3007_Lab3_clustering.R) | |
 
 

diff --git a/lab/presentation/Lab_pca.pdf b/lab/presentation/Lab_pca.pdf