Introduction

The example is split into 2 Parts:

Part 1 must be completed first to create a file, SNPs_cleaned.csv, that has been completely prepared for analysis.

Now in Part 2, you will analyze the data with PCA. The steps here will be:

  1. Center the data (scale())
  2. Run a PCA analysis (prcomp())
  3. Evaluate the scree plot from the PCA (screeplot())
  4. Evaluate the amount of variation explained by the first 2 PCs.
  5. Extract the PCA scores for plotting (vegan::scores())
  6. Plot the data

Tasks

In the code below all code is provided. Your tasks will be to do 4 things:

  1. Give a meaningful title to all sections marked “TODO: TITLE”
  2. Write 1 to 2 sentences describing what is being done and why in all sections marked “TODO: EXPLAIN”
  3. Add titles and axes to plots in all sections marked “TODO: UPDATE PLOT”
  4. Write 1 or 2 sentences interpreting the output from R in all sections marked “TODO: INTERPRET”

Preliminaries

Load the vcfR package with library()

library(vcfR) # KEY
## 
##    *****       ***   vcfR   ***       *****
##    This is vcfR 1.13.0 
##      browseVignettes('vcfR') # Documentation
##      citation('vcfR') # Citation
##    *****       *****      *****       *****
library(vegan)
## Loading required package: permute
## Loading required package: lattice
## This is vegan 2.6-4
library(ggplot2)
library(ggpubr)

Set the working directory

Load the data

SNPs_cleaned <- read.csv(file = "SNPs_cleaned.csv")

warning("If this didn't work, its may be because you didn't set your working directory.")
## Warning: If this didn't work, its may be because you didn't set your working
## directory.

Data analysis

TODO: TITLE: Scale data

TODO: EXPLAIN Scaling the data by standard deviation

SNPs_scaled <- scale(SNPs_cleaned)

TODO: TITLE: find principle components

TODO: EXPLAIN find the principle components to see data representation along the axis

pca_scaled <- prcomp(SNPs_scaled)

TODO: TITLE Show Screenplot

TODO: EXPLAIN Looking at the screeplot to see the principle components TODO: UPDATE PLOT WITH TITLE

screeplot(pca_scaled, 
          ylab  = "Relative importance",
          main = "SNP data screeplot")

TODO: INTERPRET SCREEPLOT From the screeplot, we are able to see that PC1 has the highest relative importance than the others.

TODO: TITLE: Summarization by percent variation

TODO: EXPLAIN Determine the best principle component by using percent variation

summary_out_scaled <- summary(pca_scaled)
PCA_variation <- function(pca_summary, PCs = 2){
  var_explained <- pca_summary$importance[2,1:PCs]*100
  var_explained <- round(var_explained,1)
  return(var_explained)
}
var_out <- PCA_variation(summary_out_scaled,PCs = 10)
N_columns <- ncol(SNPs_scaled)
barplot(var_out,
        main = "Percent variation Scree plot",
        ylab = "Percent variation explained")
abline(h = 1/N_columns*100, col = 2, lwd = 2)

TODO: INTERPRET VARIANCE EXPLAINED According to the data, we are able to see that PC1 is the best principle component in the context of percent variation

TODO: TITLE: Show Biplot

TODO: EXPLAIN Creating biplot to show the relationships between different features.

biplot(pca_scaled)

TODO: EXPLAIN WHY THIS IS A BAD IDEA However, since the plot is not readable because it has so many features, we are not able to interpret the data.

TODO: TITLE: Building Dataframe with PCA

TODO: EXPLAIN To make scatterplot, we must first convert PCA with their IDs to dataframes

pca_scores <- vegan::scores(pca_scaled)

TODO: EXPLAIN These are IDs for each pattern. We save it into a object called “pop_id”

pop_id <- c("Nel","Nel","Nel","Nel","Nel","Nel","Nel","Nel",
"Nel", "Nel", "Nel", "Nel", "Nel", "Nel", "Nel", "Alt",
"Alt", "Alt", "Alt", "Alt", "Alt", "Alt", "Alt", "Alt",
"Alt", "Alt", "Alt", "Alt", "Alt", "Alt", "Sub", "Sub",
"Sub", "Sub", "Sub", "Sub", "Sub", "Sub", "Sub", "Sub",
"Sub", "Cau", "Cau", "Cau", "Cau", "Cau", "Cau", "Cau",
"Cau", "Cau", "Cau", "Cau", "Cau", "Div", "Div", "Div",
"Div", "Div", "Div", "Div", "Div", "Div", "Div", "Div",
"Div", "Div", "Div", "Div")

TODO: EXPLAIN convert both the PCA and the IDs to dataframes

pca_scores2 <- data.frame(pop_id,
                              pca_scores)

TODO: TITLE: Building Scatterplot using PCA scores

TODO: EXPLAIN We create scatterplot using PCA scores and assign different colors according to different categories

TODO: UPDATE PLOT WITH TITLE TODO: UPDATE X and Y AXES WITH AMOUNT OF VARIATION EXPLAINED

ggpubr::ggscatter(data = pca_scores2,
                  y = "PC2",
                  x = "PC1",
                  color = "pop_id",
                  shape = "pop_id",
                  xlab = "PC1 (19.9% variation)",
                  ylab = "PC2 (2.3% variation)",
                  main = "Bird SNPs PCA Scatterplot")

TODO: INTERPRET PLOT According to the scatterplot, we are able to interpret that among all types of birds, Cau and Div are strongly correlated with each other.