2022 R Final Project

This repository is for the final project of course 22100 by Group 9: Huang Yi (s210304), Liu Yu (S212610), Anne Rossen (s194687) and Synne Spjelkavik (s174526).

Introduction

This project is based on a publicly available Breast Cancer proteomic dataset. Link: https://www.nature.com/articles/nature18003

The aim of the project is to investigate whether it is possible to detect and correctly classify the 4 breast cancer subtypes using different sets of protein expression data and compare the results of the classification. Analyzes used are GLM, PCA and K-means.

Project Structure

Data: All data files, which are derivatives from the raw data generated by project scripts
patients <- read_csv(file = "data/01_patients.csv") #
PAM50_clean <- read_csv(file = "data/02_PAM50_clean.csv") #
proteomes_clean <- read_csv(file = "data/02_proteomes_clean.csv") #
BC_data_clean_aug <- read_csv(file = "data/03_BC_data_clean_aug.csv") # Merge of patients and proteomes
BC_data_PAM50_clean <- read_csv(file = "data/02_BC_data_PAM50_clean.csv") # Merge of BC_data and PAM50\

R: All project scripts
file.edit("R/01_load.R") #
file.edit("R/02_clean.R") #
file.edit("R/03_augment.R") #
file.edit("R/04_PlotData.R") #
file.edit("R/05_analysis_glm.R") #
file.edit("R/06_analysis_PickGenes.R") #
file.edit("R/07_analysis_PAM50_clustering.R") #\

Results: All the plots and tables generated by the scripts
#Data Ploted
library(magick) #
image_read("results/04_plot_AgeVitalCancerType.png") #\

#GLM results
read_csv(file = "results/05_Basal_glm.csv") #
read_csv(file = "results/05_Her2_glm.csv") #
read_csv(file = "results/05_LumA_glm.csv") #
read_csv(file = "results/05_LumB_glm.csv") #
image_read("results/05_subtype_heatmap.png") #
image_read("results/05_venndiagram.png") #
read_csv(file = "results/06_BC_overlap_genes.csv") #\

#PCA and Kmeans
image_read("results/06_BC_overlap_PCA.png") #
image_read("results/07_BC_data_cumulative_kmeans.png") #
image_read("results/07_BC_data_PAM50_cumulative_kmeans.png") #\

doc: Final presentation
rmarkdown::render(input = "doc/presentation.Rmd", output_file="../doc/group9_presentation.html", knit_root_dir = "..")

Running the Scripts

source(file = "R/00_doit.R")