library(rmarkdown)
knitr::opts_chunk$set(echo = TRUE, message=FALSE,warning=FALSE,collapse = TRUE)
library(reshape2)
library(ggplot2)
library(dplyr)
library(plotly)
library(viridis)
library(data.table)
library(pheatmap)
library(tidyverse)
library(ggthemes)
library(clipr)
library(tidyr)
library(Rcpp)
mycolors<-c(viridis(15))
felix_cols<-mycolors[c(5,2)]
felix_4cols<-mycolors[c(15,10,8,2)]
plain_cols1<-c("blue","gray")
plain_cols2<-c("red","gray")
pats_cols<-colorRampPalette(c("#FDE725FF", "white","#440154FF"))(21)
leos_cols<-colorRampPalette(c("white","blue"))(10)
BC_data<-read_csv(file="breast_cancer_cells.csv")
#and then view it
view(BC_data)
BC_mat1<-BC_data %>% select(MCF10A_1:SKBR3_2) %>% as.matrix() %>% round(.,2)
head(BC_mat1)
## MCF10A_1 MCF10A_2 MCF7_1 MCF7_2 MDA231_1 MDA231_2 MDA468_1 MDA468_2
## [1,] 9.54 4.58 5.07 5.42 25.43 27.42 4.56 3.88
## [2,] 14.00 11.58 6.49 6.64 9.80 10.31 6.84 8.75
## [3,] 10.22 8.29 11.55 10.82 12.48 10.11 9.54 8.46
## [4,] 9.00 6.35 13.40 14.82 14.94 11.33 6.87 7.92
## [5,] 8.21 4.44 12.08 9.82 16.51 12.34 15.43 8.50
## [6,] 12.51 15.84 8.05 8.38 6.78 8.46 4.48 7.18
## SKBR3_1 SKBR3_2
## [1,] 8.46 5.64
## [2,] 11.91 13.67
## [3,] 9.10 9.43
## [4,] 8.54 6.82
## [5,] 7.93 4.75
## [6,] 13.27 15.05
pheatmap(BC_mat1, color=pats_cols,cellwidth=30,cellheight=.03,cluster_cols=FALSE,cluster_rows=TRUE,legend=TRUE,fontsize = 7,scale="column")
BC_data2<-BC_data %>% mutate(
mean_MCF10A= ((MCF10A_1 + MCF10A_2)/2),
mean_MCF7= ((MCF7_1 + MCF7_2)/2),
mean_MDA231= ((MDA231_1 + MDA231_2)/2),
mean_MDA468= ((MDA468_1 + MDA468_2)/2),
mean_SKBR3= ((SKBR3_1 + SKBR3_2)/2))
view(BC_data2)
BC_data2<-BC_data2 %>% mutate(
log_MCF7=log2(mean_MCF7/mean_MCF10A),
log_MDA231=log2(mean_MDA231/mean_MCF10A),
log_MDA468=log2(mean_MDA468/mean_MCF10A),
log_SKBR3=log2(mean_SKBR3/mean_MCF10A))
colnames(BC_data2)
## [1] "Gene_Symbol" "Description"
## [3] "Peptides" "MCF10A_1"
## [5] "MCF10A_2" "MCF7_1"
## [7] "MCF7_2" "MDA231_1"
## [9] "MDA231_2" "MDA468_1"
## [11] "MDA468_2" "SKBR3_1"
## [13] "SKBR3_2" "pvalue_MCF7_vs_MCF10A"
## [15] "pvalue_MDA231_vs_MCF10A" "pvalue_MDA468_vs_MCF10A"
## [17] "pvalue_SKBR3_vs_MCF10A" "mean_MCF10A"
## [19] "mean_MCF7" "mean_MDA231"
## [21] "mean_MDA468" "mean_SKBR3"
## [23] "log_MCF7" "log_MDA231"
## [25] "log_MDA468" "log_SKBR3"
BC_mat2<-BC_data2 %>% select(log_MCF7:log_SKBR3) %>% as.matrix() %>% round(.,2)
pheatmap(BC_mat2, color=pats_cols,cellwidth=30,cellheight=.03,cluster_cols=FALSE,cluster_rows=TRUE,legend=TRUE,fontsize = 7,scale="column")
BC_data2<-BC_data2 %>% mutate(neglog_SKBR3=-log10(pvalue_SKBR3_vs_MCF10A))
## Use ggplot to plot the log ratio of ____ against the -log p-value ____
volcano_plot<-BC_data2 %>% ggplot(aes(x=log_SKBR3,y=neglog_SKBR3,description=Gene_Symbol))+
geom_point(alpha=0.7,color="blue")
#to view it, type:
volcano_plot
BC_data2<-BC_data2 %>% mutate(significance=ifelse((log_SKBR3>2.1 & neglog_SKBR3>2.99),"UP", ifelse((log_SKBR3<c(-2.1) & neglog_SKBR3>2.99),"DOWN","NOT SIG")))
## Some standard colors
plain_cols3<-c("red","gray","blue")
## volcano plot as before with some added things
better_volcano_plot<-BC_data2 %>% ggplot(aes(x=log_SKBR3,y=neglog_SKBR3,description=Gene_Symbol,color=significance))+
geom_point(alpha=0.7)+
scale_color_manual(values=plain_cols3)+
xlim(-6,6)+
theme_bw()+
theme(axis.text = element_text(colour = "black",size=14))+
theme(text = element_text(size=14))+
labs(x="log ratio of SKBR3 compared to control",y="-log(p-value)")
#to view it, type
better_volcano_plot
ggplotly(better_volcano_plot)
BC_long<-pivot_longer(BC_data2, cols = c(MCF10A_1:SKBR3_2), names_to = 'variable')%>% select(-c(pvalue_SKBR3_vs_MCF10A:significance))%>%select(-Description,-Peptides)
head(BC_long)
## # A tibble: 6 × 6
## Gene_Symbol pvalue_MCF7_vs_MCF10A pvalue_MDA231_vs_MCF10A
## <chr> <dbl> <dbl>
## 1 NES 0.542 0.0185
## 2 NES 0.542 0.0185
## 3 NES 0.542 0.0185
## 4 NES 0.542 0.0185
## 5 NES 0.542 0.0185
## 6 NES 0.542 0.0185
## # ℹ 3 more variables: pvalue_MDA468_vs_MCF10A <dbl>, variable <chr>,
## # value <dbl>
Examples_Down<-BC_long %>% filter(Gene_Symbol=="APOA1" | Gene_Symbol=="HLA-A;MYO1B" | Gene_Symbol=="HMGN5")
## make barplots facetted by Gene Symbol (when working with other data sets - change x=order to x = variable)
Example_plot_down<-Examples_Down %>%
ggplot(aes(x=factor(variable,levels=c('MCF10A_1','MCF10A_2','MCF7_1','MCF7_2','MDA231_1','MDA231_2','MDA468_1','MDA468_2','SKBR3_1','SKBR3_2')),y=value))+
geom_bar(stat="identity",fill="red")+
facet_wrap(~Gene_Symbol)+
theme_bw()+
theme(axis.text = element_text(colour = "black",size=10))+
theme(text = element_text(size=14))+
theme(axis.text.x = element_text(angle=45, hjust=1))+
labs(x="sample",y="relative intensity")
Example_plot_down
## Same process for the upregulated ones
Examples_Up<-BC_long %>% filter (Gene_Symbol=="GCLC" | Gene_Symbol=="FNBP1L" | Gene_Symbol=="DENND4C"| Gene_Symbol=="KRT23"| Gene_Symbol=="TDP2")
#check viewer and / or plots to see it
Example_plot_up<-Examples_Up %>% ggplot(aes(x=variable,y=value))+
geom_bar(stat="identity",fill="royalblue")+
facet_wrap(~Gene_Symbol)+
theme_bw()+
theme(axis.text = element_text(colour = "black",size=10))+
theme(text = element_text(size=14))+
theme(axis.text.x = element_text(angle=45, hjust=1))+
labs(x="sample",y="relative intensity")
Example_plot_up
Example_plot_down<-Examples_Down %>%
ggplot(aes(x=variable,y=value))+
geom_bar(stat="identity",fill="red")+
facet_wrap(~Gene_Symbol)+
theme_bw()+
theme(axis.text = element_text(colour = "black",size=10))+
theme(text = element_text(size=14))+
theme(axis.text.x = element_text(angle=45, hjust=1))+
labs(x="sample",y="relative intensity")