####Run through this example and try to understand what is going on with the data

####So, lets load RMarkdown

library(rmarkdown)

set messages to FALSE on everything (prevents certain boring things from being shown in the results)

knitr::opts_chunk$set(echo = TRUE, message=FALSE,warning=FALSE,collapse = TRUE)

PACKAGES

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)

COLORS

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)

LOAD DATA AND MAKE AN OVERALL HEATMAP

## load the dataset (cells treated with COVID)
C19_data<-read_csv(file="COVID_test.csv")

## make some new columns that store the log ratios  

C19_data<-C19_data %>% mutate(
  log_2h=log2((Virus_2h_1+Virus_2h_2)/(Control_2h_1+Control_2h_2)), 
  log_6h=log2((Virus_6h_1+Virus_6h_2)/(Control_2h_1+Control_2h_2)), 
  log_10h=log2((Virus_10h_1+Virus_10h_2)/(Control_2h_1+Control_2h_2)),
  log_24h=log2((Virus_24h_1+Virus_24h_2)/(Control_2h_1+Control_2h_2))) 

## make a matrix of just the data

C19_mat1<-C19_data %>% select(log_2h:log_24h) %>% as.matrix() %>% round(.,2)
## make a heatmap from the data

pheatmap(C19_mat1, color=pats_cols,cellwidth=30,cellheight=.03,cluster_cols=FALSE,cluster_rows=FALSE,legend=TRUE,fontsize = 7,scale="column")


#looks like 24hrs is different than the other cells lines.  Let's take a closer look at that

PLOTTING A VOLCANO

##BASIC VOLCANO PLOT

## Add a column that stores the negative log of the pvalue of interest (in this case, Virus_24h)
C19_data<-C19_data %>% mutate(neglog_24h=-log10(P_value_24h))

## Use ggplot to plot the log ratio at 24 hours against the -log p-value at 24 hours
volcano_plot<-C19_data %>% ggplot(aes(x=log_24h,y=neglog_24h,description=Gene_Symbol))+
  geom_point(alpha=0.7,color="royalblue")

volcano_plot


  
## BETTER VOLCANO PLOT

## Define the significant ones so they can be colored
C19_data<-C19_data %>% mutate(significance=ifelse((log_24h>2 & neglog_24h>2.99),"UP", ifelse((log_24h<c(-2) & neglog_24h>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<-C19_data %>% ggplot(aes(x=log_24h,y=neglog_24h,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 at 24 hrs COVID-19 infection compared to control",y="-log(p-value)")
  

better_volcano_plot

## use ggplotly to see hover over the points to see the gene names. Record these for the next step 
ggplotly(better_volcano_plot)

Barplots of significant points of interest

EXAMPLES OF A COUPLE PROTEINS or GENES

# Make a pivot longer table of the C19 data for Control_2h_1:Virus_24h_2
C19_long<-pivot_longer(C19_data, cols = c(Control_2h_1:Virus_24h_2), names_to = 'variable')%>% select(-c(P_value_2h:significance))%>%select(-Accession)

## Because ggplot will automatically plot alphabetically on the x-axis, we need to make sure that our timepoints will be in the proper order by adding a new column using the factor() function
#this should be excluded in other datasets where order is not needed

C19_long$order <- as.character(C19_long$variable)
C19_long$order <- factor(C19_long$order,levels=c('Control_2h_1','Control_2h_2','Virus_2h_1','Virus_2h_2','Virus_6h_1','Virus_6h_2','Virus_10h_1','Virus_10h_2','Virus_24h_1','Virus_24h_2'))


## based on the gene symbols from plotly, select a few proteins from the table. Select just the data and gene symbols and pivot longer
Examples_Down<-C19_long %>% filter(Gene_Symbol=="SHH" | Gene_Symbol=="APLP2;APP" | Gene_Symbol=="TSPYL1") 
  
## 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=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")
  
Example_plot_down

## Same process for the upregulated ones
Examples_Up<-C19_long %>% filter(Gene_Symbol=="Viral_Protein_4" | Gene_Symbol=="Viral_Protein_8")

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