DEG-GO_analysis.Rmd

---
title: "GO analysis"
author: "ERM"
date: "`r Sys.Date()`"
output: html_document
---

```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = FALSE, warning = FALSE)
```

# GO Analysis

I have created several files from the RNA analysis that contain the significant genes(determined by adj.P.val \< 0.1) from each Time and Condition. The names of the files are in the following format: 'sigV'+Drug(2 letters)+time.

example: 'sigVDA3.txt' means this file contains the significant DE genes from the Daunorubicin 3 hour compared to Vehicle Control 3 hour analysis

```{r Import libraries, message=FALSE, warning=FALSE}
library(tidyverse)
library(gprofiler2)
library(readr)
library(BiocGenerics)
library(gridExtra)
library(VennDiagram)
library(kableExtra)
library(scales)
library(ggVennDiagram)
library(cowplot)
```

```{r Import data, echo=FALSE, message=FALSE, warning=FALSE}
 
##note, in this chunk the .rmd file is in the cm/analysis folder.   need to back out with .. 
##this code checks the directory and searches for the pattern, returning the full name of the file
file.names <- list.files(path = "data/", pattern = "sig*", ignore.case = TRUE,full.names = TRUE)

##Next I use the list of names and lapply to read all files into a list
##made the csv file to reimport
#write.csv(filenameonly,"data/filenameonly.txt", col.names = FALSE,row.names = FALSE)
##readin the filenameonly.txt file from data to use for naming and filtering.
filenameonly <- read_csv("data/filenameonly.txt")
#loop through the list of files and make a separate dataframe for each file under the 'real'  name of the data set
for (k in 1:length(file.names)){
  
  assign(paste0(filenameonly$x[k]) , read.csv(file.names[k]))
}

##rename the columns to the previous names

colnames(sigVDA24)<- c("ENTREZID","SYMBOL","logFC","AveExpr","t","P.Value","adj.P.Val","B")
colnames(sigVDX24)<- c("ENTREZID","SYMBOL","logFC","AveExpr","t","P.Value","adj.P.Val","B")
colnames(sigVEP24)<- c("ENTREZID","SYMBOL","logFC","AveExpr","t","P.Value","adj.P.Val","B")
colnames(sigVMT24)<- c("ENTREZID","SYMBOL","logFC","AveExpr","t","P.Value","adj.P.Val","B")
colnames(sigVTR24)<- c("ENTREZID","SYMBOL","logFC","AveExpr","t","P.Value","adj.P.Val","B")
colnames(sigVDA3)<- c("ENTREZID","SYMBOL","logFC","AveExpr","t","P.Value","adj.P.Val","B")
colnames(sigVDX3)<- c("ENTREZID","SYMBOL","logFC","AveExpr","t","P.Value","adj.P.Val","B")
colnames(sigVEP3)<- c("ENTREZID","SYMBOL","logFC","AveExpr","t","P.Value","adj.P.Val","B")
colnames(sigVMT3)<- c("ENTREZID","SYMBOL","logFC","AveExpr","t","P.Value","adj.P.Val","B")
colnames(sigVTR3)<- c("ENTREZID","SYMBOL","logFC","AveExpr","t","P.Value","adj.P.Val","B")
```

The analysis is based on all genes that passed the rowMeans\>0 from the previous page [link](https://reneeisnowhere.github.io/Cardiotoxicity/RNAseqrun_1_analysis.html)

```{r uploading the background genes, echo=FALSE,  message=FALSE, warning=FALSE, paged.print=FALSE}
#backGL <- efit2$genes   ***making the list

#write.csv(backGL, "data/backGL.txt", row.names = FALSE)
##read the  .csv the render into txt in the data file!
 backGL <- read_csv("data/backGL.txt", 
    col_types = cols(...1 = col_skip()))
```

Below is the analysis of differentially expressed genes for each treatment at 3 hours and 24 hours.
##Daunorubicin
```{r intial analysis DA3 and DA24, echo=FALSE, message=FALSE, warning=FALSE}
DAspgostres3 <- gost(query = sigVDA3$ENTREZID, 
                    organism = "hsapiens",
                    ordered_query = TRUE,
                    domain_scope = "custom",
                    measure_underrepresentation = FALSE,
                    evcodes = FALSE,
                    user_threshold = 0.05,
                    correction_method = c("fdr"),
                    custom_bg = backGL$ENTREZID,
                    sources=c("GO:BP","KEGG"))
DAspp3 <- gostplot(DAspgostres3, capped = FALSE, interactive = TRUE)
DAspp3

DAsptable3 <- DAspgostres3$result %>% 
  dplyr::select(c(source, term_id,
                  term_name,intersection_size, 
                   term_size, p_value))# %>% 
 
DAtable1 %>% mutate_at(.cols = 6, .funs= scientific_format()) %>% 
  kable(.,) %>% 
  kable_paper("striped", full_width = FALSE) %>%  
  kable_styling(full_width = FALSE, position = "left",bootstrap_options = c("striped","hover")) %>% 
  scroll_box(width = "100%", height = "400px")
```

###daun 3hour sig DEG

```{r daun 3 hour sig DEG,}


total3 <- list(sigVDA3$ENTREZID,sigVDX3$ENTREZID,sigVEP3$ENTREZID,sigVMT3$ENTREZID)
ggVennDiagram(total3,
              category.names = c("Daunorubicin","Doxorubicin","Epirubicin","Mitoxantrone"),
              show_intersect = FALSE,
              set_color = "black",
              label = "both",
              label_percent_digit = 1,
              label_size = 3.5,
              label_alpha = 0,
              label_color = "black",
              edge_lty = "solid") +
  scale_x_continuous(expand = expansion(mult = .2))+
  scale_fill_gradient(low = "red2", high = "yellow")+
  labs(title = "3 hour comparison of significant genes", caption = "n = 554 genes")+
  theme(plot.title = element_text(size = rel(1.5), hjust = 0.5))



list3totvenn3 <- get.venn.partitions(total3)
list24totvenn24 <- get.venn.partitions(total24)



Daun3sp <- list3totvenn3$..values..[[15]]
Daun24sp <- list24totvenn24$..values..[[15]]





DAspgostres3 <- gost(query = Daun3sp, 
                    organism = "hsapiens",
                    ordered_query = TRUE,
                    domain_scope = "custom",
                    measure_underrepresentation = FALSE,
                    evcodes = FALSE,
                    user_threshold = 0.05,
                    correction_method = c("fdr"),
                    custom_bg = backGL$ENTREZID,
                    sources=c("GO:BP","KEGG"))
DAspp3 <- gostplot(DAspgostres3, capped = FALSE, interactive = TRUE)
DAspp3

DAsptable3 <- DAspgostres3$result %>% 
  dplyr::select(c(source, term_id,
                  term_name,intersection_size, 
                   term_size, p_value))# %>% 
 DAspp3 <- gostplot(DAspgostres3, capped = FALSE, interactive = TRUE)
DAspp3

DAsptable3 <- DAspgostres3$result %>% 
  dplyr::select(c(source, term_id,
                  term_name,intersection_size, 
                   term_size, p_value))# %>% 

  DAsptable3 %>% 
    dplyr::filter(source=="GO:BP") %>% 
    dplyr::select(p_value,term_name,intersection_size) %>%
    slice_min(., n=10 ,order_by=p_value) %>%
    mutate(log_val = -log10(p_value)) %>%
   # slice_max(., n=10,order_by = p_value) %>%
   ggplot(., aes(x = log_val, y =reorder(term_name,p_value), col= intersection_size)) +
    geom_point(aes(size = intersection_size)) +
    scale_y_discrete(labels = wrap_format(30))+
    guides(col="none", size=guide_legend(title = "# of intersected \n terms"))+
    ggtitle('Daunorubicin 3 hour GO:BP terms') +
   xlab(expression(" -"~log[10]~(adj. p-value)))+
    ylab("GO: BP term")+
      theme_bw()+
    theme(plot.title = element_text(size = rel(1.5), hjust = 0.5),
        axis.title = element_text(size = 15, color = "black"),
        axis.ticks = element_line(linewidth = 1.5),
        axis.line = element_line(linewidth = 1.5),
        axis.text = element_text(size = 10, color = "black", angle = 0),
        strip.text.x = element_text(size = 15, color = "black", face = "bold"))
DAtable1 %>% mutate_at(.cols = 6, .funs= scientific_format()) %>% 
  kable(.,) %>% 
  kable_paper("striped", full_width = FALSE) %>%  
  kable_styling(full_width = FALSE, position = "left",bootstrap_options = c("striped","hover")) %>% 
  scroll_box(width = "100%", height = "400px")
#---------------
```




```{r daun specific Venn genes}
# 3 hour daun specific genes (venn)
  DAgostres <- gost(query = Daun3sp,  organism = "hsapiens",
  ordered_query = TRUE,
  domain_scope = "custom",
  measure_underrepresentation = FALSE,
  evcodes = FALSE,
  user_threshold = 0.0001,
  correction_method = c("fdr"),
  custom_bg = backGL$ENTREZID,
  sources=c("GO:BP","KEGG"))

DAspp3 <- gostplot(DAspgostres3, capped = FALSE, interactive = TRUE)
DAspp3

DAsptable3 <- DAspgostres3$result %>% 
  dplyr::select(c(source, term_id, term_name,intersection_size, term_size, p_value))
  DAsptable3 %>% 
    dplyr::filter(source=="GO:BP") %>% 
    dplyr::select(p_value,term_name,intersection_size) %>%
    slice_min(., n=10 ,order_by=p_value) %>%
    mutate(log_val = -log10(p_value)) %>%
   # slice_max(., n=10,order_by = p_value) %>%
   ggplot(., aes(x = log_val, y =reorder(term_name,p_value), col= intersection_size)) +
    geom_point(aes(size = intersection_size)) +
    scale_y_discrete(labels = wrap_format(30))+
    guides(col="none", size=guide_legend(title = "# of intersected \n terms"))+
    ggtitle('Daunorubicin specific 3 hour GO:BP \nterms n = 350') +
    xlab(expression(" -"~log[10]~("adj. p-value")))+
    ylab("GO: BP term")+
      theme_bw()+
    theme(plot.title = element_text(size = rel(1.5), hjust = 0.5),
        axis.title = element_text(size = 15, color = "black"),
        axis.ticks = element_line(linewidth = 1.5),
        axis.line = element_line(linewidth = 1.5),
        axis.text = element_text(size = 10, color = "black", angle = 0),
        strip.text.x = element_text(size = 15, color = "black", face = "bold"))
  
  
DAsptable3 %>% mutate_at(.cols = 6, .funs= scientific_format()) %>% 
  kable(.,) %>% 
  kable_paper("striped", full_width = FALSE) %>%  
  kable_styling(full_width = FALSE, position = "left",bootstrap_options = c("striped","hover")) %>% 
  scroll_box(width = "100%", height = "400px")




ggVennDiagram::ggVennDiagram(list(Daun3sp, sigVDA24$ENTREZID),
              category.names = c("3 hours","24 hours"),
              show_intersect = FALSE,
              set_color = "black",
              label = "both",
              label_percent_digit = 1,
              label_size = 3.5,
              label_alpha = 0,
              label_color = "black",
              edge_lty = "solid") +
  scale_x_continuous(expand = expansion(mult = .2))+
  scale_fill_gradient(low = "red4", high = "tan1")+
  labs(title = "3 hour venn with 24 hour Daun", caption = "n = 554 genes")+
  theme(plot.title = element_text(size = rel(1.5), hjust = 0.5))

Dau3sp_Dau24sig <- get.venn.partitions(list(Daun3sp, sigVDA24$ENTREZID,DDEresp ))
dau3hoursp_ddesp <- Dau3sp_Dau24sig$..values..[[1]]
##gives 150 genes expressed in Daun only at 3 hours and in the DDE set of 4400 at 24 hours.
 DAgostressp3 <- gost(query = dau3hoursp_ddesp,  organism = "hsapiens",
  ordered_query = TRUE,
  domain_scope = "custom",
  measure_underrepresentation = FALSE,
  evcodes = FALSE,
  user_threshold = 0.05,
  correction_method = c("fdr"),
  custom_bg = backGL$ENTREZID,
  sources=c("GO:BP","KEGG"))

DAspp3 <- gostplot(DAgostressp3, capped = FALSE, interactive = TRUE)
DAspp3

DAsptable3 <- DAgostressp3$result %>% 
  dplyr::select(c(source, term_id, term_name,intersection_size, term_size, p_value))
  DAsptable3 %>% 
    dplyr::filter(source=="GO:BP") %>% 
    dplyr::select(p_value,term_name,intersection_size) %>%
    slice_min(., n=20,order_by=p_value) %>%
    mutate(log_val = -log10(p_value)) %>%
   # slice_max(., n=10,order_by = p_value) %>%
   ggplot(., aes(x = log_val, y =reorder(term_name,p_value), col= intersection_size)) +
    geom_point(aes(size = intersection_size)) +
    scale_y_discrete(labels = wrap_format(30))+
    guides(col="none", size=guide_legend(title = "# of intersected \n terms"))+
    ggtitle('Daunorubicin 3 hour sp in DDE GO:BP \nterms n = 150') +
    xlab(expression(" -"~log[10]~("adj. p-value")))+
    ylab("GO: BP term")+
      theme_bw()+
    theme(plot.title = element_text(size = rel(1.5), hjust = 0.5),
        axis.title = element_text(size = 15, color = "black"),
        axis.ticks = element_line(linewidth = 1.5),
        axis.line = element_line(linewidth = 1.5),
        axis.text = element_text(size = 10, color = "black", angle = 0),
        strip.text.x = element_text(size = 15,  color = "black", face = "bold"))
  
  
 # ---------------------------------------------------------
```
###Daunorubicin 24 sigDEG
```{r DA sig deg 24}
  DAgostres24 <- gost(query = sigVDA24$ENTREZID,  
                    organism = "hsapiens",
                    ordered_query = TRUE,
                    domain_scope = "custom",
                    measure_underrepresentation = FALSE,
                    evcodes = FALSE,
                    user_threshold = 0.0001,
                    correction_method = c("fdr"),
                    custom_bg = backGL$ENTREZID,
                    sources=c("GO:BP","KEGG"))
DAp24 <- gostplot(DAgostres24, capped = FALSE, interactive = TRUE)
DAp24

DAtable124 <- DAgostres24$result %>% 
  dplyr::select(c(source, term_id, term_name,intersection_size, term_size, p_value))  
  

  DAtable124 %>% 
  dplyr::filter(source=="GO:BP") %>% 
    dplyr::select(p_value,term_name,intersection_size) %>%
    slice_min(., n=10 ,order_by=p_value) %>%
    mutate(log_val = -log10(p_value)) %>%
   # slice_max(., n=10,order_by = p_value) %>%
   ggplot(., aes(x = log_val, y =reorder(term_name,p_value), col= intersection_size)) +
    geom_point(aes(size = intersection_size)) +
    scale_y_discrete(labels = wrap_format(30))+
    guides(col="none", size=guide_legend(title = "# of intersected \n terms"))+
    ggtitle('Daunorubicin 24 hour GO:BP terms') +
    xlab(expression(" -"~log[10]~(adj. p-value)))+
    ylab("GO: BP term")+
      theme_bw()+ 
    theme(plot.title = element_text(size = rel(1.5), hjust = 0.5),
        axis.title = element_text(size = 15, color = "black"),
        axis.ticks = element_line(linewidth = 1.5),
        axis.line = element_line(linewidth = 1.5),
        axis.text = element_text(size = 10, color = "black", angle = 0),
        strip.text.x = element_text(size = 15, color = "black", face = "bold"))
  
  DAtable124 %>% mmutate_at(.cols = 6, .funs= scientific_format()) %>% 
  kable(.,) %>% 
  kable_paper("striped", full_width = FALSE) %>%  
  kable_styling(full_width = FALSE, position = "left",bootstrap_options = c("striped","hover")) %>% 
  scroll_box(width = "100%", height = "400px")
  
#------------------------------------------------------
```
  
###Daunorubicin 3 v 24 intersection

```{r DA 3v24 int}
daun3v24int <- intersect(sigVDA3$ENTREZID,sigVDA24$ENTREZID)
DAgostresint <- gost(query = daun3v24int,  
                    organism = "hsapiens",
                    ordered_query = TRUE,
                    domain_scope = "custom",
                    measure_underrepresentation = FALSE,
                    evcodes = FALSE,
                    user_threshold = 0.0001,
                    correction_method = c("fdr"),
                    custom_bg = backGL$ENTREZID,
                    sources=c("GO:BP","KEGG"))
DApint <- gostplot(DAgostresint, capped = FALSE, interactive = TRUE)
DApint

DAtable_int <- DAgostresint$result %>% 
  dplyr::select(c(source, term_id, term_name,intersection_size, term_size, p_value))  
 
  DAtable_int %>% 
    dplyr::filter(source=="GO:BP") %>% 
    dplyr::select(p_value,term_name,intersection_size) %>%
    slice_min(., n=10 ,order_by=p_value) %>%
    mutate(log_val = -log10(p_value)) %>%
    ggplot(., aes(x = log_val, y =reorder(term_name,p_value), 
                  col= intersection_size)) +
    geom_point(aes(size = intersection_size)) +
    scale_y_discrete(labels = wrap_format(30))+
    guides(col="none", 
           size=guide_legend(title = "# of intersected \n terms"))+
    ggtitle('Daunorubicin intersect 3 v 24 hour GO:BP terms') +
    xlab(expression(" -"~log[10]~(adj. p-value)))+
    ylab("GO: BP term")+
    theme_bw()+
    theme(plot.title = element_text(size = rel(1.5), hjust = 0.5),
          axis.title = element_text(size = 15, color = "black"),
          axis.ticks = element_line(linewidth = 1.5),
          axis.line = element_line(linewidth = 1.5),
          axis.text = element_text(size = 10, color = "black", angle = 0),
          strip.text.x = element_text(size = 15, color = "black", face = "bold"))
  
  DAtable_int %>% 
    mutate_at(.cols = 6, .funs= scientific_format()) %>% 
    kable(.,) %>% 
    kable_paper("striped", full_width = FALSE) %>%
    kable_styling(full_width = FALSE, position = "left",bootstrap_options = c("striped","hover")) %>%
    scroll_box(width = "100%", height = "400px") 
  
  
  
  
```

##Doxorubicin 3 hour and 24 hour
```{r intial analysis DX3 and DX24, echo=FALSE, message=FALSE, warning=FALSE}
DXgostres <- gost(query = sigVDX3$ENTREZID,  
                  organism = "hsapiens",
                  ordered_query = TRUE,
                  domain_scope = "custom",
                  measure_underrepresentation = FALSE,
                  evcodes = FALSE,
                  user_threshold = 0.05,
                  correction_method = c("fdr"),
                  custom_bg = backGL$ENTREZID,
                  sources=c("GO:BP","KEGG"))

DXp <- gostplot(DXgostres, capped = FALSE, interactive = TRUE)
DXp

DXtable1 <- DXgostres$result %>% 
  dplyr::select(c(source, term_id, term_name,intersection_size, term_size, p_value)) 


  DXtable1 %>% 
    dplyr::filter(source=="GO:BP") %>% 
    dplyr::select(p_value,term_name,intersection_size) %>%
    slice_min(., n=10 ,order_by=p_value) %>%
    mutate(log_val = -log10(p_value)) %>%
   # slice_max(., n=10,order_by = p_value) %>%
   ggplot(., aes(x = log_val, y =reorder(term_name,p_value), col= intersection_size)) +
    geom_point(aes(size = intersection_size)) +
    scale_y_discrete(labels = wrap_format(30))+
    guides(col="none", size=guide_legend(title = "# of intersected \n terms"))+
    ggtitle('Doxorubicin 3 hour GO:BP terms') +
    xlab(expression(" -"~log[10]~(adj. p-value)))+
    ylab("GO: BP term")+
      theme_bw()+
    theme(plot.title = element_text(size = rel(1.5), hjust = 0.5),
        axis.title = element_text(size = 15, color = "black"),
        axis.ticks = element_line(linewidth = 1.5),
        axis.line = element_line(linewidth = 1.5),
        axis.text = element_text(size = 9, color = "black", angle = 0),
        strip.text.x = element_text(size = 15, color = "black", face = "bold"))

  DXtable1%>% mutate_at(.cols = 6, .funs= scientific_format()) %>% 
  kable(.,) %>% 
  kable_paper("striped", full_width = FALSE) %>%  
  kable_styling(full_width = FALSE, position = "left",bootstrap_options = c("striped","hover")) %>% 
  scroll_box(width = "100%", height = "400px")
  
 # ---------------------------------------------------------
  
  gostres24 <- gost(query = sigVDX24$ENTREZID,  
                    organism = "hsapiens",
                    ordered_query = TRUE,
                    domain_scope = "custom",
                    measure_underrepresentation = FALSE,
                    evcodes = FALSE,
                    user_threshold = 0.0001,
                    correction_method = c("fdr"),
                    custom_bg = backGL$ENTREZID,
                    sources=c("GO:BP","KEGG"))
p24 <- gostplot(gostres24, capped = FALSE, interactive = TRUE)
p24

DXtable124 <- gostres24$result %>% 
  dplyr::select(c(source, term_id, term_name,intersection_size, term_size, p_value)) 



  table124 %>% 
    dplyr::filter(source=="GO:BP") %>% 
    dplyr::select(p_value,term_name,intersection_size) %>%
    slice_min(., n=10 ,order_by=p_value) %>%
    mutate(log_val = -log10(p_value)) %>%
   # slice_max(., n=10,order_by = p_value) %>%
   ggplot(., aes(x = log_val, y =reorder(term_name,p_value), col= intersection_size)) +
    geom_point(aes(size = intersection_size)) +
    scale_y_discrete(labels = wrap_format(30))+
    guides(col="none", size=guide_legend(title = "# of intersected\n terms"))+
    ggtitle('Doxorubicin 24 hour GO:BP terms') +
    xlab(expression(" -"~log[10]~(adj. p-value)))+
    ylab("GO: BP term")+
      theme_bw()+
    theme(plot.title = element_text(size = rel(1.5), hjust = 0.5),
        axis.title = element_text(size = 15, color = "black"),
        axis.ticks = element_line(linewidth = 1.5),
        axis.line = element_line(linewidth = 1.5),
        axis.text = element_text(size = 10, color = "black", angle = 0),
        strip.text.x = element_text(size = 15, color = "black", face = "bold"))
DXtable124 %>% 
  mutate_at(.cols = 6, .funs= scientific_format()) %>% 
  kable(.,) %>% 
  kable_paper("striped", full_width = FALSE) %>%  
  kable_styling(full_width = FALSE, position = "left",bootstrap_options = c("striped","hover")) %>% 
  scroll_box(width = "100%", height = "400px")
```

##Epirubicin


```{r intial analysis EP3 and EP24, echo=FALSE, message=FALSE, warning=FALSE}






EPgostres <- gost(query = sigVEP3$ENTREZID,  organism = "hsapiens",
  ordered_query = TRUE,
  domain_scope = "custom",
  measure_underrepresentation = FALSE,
  evcodes = FALSE,
  user_threshold = 0.01,
  correction_method = c("fdr"),
  custom_bg = backGL$ENTREZID,
  sources=c("GO:BP","KEGG"))
EPp <- gostplot(EPgostres, capped = FALSE, interactive = TRUE)
EPp 

EPtable1 <- EPgostres$result %>% 
  dplyr::select(c(source, term_id, term_name,intersection_size, term_size, p_value))  




  EPtable1 %>% 
    dplyr::filter(source=="GO:BP") %>% 
    dplyr::select(p_value,term_name,intersection_size) %>%
    slice_min(., n=10 ,order_by=p_value) %>%
    mutate(log_val = -log10(p_value)) %>%
   # slice_max(., n=10,order_by = p_value) %>%
   ggplot(., aes(x = log_val, y =reorder(term_name,p_value), col= intersection_size)) +
    geom_point(aes(size = intersection_size)) +
    scale_y_discrete(labels = wrap_format(30))+
    guides(col="none", size=guide_legend(title = "# of intersected \n terms"))+
    ggtitle('Epirubicin 3 hour top ten GO:BP terms') +
    xlab(expression(" -"~log[10]~(adj. p-value)))+
    ylab("GO: BP term")+
      theme_bw()+
    theme(plot.title = element_text(size = rel(1.5), hjust = 0.5),
        axis.title = element_text(size = 15, color = "black"),
        axis.ticks = element_line(linewidth = 1.5),
        axis.line = element_line(linewidth = 1.5),
        axis.text = element_text(size = 10, color = "black", angle = 0),
        strip.text.x = element_text(size = 15, color = "black", face = "bold"))
  EPtable1%>%
    mutate_at(.cols = 6, .funs= scientific_format()) %>% 
  kable(.,) %>% 
  kable_paper("striped", full_width = FALSE) %>%  
  kable_styling(full_width = FALSE, position = "left",bootstrap_options = c("striped","hover")) %>% 
  scroll_box(width = "100%", height = "400px")
  
 # ---------------------------------------------------------
  
  EPgostres24 <- gost(query = sigVEP24$ENTREZID,  
                    organism = "hsapiens",
                    ordered_query = TRUE,
                    domain_scope = "custom",
                    measure_underrepresentation = FALSE,
                    evcodes = FALSE,
                    user_threshold = 0.0001,
                    correction_method = c("fdr"),
                    custom_bg = backGL$ENTREZID,
                    sources=c("GO:BP","KEGG"))
EPp24 <- gostplot(EPgostres24, capped = FALSE, interactive = TRUE)
EPp24

EPtable124 <- EPgostres24$result %>% 
  dplyr::select(c(source, term_id, term_name,intersection_size, term_size, p_value)) 
  EPtable124 %>% 
    dplyr::filter(source=="GO:BP") %>% 
    dplyr::select(p_value,term_name,intersection_size) %>%
    slice_min(., n=10 ,order_by=p_value) %>%
    mutate(log_val = -log10(p_value)) %>%
   # slice_max(., n=10,order_by = p_value) %>%
   ggplot(., aes(x = log_val, y =reorder(term_name,p_value), col= intersection_size)) +
    geom_point(aes(size = intersection_size)) +
    scale_y_discrete(labels = wrap_format(30))+
    guides(col="none", size=guide_legend(title = "# of intersected \n terms"))+
    ggtitle('Epirubicin 24 hour top 10 GO:BP terms') +
    xlab(expression(" -"~log[10]~(adj. p-value)))+
    ylab("GO: BP term")+
      theme_bw()+
    theme(plot.title = element_text(size = rel(1.5), hjust = 0.5),
        axis.title = element_text(size = 15, color = "black"),
        axis.ticks = element_line(linewidth = 1.5),
        axis.line = element_line(linewidth = 1.5),
        axis.text = element_text(size = 10, color = "black", angle = 0),
        strip.text.x = element_text(size = 15, color = "black", face = "bold"))
EPtable124%>% 
  mutate_at(.cols = 6, .funs= scientific_format()) %>% 
  kable(.,) %>% 
  kable_paper("striped", full_width = FALSE) %>%  
  kable_styling(full_width = FALSE, position = "left",bootstrap_options = c("striped","hover")) %>% 
  scroll_box(width = "100%", height = "400px")



##epi 24 specifi n = 533
epsp24 <- list24totvenn24$..values..[[12]]
mtsp24 <- list24totvenn24$..values..[[8]]

EPgostres24sp <- gost(query = epsp24,  
                    organism = "hsapiens",
                    ordered_query = TRUE,
                    domain_scope = "custom",
                    measure_underrepresentation = FALSE,
                    evcodes = FALSE,
                    user_threshold = 0.05,
                    correction_method = c("fdr"),
                    custom_bg = backGL$ENTREZID,
                    sources=c("GO:BP","KEGG"))

EPsptable124 <- EPgostres24sp$result %>% 
  dplyr::select(c(source, term_id, term_name,intersection_size, term_size, p_value)) 

MTgostres24sp <- gost(query = mtsp24,  
                    organism = "hsapiens",
                    ordered_query = TRUE,
                    domain_scope = "custom",
                    measure_underrepresentation = FALSE,
                    evcodes = FALSE,
                    user_threshold = 0.05,
                    correction_method = c("fdr"),
                    custom_bg = backGL$ENTREZID,
                    sources=c("GO:BP","KEGG"))
MTsptable124 <- MTgostres24sp$result %>% 
  dplyr::select(c(source, term_id, term_name,intersection_size, term_size, p_value)) 


MTsptable124%>% 
    dplyr::filter(source=="GO:BP") %>% 
    dplyr::select(p_value,term_name,intersection_size) %>%
    slice_min(., n=10 ,order_by=p_value) %>%
    mutate(log_val = -log10(p_value)) %>%
   # slice_max(., n=10,order_by = p_value) %>%
   ggplot(., aes(x = log_val, y =reorder(term_name,p_value), col= intersection_size)) +
    geom_point(aes(size = intersection_size)) +
    scale_y_discrete(labels = wrap_format(30))+
    guides(col="none", size=guide_legend(title = "# of intersected \n terms"))+
    ggtitle("Mitoxantrone specific 24 hour top GO:BP terms") +
     xlab(expression(" -"~log[10]~"(adj p-value)"))+
    ylab("GO: BP term")+
      theme_bw()+
    theme(plot.title = element_text(size = rel(1.5), hjust = 0.5),
        axis.title = element_text(size = 15, color = "black"),
        axis.ticks = element_line(linewidth = 1.5),
        axis.line = element_line(linewidth = 1.5),
        axis.text = element_text(size = 10, color = "black", angle = 0),
        strip.text.x = element_text(size = 15, color = "black", face = "bold"))



```

##Mitoxantrone


```{r intial analysis MT3 and MT24, echo=FALSE, message=FALSE, warning=FALSE}
MTgostres <- gost(query = sigVMT3$ENTREZID,  organism = "hsapiens",
  ordered_query = TRUE,
  domain_scope = "custom",
  measure_underrepresentation = FALSE,
  evcodes = FALSE,
  user_threshold = 0.0001,
  correction_method = c("fdr"),
  custom_bg = backGL$ENTREZID,
  sources=c("GO:BP","KEGG"))
MTp <- gostplot(MTgostres, capped = FALSE, interactive = TRUE)
MTp

MTtable1 <- MTgostres$result %>% 
  dplyr::select(c(source, term_id, term_name,intersection_size, term_size, p_value))
MTtable1 %>% 
 dplyr::filter(source=="GO:BP") %>% 
    dplyr::select(p_value,term_name,intersection_size) %>%
    slice_min(., n=10 ,order_by=p_value) %>%
    mutate(log_val = -log10(p_value)) %>%
   # slice_max(., n=10,order_by = p_value) %>%
   ggplot(., aes(x = log_val, y =reorder(term_name,p_value), col= intersection_size)) +
    geom_point(aes(size = intersection_size)) +
    scale_y_discrete(labels = wrap_format(30))+
    guides(col="none", size=guide_legend(title = "# of intersected \n terms"))+
    ggtitle('Mitoxatrone 3 hour GO:BP terms') +
    xlab(expression(" -"~log[10]~(adj p-value)))+
    ylab("GO: BP term")+
      theme_bw()+
    theme(plot.title = element_text(size = rel(1.5), hjust = 0.5),
        axis.title = element_text(size = 15, color = "black"),
        axis.ticks = element_line(linewidth = 1.5),
        axis.line = element_line(linewidth = 1.5),
        axis.text = element_text(size = 10, color = "black", angle = 0),
        strip.text.x = element_text(size = 15, color = "black", face = "bold")) 
  MTtable1 %>% 
    mutate_at(.cols = 6, .funs= scientific_format()) %>% 
  kable(.,) %>% 
  kable_paper("striped", full_width = FALSE) %>%  
  kable_styling(full_width = FALSE, position = "left",bootstrap_options = c("striped","hover")) %>% 
  scroll_box(width = "100%", height = "400px")
 # ---------------------------------------------------------
  
  MTgostres24 <- gost(query = sigVMT24$ENTREZID,  
                    organism = "hsapiens",
                    ordered_query = TRUE,
                    domain_scope = "custom",
                    measure_underrepresentation = FALSE,
                    evcodes = FALSE,
                    user_threshold = 0.0001,
                    correction_method = c("fdr"),
                    custom_bg = backGL$ENTREZID,
                    sources=c("GO:BP","KEGG"))
MTp24 <- gostplot(MTgostres24, capped = FALSE, interactive = TRUE)
MTp24

MTtable124 <- MTgostres24$result %>% 
  dplyr::select(c(source, term_id, term_name,intersection_size, term_size, p_value)) 

MTtable124 %>% dplyr::filter(source=="GO:BP") %>% 
    dplyr::select(p_value,term_name,intersection_size) %>%
    slice_min(., n=10 ,order_by=p_value) %>%
    mutate(log_val = -log10(p_value)) %>%
   # slice_max(., n=10,order_by = p_value) %>%
   ggplot(., aes(x = log_val, y =reorder(term_name,p_value), col= intersection_size)) +
    geom_point(aes(size = intersection_size)) +
    scale_y_discrete(labels = wrap_format(30))+
    guides(col="none", size=guide_legend(title = "# of intersected \n terms"))+
    ggtitle('Mitoxantrone 24 hour GO:BP terms') +
    xlab(expression(" -"~log[10]~(adj. p-value)))+
    ylab("GO: BP term")+
      theme_bw()+
    theme(plot.title = element_text(size = rel(1.5), hjust = 0.5),
        axis.title = element_text(size = 15, color = "black"),
        axis.ticks = element_line(linewidth = 1.5),
        axis.line = element_line(linewidth = 1.5),
        axis.text = element_text(size = 10, color = "black", angle = 0),
        strip.text.x = element_text(size = 15, color = "black", face = "bold"))

  
  MTtable124 %>% 
    mutate_at(.cols = 6, .funs= scientific_format()) %>% 
  kable(.,) %>% 
  kable_paper("striped", full_width = FALSE) %>%  
  kable_styling(full_width = FALSE, position = "left",bootstrap_options = c("striped","hover")) %>% 
  scroll_box(width = "100%", height = "400px")
  

  
  
  
  

```




```{r up down trial data3, eval=FALSE, message=FALSE, warning=FALSE, include=FALSE}
#results_sig = subset(sigVDA3,adj.P.Val < 0.05)
# # get the significant up-regulated genes
#   up = subset(sigVDA3, logFC > 0)
# # get the significant down-regulated genes
# down = subset(sigVDA3, logFC < 0)
# gp_up = gost(query = up$ENTREZID,  organism = "hsapiens",
#   ordered_query = TRUE,
#   domain_scope = "custom",
#   measure_underrepresentation = FALSE,
#   evcodes = FALSE,
#   user_threshold = 0.001,
#   correction_method = c("fdr"),
#   custom_bg = backGL$ENTREZID,
#   sources=c("GO:BP","GO:MF", "GO:CC"))
#   
#   
# #gp_down = gost(query = down$ENTREZID,  organism = "hsapiens",
#   # ordered_query = TRUE,
#   # domain_scope = "custom",
#   # measure_underrepresentation = FALSE,
#   # evcodes = FALSE,
#   # user_threshold = 0.05,
#   # correction_method = c("bonferroni"),
#   # custom_bg = backGL$ENTREZID,
#   # sources=c("GO:BP","GO:MF", "GO:CC"))
# p2_up <- gostplot(gp_up, capped = FALSE, interactive = TRUE)
# 
# p2_up #+ ggtitle("Daunorubicin up regulated gene enrichment at 3 hours")

```



```{r showing it all, eval = FALSE, echo=FALSE, message=FALSE, warning=FALSE}

#p2_down <- gostplot(gp_down, capped = FALSE, interactive = TRUE)

#p2_down #+ ggtitle("Daunorubicin down regulated gene enrichment at 3 hours")


```

##Venn Diagrams
```{r venndiagram, echo=FALSE, message=FALSE, warning=FALSE}
total24 <-list(sigVDA24$ENTREZID,sigVDX24$ENTREZID,sigVEP24$ENTREZID,sigVMT24$ENTREZID)
in_common24 <-c(sigVDA24$ENTREZID,sigVDX24$ENTREZID,sigVEP24$ENTREZID,sigVMT24$ENTREZID)

length(unique(in_common24))

ggVennDiagram(total24,
              category.names = c("Daunorubicin\n n = 7518",
                                 "Doxorubicin\n n = 6793","Epirubicin\n n = 6783","Mitoxantrone\n n = 1213"),
              show_intersect = FALSE,
              set_color = "black",
              catagory_size = c(6,6,6,6),
              label = "both",
              label_percent_digit = 1,
              label_size = 3,
              label_alpha = 0,
              label_color = "black",
              edge_lty = "solid", set_size = )+
  scale_x_continuous(expand = expansion(mult = .3))+
   scale_y_continuous(expand = expansion(mult = .2))+
  #scale_color_brewer(palette = "Dark2",name = "Individual",)
scale_fill_distiller(palette="RdYlBu")+
  labs(title = "24 hour comparison of significant DE genes", caption = "n = 8887 genes")+
  theme(plot.title = element_text(size = rel(1.6), hjust = 0.5, vjust =1))


```



```{r 3 hours, echo=FALSE, message=FALSE, warning=FALSE}
# total 3 ----------------------------------------------------------------

total3 <- list(sigVDA3$ENTREZID,sigVDX3$ENTREZID,sigVEP3$ENTREZID,sigVMT3$ENTREZID)
totalin_common3 <- c(sigVDA3$SYMBOL,sigVDX3$SYMBOL,sigVEP3$SYMBOL,sigVMT3$SYMBOL)
length(unique(totalin_common3))

ggVennDiagram(total3,
              category.names = c("Daunorubicin\nn = 523",
                                 "Doxorubicin\nn = 25",
                                 "Epirubicin\nn = 179",
                                 "Mitoxantrone\nn = 52"),
                show_intersect = FALSE,
              set_color = "black",
              catagory_size = c(6,6,6,6),
              label = "both",
              label_percent_digit = 1,
              label_size = 3,
              label_alpha = 0,
              label_color = "black",
              edge_lty = "solid", set_size = )+
  scale_x_continuous(expand = expansion(mult = .3))+
   scale_y_continuous(expand = expansion(mult = .2))+
  #scale_color_brewer(palette = "Dark2",name = "Individual",)
scale_fill_distiller(palette="RdYlBu")+
  labs(title = "3 hour comparison of significant DE genes", caption = "n = 554 genes")+
  theme(plot.title = element_text(size = rel(1.6), hjust = 0.5, vjust =1))
```

```{r 3V 24 BY DRUG, eval=FALSE, include=FALSE}
# Dauno comp --------------------------------------------------------------

Dauncomp <- list(sigVDA24$ENTREZID,sigVDA3$ENTREZID)
in_commonDa <- c(sigVDA24$ENTREZID,sigVDA3$ENTREZID)
length(unique(in_commonDa))
ggVennDiagram(Dauncomp,
              category.names = c("Daunorubicin-24","Daunorubicin-3"),
              show_intersect = FALSE,
              set_color = "black",
              label = "both",
              label_percent_digit = 1,
              label_size = 3,
              label_alpha = 0,
              label_color = "black",
              edge_lty = "solid", set_size = )+
  scale_x_continuous(expand = expansion(mult = .2))+
  scale_fill_gradient(low = "light blue", high = "yellow")+
  labs(title = "Comparision of Dauno 3h v 24h", caption = "n = 7732 genes")+
  theme(plot.title = element_text(size = rel(1.5), hjust = 0.5))
#Davenlist <- intersect(Dauncomp[[1]],Dauncomp[[2]])


# Doxocomp ----------------------------------------------------------------
Doxcomp <- list(sigVDX24$ENTREZID,sigVDX3$ENTREZID)
in_commonDx <- c(sigVDX24$ENTREZID,sigVDX3$ENTREZID)
length(unique(in_commonDx))
ggVennDiagram(Doxcomp,
              category.names = c("Doxorubicin-24","Doxorubicin-3"),
              show_intersect = FALSE,
              set_color = "black",
              label = "both",
              label_percent_digit = 1,
              label_size = 3,
              label_alpha = 0,
              label_color = "black",
              edge_lty = "solid", set_size = )+
  scale_x_continuous(expand = expansion(mult = .2))+
  scale_fill_gradient(low = "light blue", high = "yellow")+
  labs(title = "Comparision of Doxo 3h v 24h", caption = "n = 6808 genes")+
  theme(plot.title = element_text(size = rel(1.5), hjust = 0.5))
#Dxvenlist <- intersect(Doxcomp[[1]],Doxcomp[[2]])
#length(intersect(Dxvenlist,Davenlist))## 7 of DX are in DA
# Epi Comp ----------------------------------------------------------------

Epicomp <- list(sigVEP24$ENTREZID,sigVEP3$ENTREZID)
in_commonEp <- c(sigVEP24$ENTREZID,sigVEP3$ENTREZID)
length(unique(in_commonEp))

ggVennDiagram(Epicomp,
              category.names = c("Epirubicin-24","Epirubicin-3"),
              show_intersect = FALSE,
              set_color = "black",
              label = "both",
              label_percent_digit = 1,
              label_size = 3,
              label_alpha = 0,
              label_color = "black",
              edge_lty = "solid", set_size = )+
  scale_x_continuous(expand = expansion(mult = .2))+
  scale_fill_gradient(low = "light blue", high = "yellow")+
  labs(title = "Comparision of Epi 3h v 24h", caption = "n = 6858 genes")+
  theme(plot.title = element_text(size = rel(1.5), hjust = 0.5))

#Epvenlist <- intersect(Epicomp[[1]],Epicomp[[2]])
# qA <- (intersect(Epvenlist , Davenlist))##76 are are in da
# qB <- (intersect(Dxvenlist, Epvenlist))#y are in Ep from Dx
# #ACintersect <- intersect(qA,qB)##total of 6 are in all 3 comparisons
# #ggVennDiagram(qA,qB)
#               
# qA
# qB
# Mito comp ---------------------------------------------------------------
Mitocomp <- list(sigVMT24$ENTREZID,sigVMT3$ENTREZID)
in_commonMt <-c(sigVMT24$ENTREZID,sigVMT3$ENTREZID)
length(unique(in_commonMt))

ggVennDiagram(Mitocomp,
              category.names = c("Mitoxantrone-24","Mitoxantrone-3"),
              show_intersect = FALSE,
              set_color = "black",
              label = "both",
              label_percent_digit = 1,
              label_size = 3,
              label_alpha = 0,
              label_color = "black",
              edge_lty = "solid", set_size = )+
  scale_x_continuous(expand = expansion(mult = .21))+
  scale_fill_gradient(low = "light blue", high = "yellow")+
  labs(title = "Comparision of Mito 3h v 24h", caption = "n = 1251 genes")+
  theme(plot.title = element_text(size = rel(1.5), hjust = 0.5))
#Mtvenlist <- intersect(Mitocomp[[1]],Mitocomp[[2]])
#intersect(Mtvenlist,ACintersect)
#

# AC comparison -----------------------------------------------------------

ACcomp <- list(sigVDA24$ENTREZID,sigVDX24$ENTREZID,sigVEP24$ENTREZID,sigVDA3$ENTREZID,sigVDX3$ENTREZID,sigVEP3$ENTREZID)
in_commonAC <-c(sigVDA24$ENTREZID,sigVDX24$ENTREZID,sigVEP24$ENTREZID,sigVDA3$ENTREZID,sigVDX3$ENTREZID,sigVEP3$ENTREZID)
length(unique(in_commonAC))

ggVennDiagram(ACcomp,
              category.names = c("Daunorubicin-24","Doxorubicin-24", "Epirubicin-24","Daunorubicin-3","Doxorubicin-3", "Epirubicin-3"),
              show_intersect = FALSE,
              set_color = "black",
              label = "both",
              label_percent_digit = 1,
              label_size = 3,
              label_alpha = 0,
              label_color = "black",
              edge_lty = "solid", set_size = )+
  scale_x_continuous(expand = expansion(mult = .21))+
  scale_fill_gradient(low = "purple", high = "yellow")+
  labs(title = "Comparision AC 3h v 24h", caption = "n = 8925 genes")+
  theme(plot.title = element_text(size = rel(1.5), hjust = 0.5))



```
Tell me more!  what are those genes at 24 hours??  give me some GO!
# Graphing GO of the Venn results

```{r Making Go lists from SigDE genes, message=FALSE, warning=FALSE, paged.print=TRUE}
list24totvenn <- get.venn.partitions(total24)
# > list24totvenn$..count..
#  [1]  960   13   19   15   37   18   25  126 4440  265  538  533  661  399  838
# > list24totvenn$..values..[?]
# ##From Cormotif
##pairwise
DDEresp<- list24totvenn$..values..[[9]]
write.csv(DDEresp,"data/DDEresp_list.csv")
DDEMresp<- list24totvenn$..values..[[1]]

DXsprespon<- list24totvenn$..values..[[14]]

# top2bi <-  read_csv("data/Top2biresp_cluster24h.csv",col_select = x, col_types = "c")
# NRresp <- read_csv("data/nonresponse_cluster24h.csv",col_select =x,col_types = "c")


complete <- c(sigVDA24$ENTREZID,sigVDX24$ENTREZID,sigVEP24$ENTREZID,sigVMT24$ENTREZID, sigVDA3$ENTREZID, sigVDX3$ENTREZID, sigVEP3$ENTREZID, sigVMT3$ENTREZID)
complete <- as.data.frame(unique(complete))
colnames(complete) <- "ENTREZID"  #9047 genes

NoResp <- backGL %>% anti_join(.,(complete), by = "ENTREZID") #5776 genes
#5776
       
```
##DDE response of 4400
```{r GO DDEresp genelist of SDEGs, echo=FALSE, message=FALSE, warning=FALSE}


DDEgostres24 <- gost(query = DDEresp,  
                    organism = "hsapiens",
                    ordered_query = TRUE,
                    domain_scope = "custom",
                    measure_underrepresentation = FALSE,
                    evcodes = FALSE,
                    user_threshold = 0.05,
                    correction_method = c("fdr"),
                    custom_bg = backGL$ENTREZID,
                    sources=c("GO:BP","KEGG"))
DDEp24 <- gostplot(DDEgostres24, capped = FALSE, interactive = TRUE)
DDEp24

DDEtable24 <- DDEgostres24$result %>% 
  dplyr::select(c(source, term_id, term_name,intersection_size, term_size, p_value)) 

DDEtable24 %>% dplyr::filter(source=="GO:BP") %>% 
    dplyr::select(p_value,term_name,intersection_size) %>%
    slice_min(., n= 10 ,order_by=p_value) %>%
    mutate(log_val = -log10(p_value)) %>%
   # slice_max(., n=10,order_by = p_value) %>%
   ggplot(., aes(x = log_val, y =reorder(term_name,p_value), col= intersection_size)) +
    geom_point(aes(size = intersection_size)) +
    scale_y_discrete(labels = wrap_format(30))+
    guides(col="none", size=guide_legend(title = "# of intersected \n terms"))+
    ggtitle('Anthracycline DEGs (n=4400) 24 hr GO:BP terms') +
    xlab(expression(" -"~log[10]~"(adj p-value)"))+
    ylab("GO: BP term")+
      theme_bw()+
    theme(plot.title = element_text(size = rel(1.2), hjust = 0.5),
        axis.title = element_text(size = 12, color = "black"),
        axis.ticks = element_line(linewidth = 1.5),
        axis.line = element_line(linewidth = 1.5),
        axis.text = element_text(size = 8, color = "black", angle = 0),
        strip.text.x = element_text(size = 15, color = "black", face = "bold"))

  
  DDEtable24 %>% 
    mutate_at(.cols = 6, .funs= scientific_format()) %>% 
  kable(.,) %>% 
  kable_paper("striped", full_width = FALSE) %>%  
  kable_styling(full_width = FALSE, position = "left",bootstrap_options = c("striped","hover")) %>% 
  scroll_box(width = "100%", height = "400px")

``` 
##DDEM response of 952
```{r GO DDEM 952 genelist of SDEGs, echo=FALSE, message=FALSE, warning=FALSE}


DDEMgostres24 <- gost(query = DDEMresp,  
                    organism = "hsapiens",
                    ordered_query = TRUE,
                    domain_scope = "custom",
                    measure_underrepresentation = FALSE,
                    evcodes = FALSE,
                    user_threshold = 0.05,
                    correction_method = c("fdr"),
                    custom_bg = backGL$ENTREZID,
                    sources=c("GO:BP","KEGG"))
DDEMp24 <- gostplot(DDEMgostres24, capped = FALSE, interactive = TRUE)
DDEMp24

DDEMtable24 <- DDEMgostres24$result %>% 
  dplyr::select(c(source, term_id, term_name,intersection_size, term_size, p_value)) 

DDEMtable24 %>% dplyr::filter(source=="GO:BP") %>% 
    dplyr::select(p_value,term_name,intersection_size) %>%
    slice_min(., n= 10 ,order_by=p_value) %>%
    mutate(log_val = -log10(p_value)) %>%
   # slice_max(., n=10,order_by = p_value) %>%
   ggplot(., aes(x = log_val, y =reorder(term_name,p_value), col= intersection_size)) +
    geom_point(aes(size = intersection_size)) +
    scale_y_discrete(labels = wrap_format(30))+
    guides(col="none", size=guide_legend(title = "# of intersected \n terms"))+
    ggtitle('Top2B inhibitor DEGs (n=952) 24 hr GO:BP terms') +
    xlab(expression(" -"~log[10]~"(adj. p-value)"))+
    ylab("GO: BP term")+
      theme_bw()+
    theme(plot.title = element_text(size = rel(1.2), hjust = 0.5),
        axis.title = element_text(size = 12, color = "black"),
        axis.ticks = element_line(linewidth = 1.5),
        axis.line = element_line(linewidth = 1.5),
        axis.text = element_text(size = 8, color = "black", angle = 0),
        strip.text.x = element_text(size = 15, color = "black", face = "bold"))


DDEMtable24 %>% dplyr::filter(source=="KEGG") %>% 
    dplyr::select(p_value,term_name,intersection_size) %>%
    slice_min(., n= 10 ,order_by=p_value) %>%
    mutate(log_val = -log10(p_value)) %>%
   # slice_max(., n=10,order_by = p_value) %>%
   ggplot(., aes(x = log_val, y =reorder(term_name,p_value), col= intersection_size)) +
    geom_point(aes(size = intersection_size, col = "red")) +
    scale_y_discrete(labels = wrap_format(30))+
    guides(col="none", size=guide_legend(title = "# of intersected \n terms"))+
    ggtitle('Top2B inhibitor DEGs (n=952) 24 hr KEGG terms') +
    xlab(expression(" -"~log[10]~"(adj. p-value)"))+
    ylab("KEGG term")+
      theme_bw()+
    theme(plot.title = element_text(size = rel(1.2), hjust = 0.5),
        axis.title = element_text(size = 12, color = "black"),
        axis.ticks = element_line(linewidth = 1.5),
        axis.line = element_line(linewidth = 1.5),
        axis.text = element_text(size = 8, color = "black", angle = 0),
        strip.text.x = element_text(size = 15, color = "black", face = "bold"))
  
  DDEMtable24 %>% 
    mutate_at(.cols = 6, .funs= scientific_format()) %>% 
  kable(.,) %>% 
  kable_paper("striped", full_width = FALSE) %>%  
  kable_styling(full_width = FALSE, position = "left",bootstrap_options = c("striped","hover")) %>% 
  scroll_box(width = "100%", height = "400px")
  
```
 
##dauno sp response 3 hour
```{r GO Daun3sp genelist of SDEGs, echo=FALSE, message=FALSE, warning=FALSE}


DAspgostres3 <- gost(query = Daun3sp, 
                    organism = "hsapiens",
                    ordered_query = TRUE,
                    domain_scope = "custom",
                    measure_underrepresentation = FALSE,
                    evcodes = FALSE,
                    user_threshold = 0.05,
                    correction_method = c("fdr"),
                    custom_bg = backGL$ENTREZID,
                    sources=c("GO:BP","KEGG"))
DAspp3 <- gostplot(DAspgostres3, capped = FALSE, interactive = TRUE)
DAspp3

DAsptable3 <- DAspgostres3$result %>% 
  dplyr::select(c(source, term_id, term_name,intersection_size, term_size, p_value)) 

DAsptable3 %>% dplyr::filter(source=="GO:BP") %>% 
    dplyr::select(p_value,term_name,intersection_size) %>%
    slice_min(., n= 10 ,order_by=p_value) %>%
    mutate(log_val = -log10(p_value)) %>%
   # slice_max(., n=10,order_by = p_value) %>%
   ggplot(., aes(x = log_val, y =reorder(term_name,p_value), col= intersection_size)) +
    geom_point(aes(size = intersection_size)) +
    scale_y_discrete(labels = wrap_format(30))+
    guides(col="none", size=guide_legend(title = "# of intersected \n terms"))+
    ggtitle('Daunorubicin specific DEGs (n=350) 3 hr GO:BP terms') +
    xlab(expression(" -"~log[10]~"(adj. p-value)"))+
    ylab("GO: BP term")+
      theme_bw()+
    theme(plot.title = element_text(size = rel(1.2), hjust = 0.5),
        axis.title = element_text(size = 12, color = "black"),
        axis.ticks = element_line(linewidth = 1.5),
        axis.line = element_line(linewidth = 1.5),
        axis.text = element_text(size = 8, color = "black", angle = 0),
        strip.text.x = element_text(size = 15, color = "black", face = "bold"))


DAsptable3 %>% dplyr::filter(source=="KEGG") %>% 
    dplyr::select(p_value,term_name,intersection_size) %>%
    slice_min(., n= 10 ,order_by=p_value) %>%
    mutate(log_val = -log10(p_value)) %>%
   # slice_max(., n=10,order_by = p_value) %>%
   ggplot(., aes(x = log_val, y =reorder(term_name,p_value), col= intersection_size)) +
    geom_point(aes(size = intersection_size, col = "red")) +
    scale_y_discrete(labels = wrap_format(30))+
    guides(col="none", size=guide_legend(title = "# of intersected \n terms"))+
    ggtitle('Daunorubicin specific DEGs (n=350) 3 hr KEGG terms') +
    xlab(expression(" -"~log[10]~"(adj. p-value)"))+
    ylab("KEGG term")+
      theme_bw()+
    theme(plot.title = element_text(size = rel(1.2), hjust = 0.5),
        axis.title = element_text(size = 12, color = "black"),
        axis.ticks = element_line(linewidth = 1.5),
        axis.line = element_line(linewidth = 1.5),
        axis.text = element_text(size = 8, color = "black", angle = 0),
        strip.text.x = element_text(size = 15, color = "black", face = "bold"))
  
  DAsptable3 %>% 
    mutate_at(.cols = 6, .funs= scientific_format()) %>% 
  kable(.,) %>% 
  kable_paper("striped", full_width = FALSE) %>%  
  kable_styling(full_width = FALSE, position = "left",bootstrap_options = c("striped","hover")) %>% 
  scroll_box(width = "100%", height = "400px")
  



``` 











##DXspecific response

###no erichment
```{r GO DXsprespon$ENTREZID genelist of SDEGs, echo=FALSE, message=FALSE, warning=FALSE}

### no enrichment
DXspgostres24 <- gost(query = DXsprespon, 
                    organism = "hsapiens",
                    ordered_query = TRUE,
                    domain_scope = "custom",
                    measure_underrepresentation = FALSE,
                    evcodes = FALSE,
                    user_threshold = 0.05,
                    correction_method = c("fdr"),
                    custom_bg = backGL$ENTREZID,
                    sources=c("GO:BP","KEGG"))
DXspp24 <- gostplot(DXspgostres24, capped = FALSE, interactive = TRUE)
DXspp24

DXsptable24 <- DXspgostres24$result %>% 
  dplyr::select(c(source, term_id, term_name,intersection_size, term_size, p_value)) 

DXsptable24 %>% dplyr::filter(source=="GO:BP") %>% 
    dplyr::select(p_value,term_name,intersection_size) %>%
    slice_min(., n= 10 ,order_by=p_value) %>%
    mutate(log_val = -log10(p_value)) %>%
   # slice_max(., n=10,order_by = p_value) %>%
   ggplot(., aes(x = log_val, y =reorder(term_name,p_value), col= intersection_size)) +
    geom_point(aes(size = intersection_size)) +
    scale_y_discrete(labels = wrap_format(30))+
    guides(col="none", size=guide_legend(title = "# of intersected \n terms"))+
    ggtitle('Doxorubicin specific DEGs (n=400) 24 hr GO:BP terms') +
    xlab(expression(" -"~log[10]~(adj. p-value)))+
    ylab("GO: BP term")+
      theme_bw()+
    theme(plot.title = element_text(size = rel(1.2), hjust = 0.5),
        axis.title = element_text(size = 12, color = "black"),
        axis.ticks = element_line(linewidth = 1.5),
        axis.line = element_line(linewidth = 1.5),
        axis.text = element_text(size = 8, color = "black", angle = 0),
        strip.text.x = element_text(size = 15, color = "black", face = "bold"))


DXsptable24 %>% dplyr::filter(source=="KEGG") %>% 
    dplyr::select(p_value,term_name,intersection_size) %>%
    slice_min(., n= 10 ,order_by=p_value) %>%
    mutate(log_val = -log10(p_value)) %>%
   # slice_max(., n=10,order_by = p_value) %>%
   ggplot(., aes(x = log_val, y =reorder(term_name,p_value), col= intersection_size)) +
    geom_point(aes(size = intersection_size, col = "red")) +
    scale_y_discrete(labels = wrap_format(30))+
    guides(col="none", size=guide_legend(title = "# of intersected \n terms"))+
    ggtitle('Doxorubicin specific DEGs (n=400) 24 hr KEGG terms') +
    xlab(expression(" -"~log[10]~(adj. p-value)))+
    ylab("KEGG term")+
      theme_bw()+
    theme(plot.title = element_text(size = rel(1.2), hjust = 0.5),
        axis.title = element_text(size = 12, color = "black"),
        axis.ticks = element_line(linewidth = 1.5),
        axis.line = element_line(linewidth = 1.5),
        axis.text = element_text(size = 8, color = "black", angle = 0),
        strip.text.x = element_text(size = 15, color = "black", face = "bold"))
  
  DXsptable24 %>% 
    mutate_at(.cols = 6, .funs= scientific_format()) %>% 
  kable(.,) %>% 
  kable_paper("striped", full_width = FALSE) %>%  
  kable_styling(full_width = FALSE, position = "left",bootstrap_options = c("striped","hover")) %>% 
  scroll_box(width = "100%", height = "400px")
  



``` 

##No Response genes in DEG- 5776 genes not showing DEG (union of all DEGS taken out of background genelist)

```{r no response cluster, echo=FALSE, message=FALSE, warning=FALSE}

gostresNoResp <- gost(query = NoResp$ENTREZID,  
                    organism = "hsapiens",
                    ordered_query = FALSE,
                    domain_scope = "custom",
                    measure_underrepresentation = FALSE,
                    evcodes = FALSE,
                    user_threshold = 0.05,
                    correction_method = c("fdr"),
                    custom_bg = backGL$ENTREZID,
                    sources=c("GO:BP","KEGG"))
Noresp24 <- gostplot(gostresNoResp, capped = FALSE, interactive = TRUE)
Noresp24

Noresp24table <- gostresNoResp$result %>% 
  dplyr::select(c(source, term_id, term_name,intersection_size, term_size, p_value)) 

# Noresp24table %>% dplyr::filter(source=="GO:BP") %>% 
#     dplyr::select(p_value,term_name,intersection_size) %>%
#     slice_min(., n= 10 ,order_by=p_value) %>%
#     mutate(log_val = -log10(p_value)) %>%
#    # slice_max(., n=10,order_by = p_value) %>%
#    ggplot(., aes(x = log_val, y =reorder(term_name,p_value), col= intersection_size)) +
#     geom_point(aes(size = intersection_size)) +
#     scale_y_discrete(labels = wrap_format(30))+
#     guides(col="none", size=guide_legend(title = "# of intersected \n terms"))+
#     ggtitle('No response genes(n=5775) 24 hr GO:BP terms') +
#     xlab(expression(" -"~log[10]~(adj. p-value)))+
#     ylab("GO: BP term")+
#       theme_bw()+
#     theme(plot.title = element_text(size = rel(1.2), hjust = 0.5),
#         axis.title = element_text(size = 12, color = "black"),
#         axis.ticks = element_line(linewidth = 1.5),
#         axis.line = element_line(linewidth = 1.5),
#         axis.text = element_text(size = 8, color = "black", angle = 0),
#         strip.text.x = element_text(size = 15, color = "black", face = "bold"))


Noresp24table %>% dplyr::filter(source=="GO:BP") %>% 
    dplyr::select(p_value,term_name,intersection_size) %>%
    slice_min(., n= 10 ,order_by=p_value) %>%
    mutate(log_val = -log10(p_value)) %>%
   # slice_max(., n=10,order_by = p_value) %>%
   ggplot(., aes(x = log_val, y =reorder(term_name,p_value), col= intersection_size)) +
    geom_point(aes(size = intersection_size, col = "red")) +
    scale_y_discrete(labels = wrap_format(30))+
    guides(col="none", size=guide_legend(title = "# of intersected \n terms"))+
    ggtitle('No response genes(n=5775) 24 hr KEGG terms') +
    xlab(expression(" -"~log[10]~(adj. p-value)))+
    ylab("KEGG term")+
      theme_bw()+
    theme(plot.title = element_text(size = rel(1.2), hjust = 0.5),
        axis.title = element_text(size = 12, color = "black"),
        axis.ticks = element_line(linewidth = 1.5),
        axis.line = element_line(linewidth = 1.5),
        axis.text = element_text(size = 8, color = "black", angle = 0),
        strip.text.x = element_text(size = 15, color = "black", face = "bold"))
  
  Noresp24table %>% 
    mutate_at(.cols = 6, .funs= scientific_format()) %>% 
  kable(.,) %>% 
  kable_paper("striped", full_width = FALSE) %>%  
  kable_styling(full_width = FALSE, position = "left",bootstrap_options = c("striped","hover")) %>% 
  scroll_box(width = "100%", height = "400px")
  

  
```
##GO:BP full figure DEG analysis






```{r GO:BP dEGlong, fig.height=8, fig.width=8, message=FALSE, warning=FALSE}



###24 hour DEG sets:
complete <- c(sigVDA24$ENTREZID,sigVDX24$ENTREZID,sigVEP24$ENTREZID,sigVMT24$ENTREZID, sigVDA3$ENTREZID, sigVDX3$ENTREZID, sigVEP3$ENTREZID, sigVMT3$ENTREZID)


# DDEtable24
# DDEMtable24
# 
# DXsptable24-no go enrichment
# 
# NoRespdeg-0no enrichment

DDEdeg <- DDEtable24 %>% dplyr::filter(source=="GO:BP") %>% 
    dplyr::select(p_value,term_name,intersection_size) %>%
    slice_min(., n=6 ,order_by=p_value) %>%
    mutate(log_val = -log10(p_value)) %>%mutate(order_val=rev(as.numeric(rownames(.)))) %>% slice_max(., n=10, order_by=order_val)

DDEMdeg <-  DDEMtable24 %>% dplyr::filter(source=="GO:BP") %>% 
    dplyr::select(p_value,term_name,intersection_size) %>%
    slice_min(., n=10 ,order_by=p_value) %>%
    mutate(log_val = -log10(p_value)) %>%mutate(order_val=rev(as.numeric(rownames(.))))

DAsp3deg <- DAsptable3%>% 
  dplyr::filter(source=="GO:BP") %>% 
    dplyr::select(p_value,term_name,intersection_size) %>%
    slice_min(., n=10 ,order_by=p_value) %>%
    mutate(log_val = -log10(p_value)) %>%
  mutate(order_val=rev(as.numeric(rownames(.))))


DAdeg <-   DAtable124 %>% dplyr::filter(source=="GO:BP") %>% 
    dplyr::select(p_value,term_name,intersection_size) %>%
    slice_min(., n=10 ,order_by=p_value) %>%
    mutate(log_val = -log10(p_value)) %>%mutate(order_val=rev(as.numeric(rownames(.))))
DXdeg <-  DXtable124 %>% dplyr::filter(source=="GO:BP") %>% 
    dplyr::select(p_value,term_name,intersection_size) %>%
    slice_min(., n=10 ,order_by=p_value) %>%
    mutate(log_val = -log10(p_value)) %>%mutate(order_val=rev(as.numeric(rownames(.))))
  EPdeg <-   EPtable124 %>% dplyr::filter(source=="GO:BP") %>% 
    dplyr::select(p_value,term_name,intersection_size) %>%
    slice_min(., n=10 ,order_by=p_value) %>%
    mutate(log_val = -log10(p_value)) %>%mutate(order_val=rev(as.numeric(rownames(.))))
  MTdeg <-   MTtable124 %>% dplyr::filter(source=="GO:BP") %>% 
    dplyr::select(p_value,term_name,intersection_size) %>%
    slice_min(., n=10 ,order_by=p_value) %>%
    mutate(log_val = -log10(p_value)) %>%mutate(order_val=rev(as.numeric(rownames(.))))
  
  
#filter<- dplyr::filter 
#select <- dplyr::select
  
GODEGlong <-list("DAdeg"=DAdeg,"DXdeg"=DXdeg,"EPdeg"=EPdeg,"MTdeg"=MTdeg,"DDEdeg"=DDEdeg,"DDEMdeg"=DDEMdeg)
GODEGlongtable <- data.table::rbindlist(GODEGlong, idcol = "deg")
 p <- GODEGlongtable%>% mutate(deg = fct_inorder(deg)) %>% 
ggplot(., aes(x = log_val, y=reorder(term_name, order_val, desc=TRUE),col=deg)) +
    geom_point(aes(size = intersection_size, col=deg)) +
  scale_y_discrete(labels = wrap_format(40))+
   scale_color_manual(values=c("#8B006D","#DF707E","#F1B72B", "#3386DD","red", "blue"))+
  facet_grid(deg ~.,scales="free_y" , labeller=label_bquote(.(as.expression(
               eval(parse(text = paste0('facet_deg', '$`', deg, '`')))
               )))) +
  
    guides(col=guide_legend(title="DEG Treatment group"), size=guide_legend(title = "# of intersected \n terms"))+
    ggtitle('Top 10 enriched GO:BP terms for each DEG list') +
    xlab("-log 10 (adj. p-value)")+
    
    ylab("GO:BP term")+
      theme_bw()+
 theme(plot.title = element_text(size = rel(1.5), hjust = 0.5),
        axis.title = element_text(size = 15, color = "black"),
        axis.ticks = element_line(linewidth = 1.5),
        axis.line = element_line(linewidth = 1.5),
        axis.text = element_text(size = 8, color = "black", angle = 0),
        #strip.text.x = element_text(size = 15, color = "black", face = "bold"),
        strip.text.y = element_text(color = "white"))

 
 
 facet_deg=(
   c(DAdeg = expression(atop("Daunorubicin",~italic("n = 7518"))),
     DXdeg=expression(atop("Doxorubicin",~italic("n = 6793"))),
     EPdeg=expression(atop("Epirubicin",~italic("n = 6783"))),
MTdeg=expression(atop( "Mitoxantrone",~italic("n = 1213"))),
    DDEdeg=expression(atop("DDE Intersection",~italic("n = 4400"))),
     DDEMdeg=expression(atop( "DDEM Intersection",~italic("n = 952")))))
   

#  title_labeller_function <- function(nuisance_parameter) {
#   return(facet_degnames() )
# }
 
 
#https://stackoverflow.com/questions/41631806/change-facet-label-text-and-background-colour/60046113#60046113
 
# drug_palNoVeh <- c("#8B006D","#DF707E","#F1B72B", "#3386DD","#707031","#41B333")
 
 
g <- ggplot_gtable(ggplot_build(p))
stripr <- which(grepl('strip-r', g$layout$name))
fills <- c("#8B006D","#DF707E","#F1B72B", "#3386DD","red", "blue", alpha= 0.8)
k <- 1
for (i in stripr) {
  j <- which(grepl('rect', g$grobs[[i]]$grobs[[1]]$childrenOrder))
  g$grobs[[i]]$grobs[[1]]$children[[j]]$gp$fill <- fills[k]
  k <- k+1
}
#print(grid::grid.draw(g))
grid::grid.draw(g)
#hex <- hue_pal()(4)

#hex
#> hex (red/green/blue/purple default in ggplot)
#[1] "#F8766D" "#7CAE00" "#00BFC4" "#C77CFF"
#p



```



##Visualization of FC in cormotif sets
```{r Correlation of FC values in all values and in DEG}
library(limma)
library(corrplot)
library(pheatmap)

efit2<- readRDS(file ="data/efit2results.RDS")

# saveRDS(efit2, file ="data/efit2results.RDS")

V.DA.top= topTable(efit2, coef=1, adjust="BH", number=Inf, sort.by="p")
V.DX.top= topTable(efit2, coef=2, adjust="BH", number=Inf, sort.by="p")
V.EP.top= topTable(efit2, coef=3, adjust="BH", number=Inf, sort.by="p")
V.MT.top= topTable(efit2, coef=4, adjust="BH", number=Inf, sort.by="p")
V.TR.top= topTable(efit2, coef=5, adjust="BH", number=Inf, sort.by="p")
V.DA24.top= topTable(efit2, coef=6, adjust="BH", number=Inf, sort.by="p")
V.DX24.top= topTable(efit2, coef=7, adjust="BH", number=Inf, sort.by="p")
V.EP24.top= topTable(efit2, coef=8, adjust="BH", number=Inf, sort.by="p")
V.MT24.top= topTable(efit2, coef=9, adjust="BH", number=Inf, sort.by="p")
V.TR24.top= topTable(efit2, coef=10, adjust="BH", number=Inf, sort.by="p")

toplist3hours <-list(V.DA.top[,c(1,3)],
                              V.DX.top[,c(1,3)],
                              V.EP.top[,c(1,3)],
                              V.MT.top[,c(1,3)],
                              V.TR.top[,c(1,3)])
names(toplist3hours) <- c("Daunorubicin-3h","Doxorubicin-3h", "Epirubicin-3h","Mitoxantrone-3h", "Trastuzumab-3h")

#toplist3hours <- map_df(toplist3hours, ~as.data.frame(.x), .id="id")

FCmatrix <- subset(efit2$coefficients)
is.matrix(subset(FCmatrix))
colnames(FCmatrix) <- c("Daunorubicin-3h","Doxorubicin-3h", "Epirubicin-3h","Mitoxantrone-3h", "Trastuzumab-3h","Daunorubicin-24h","Doxorubicin-24h", "Epirubicin-24h","Mitoxantrone-24h", "Trastuzumab-24h")



 

 
corrplot(FCmatrix, order= "hclust", iscorr=FALSE, col= "RdBu")

corrFC <- cor(FCmatrix)
corrplot(corrFC, method = 'square', order = 'AOE', addCoef.col = 'black')
corrplot(corrFC,type = 'full', order = 'hclust', tl.col = 'black',
         cl.ratio = 0.2, tl.srt = 45, col = COL2('PuOr', 20), col.lim=c(-0.1,1), addrect=3)


pheatmap::pheatmap(corrFC, display_numbers = TRUE, number_format = "%.2f")




#making of the side columns for class
```
##CVRI poster figures

```{r Poster DEG GOBP}
DDEdeg <- DDEtable24 %>% dplyr::filter(source=="GO:BP") %>% 
    dplyr::select(p_value,term_name,intersection_size) %>%
    slice_min(., n=6 ,order_by=p_value) %>%
    mutate(log_val = -log10(p_value)) %>%mutate(order_val=rev(as.numeric(rownames(.)))) %>% slice_max(., n=10, order_by=order_val)

DDEMdeg <-  DDEMtable24 %>% dplyr::filter(source=="GO:BP") %>% 
    dplyr::select(p_value,term_name,intersection_size) %>%
    slice_min(., n=10 ,order_by=p_value) %>%
    mutate(log_val = -log10(p_value)) %>%mutate(order_val=rev(as.numeric(rownames(.))))
DAsp3deg <- DAsptable3%>% 
  dplyr::filter(source=="GO:BP") %>% 
    dplyr::select(p_value,term_name,intersection_size) %>%
    slice_min(., n=10 ,order_by=p_value) %>%
    mutate(log_val = -log10(p_value)) %>%
  mutate(order_val=rev(as.numeric(rownames(.))))
MTsp24 <- MTsptable124 %>% 
  dplyr::filter(source=="GO:BP") %>% 
  dplyr::select(p_value,term_name,intersection_size) %>%
  slice_min(., n=10 ,order_by=p_value) %>%
  mutate(log_val = -log10(p_value)) %>%
  mutate(order_val=rev(as.numeric(rownames(.))))





# GODEGposter <-list("DAsp3deg"=DAsp3deg,"DDEdeg"=DDEdeg,"DDEMdeg"=DDEMdeg)
GODEGposter <-list("MTsp24"=MTsp24,"DDEdeg"=DDEdeg,"DDEMdeg"=DDEMdeg)
GODEGpostertable <- data.table::rbindlist(GODEGposter, idcol = "deg")
facet_degp=(
   c(MTsp24 = expression(atop("Mitoxantrone-specific 24 hours",~italic("n = 126"))),
    DDEdeg=expression(atop("Shared Anthracycline genes 24 hours",~italic("n = 4400"))),
     DDEMdeg=expression(atop("Shared Top2"*beta*"i genes 24 hours",~italic("n = 952")))))

DEG_posterp <- GODEGpostertable%>% mutate(deg = fct_inorder(deg)) %>% 
  ggplot(., aes(x = log_val, y=reorder(term_name, order_val, desc=TRUE),col=deg)) +
    geom_point(aes(size = intersection_size, col=deg)) +
  scale_y_discrete(labels = wrap_format(40))+
   scale_color_manual(values=c("#3386DD","red3", "orange"))+
   facet_grid(deg ~.,scales="free_y" , labeller=label_bquote(.(as.expression(
               eval(parse(text = paste0('facet_degp', '$`', deg, '`')))
               )))) +
  guides(color= "none", size=guide_legend(title = "# of intersected \n terms"))+
    ggtitle('Top 10 enriched GO:BP terms\n in specific overlaping sets') +
     xlab(expression(" -"~log[10]~"(adj. p-value)"))+
      ylab("GO:BP term")+
      theme_bw()+
 theme(plot.title = element_text(size = rel(1.5), hjust = 0.5),
        axis.title = element_text(size = 15, color = "black"),
        axis.ticks = element_line(linewidth = 1.5),
        axis.line = element_line(linewidth = 1.5),
        axis.text = element_text(size = 8, color = "black", angle = 0),
        #strip.text.x = element_text(size = 15, color = "black", face = "bold"),
        strip.text.y = element_text(color = "black"))

 
 g <- ggplot_gtable(ggplot_build(DEG_posterp))
stripr <- which(grepl('strip-r', g$layout$name))
fills <- c("#3386DD","red3", "orange", alpha= 0.8)
k <- 1
for (i in stripr) {
  j <- which(grepl('rect', g$grobs[[i]]$grobs[[1]]$childrenOrder))
  g$grobs[[i]]$grobs[[1]]$children[[j]]$gp$fill <- fills[k]
  k <- k+1
}
 

plot_grid(g)

```


```{r gene reads overall, eval=FALSE, include=FALSE}



### list of genes I want to see expression from
#sample(complete$ENTREZID, 3)
library(readr)
# ensembl <- useMart("ensembl", dataset="hsapiens_gene_ensembl")
my_attributes <- c('entrezgene_id', 'ensembl_gene_id', 'hgnc_symbol')
Response_24list <- c(1119,2950,6556)
Resp24_list <- getBM(attributes=my_attributes,
                     filters ='entrezgene_id',
                     values = Response_24list, 
                     mart = ensembl)
lookupforlevel <- getBM(attributes=my_attributes,
                     filters ='entrezgene_id',
                     values =renamelvl, 
                     mart = ensembl)
lookupforlevel <- lookupforlevelunique()
NRresp_list <- getBM(attributes=my_attributes,
                     filters ='entrezgene_id',
                     values = exampNR, 
                     mart = ensembl)

exampNR <- c(4287,56946,4478)

#find the "indecies for those genes in the matrix"
exampNR,x
counts24hours <- read.csv("data/filtered_cpm_counts.csv", row.names=1)
counts24da<- read_csv("data/Da24counts.txt")
counts24dx<- read_csv("data/Dx24counts.txt")
counts24ep<- read_csv("data/Ep24counts.txt")
counts24mt<- read_csv("data/Mt24counts.txt")
counts24tr<- read_csv("data/TR24counts.txt")


levels(counts24hours$Samples) <- c("Da.24h"="Daunorubicin",
                                           "Dx.24h"="Doxorubicin",
                                           "Ep.24h"="Epirubicin",
                                           "Mt.24h"="Mitoxantrone",
                                         "Tr.24h"="Trastuzumab",
                                         "Ve.24h"="Vehicle") 
 counts24hours %>% dplyr::filter(ENTREZID %in% Response_24list) %>%
  group_by(Samples) %>%
  ggplot(.,aes(x= Samples, y=Counts))+
  geom_boxplot()+
 
  facet_grid(~ENTREZID,labeller = cols= 2, scales ="free")
idx <- match(Emmalist$ENTREZID,counts24hours$ENTREZID)


namelist <- c("1026"="CDKN1A","23411"= "SIRT1","27113"= "BBC3","4193"= "MDM2")
Emmalist <- select(org.Hs.eg.db, 
       keys = Emmalist,
       columns = c("ENTREZID", "SYMBOL"),
       keytype = "SYMBOL")






counts24hours %>% dplyr::filter(ENTREZID %in% Emmalist$ENTREZID) %>%
  group_by(Samples) %>%
  ggplot(.,aes(x= Samples, y=Counts))+
  geom_boxplot(aes(col=Samples))+
  facet_grid(~ENTREZID,labeller=labeller(ENTREZID = namelist))
  
#                       

getBM(attributes=my_attributes,
                     filters ='entrezgene_id',
                     values = Response_24list, 
                     mart = ensembl)
```








# Knowles Comparison data:
  
  pulled from 2018 eLife, supplementary 5
  
  First I will get my data organized into several sets.
  
  Set 1: 'Dx' -Doxorubicin 24 hour significant DEGs p<0.1  
  Set 2: 'DDE' -Union of Daunorubicin, Doxorubicin, Epirubicin 24 hour significant DEGs p<0.1
  Set 3: 'DDEM'-Union of Daunorubicin, Doxorubicin, Epirubicin, and Mitoxantrone 24 hour significant DEGs p<0.1
  set 4: 'NoResp' - all genes at 24 hours that are not expressed significantly in any treatment  
  set 5: 'TOP2Bi' -cormotif intersection of 24 hour samples compared to Vehicle with p>0.5
  set 6: Knowles list from supp 5 in 2018 eLife paper  eQTL
```{r  knowles data comparisons}
library(biomaRt)
# library(org.Hs.eg.db)
# library(Homo.sapiens)
# library("AnnotationHub")
# library(EnsDb.Hsapiens.v79)
# if (!require("BiocManager", quietly = TRUE))
#     install.packages("BiocManager")
#entrez <- useMart("entrez",dataset="hsapiens_gene_entrez")
ensembl <- useMart("ensembl", dataset="hsapiens_gene_ensembl")
my_chr <- c(1:22, 'M', 'X', 'Y')  ## creates a filter for each database
attributes <- listAttributes((ensembl))

Knowles_2018.elife.33480.supp5.v2 <- read.delim("~/Ward Lab/Cardiotoxicity/Manuscript/Knowles_2018-elife-33480-supp5-v2/Knowles_2018-elife-33480-supp5-v2")
# Knowles_2018.elife.33480.supp6.v2 <- read.delim("~/Ward Lab/Cardiotoxicity/Manuscript/Knowles_2018-elife-33480-supp6-v2/Knowles_2018-elife-33480-supp6-v2")
# # #  read in supp file
# supp5list <- Knowles_2018.elife.33480.supp5.v2[,1]
# supp6list <- Knowles_2018.elife.33480.supp6.v2[,1]

my_attributes <- c('entrezgene_id', 'ensembl_gene_id', 'hgnc_symbol')
knowles5 <- Knowles_2018.elife.33480.supp5.v2[,1]

knowles5 <- getBM(attributes=my_attributes,filters ='ensembl_gene_id',
                 values = knowles5, mart = ensembl)
length(unique(knowles5$ensembl_gene_id))#376+
length(unique(knowles5$entrezgene_id))#377
# 
# in_common24AC <- intersect(sigVDA24$ENTREZID,sigVDX24$ENTREZID)
# in_common24AC <- intersect(in_common24AC,sigVEP24$ENTREZID)
# AC24hoursig <- (unique(in_common24AC))  ### 5400 total genes from all 24 h AC
# AConly24hsig <- getBM(attributes=my_attributes,filters ='entrezgene_id',
#                   values = AC24hoursig, mart = ensembl)
# length(unique(AConly24hsig$ensembl_gene_id))#5736
# length(unique(AConly24hsig$entrezgene_id))#5245

totalgenes <- getBM(attributes=my_attributes,filters ='entrezgene_id',
                  values = backGL$ENTREZID, mart = ensembl)
length(unique(totalgenes$entrezgene_id))#14184
length(unique(totalgenes$ensembl_gene_id))#15596
DX24Respset<- intersect(totalgenes$ensembl_gene_id, sigDox24$ensembl_gene_id)
length(unique(DX24Respset))  #7182 (#185 in common with knowls5)
15596-7182  #=8414

comp1 <- list(knowles5$entrezgene_id,top2bi$entrezgene_id)
  ggVennDiagram(comp1,
              category.names = c("Knowles", "Top2Bi-24hours"),
              show_intersect = FALSE,
              set_color = "black",
              label = "both",
              label_percent_digit = 1,
              label_size = 3,
              label_alpha = 0,
              label_color = "black",
              edge_lty = "solid", set_size = )+
  scale_x_continuous(expand = expansion(mult = .21))+
  scale_fill_gradient(low = "red", high = "yellow")+
  labs(title = "Comparision Knowles v 24h", caption = "n = 376")+
  theme(plot.title = element_text(size = rel(1.5), hjust = 0.5))
  

  
  
  
  
  
  

```
  
  
  
  
  
  
  Seone, Jose  Chromatin gene comparison:   comes from supp data NAT. MED 2019



```{r gene comparison, eval=FALSE, include=FALSE}
library(readxl)

# Seoane_chromatinregs <- read_excel("C:/Users/renee/Downloads/Supplements folde manuscriptr/NIHMS1539805-supplement-SuppTables.xlsx", 
#     range = "A13:H469")

#write.csv(Seoane_chromatinregs, "data/Seonane2019supp1.txt")

chrom_reg_Seoane <- read_csv(file = "data/Seonane2019supp1.txt",col_types = cols(...1 = col_skip()))
                            
Seoane_2019 <- chrom_reg_Seoane[,2]
names(Seoane_2019) <- "ENTREZID"
my_attributes <- c('entrezgene_id', 'ensembl_gene_id', 'hgnc_symbol')
#genesof interest: 
GWASsnps <- c('RARG', 'HIVEP1', 'ITGB7',  'PRDM2', 'KAZN', 'GDF5', 'FRS2', 'HDDC2', 'EEF1B2', 'SLC28A3')
totalsnpgenes <- getBM(attributes=my_attributes,filters ='hgnc_symbol',
                  values = GWASsnps, mart = ensembl)



```