2d.Rmd

---
title: "2D Analysis Notebook"
output:
  html_document:
    df_print: paged
---
```{r}
library(tidyverse)
```

```{r}
UpGene <- read.csv("~/R/Alina_RNAseq/UpGene.csv", sep = "", stringsAsFactors = FALSE )
DownGene <- read.csv("C:/Users/kesha/Desktop/2danalysis/DownGenelist_infected_infectedvscontrol.csv )
```

```{r}
library(readxl)
effector <- as.data.frame(read_excel("allStrains_PresentAbsent_effectorlist_forKeshav.xlsx",
  col_types = c(
    "text", "numeric", "numeric", "numeric", "numeric",
    "numeric", "numeric", "numeric", "numeric", "numeric",
    "numeric", "numeric", "numeric", "numeric", "numeric"
  )
))

rownames(effector) <- effector[,1] # move effectors to rownames
effector <- subset(effector, select = - ...1) #remove effector symbol column
net_eff <- rowSums(effector, na.rm = TRUE, dims = 1)
effector$net_effector <- net_eff
head(effector)
```

```{r}
l2fctable_InfectedvsControl <- read.csv("~/R/Alina_RNAseq/InfectedVsControl/l2fctableSIGNIFICANT_InfectedvsControl_BL6.csv")

# filter the symbol column based on only the genes that are in UpGenes so that these are both significant and differentially regulated
# l2fctable_InfectedvsControl <- filter(l2fctable_InfectedvsControl, 
#                                       symbol %in% c(UpGene$UpGeneList, DownGene$x))
l2fctable_InfectedvsControl <- filter(l2fctable_InfectedvsControl, 
                                       symbol %in% UpGene$UpGeneList)
```

```{r}
rownames(l2fctable_InfectedvsControl) <- l2fctable_InfectedvsControl[, 2] # make the symbols as rownames
l2fctable_InfectedvsControl <- subset(l2fctable_InfectedvsControl, select = -c(X, symbol)) # remove the X and symbol columns
net_l2fc <- rowSums(l2fctable_InfectedvsControl, na.rm = TRUE, dims = 1) # sum up the values in each row for every gene to get net infected gene expression
l2fctable_InfectedvsControl$net_l2fc  <- net_l2fc

```
```{r}
library(stringr)
colnames(l2fctable_InfectedvsControl) <- str_remove_all(colnames(l2fctable_InfectedvsControl),
                                  pattern = "log2FoldChange_")

l2fctable_InfectedvsControl <- l2fctable_InfectedvsControl[order(l2fctable_InfectedvsControl$net_l2fc, decreasing = TRUE), ]
head(l2fctable_InfectedvsControl)
```

```{r}
#l2fc <- l2fctable_InfectedvsControl[ c(1:28, 227:254), ] 
# l2fc <- l2fctable_InfectedvsControl[ c(1:28, 227:254), ] 
```


```{r}
# l <- t(l2fctable_InfectedvsControl)
# e <- t(effector)
# data_frame_merge <- merge(l, e, by = 'row.names', all = TRUE)
```
```{r}
# cancor(l2fc$net_l2fc, effector$net_effector)
```


```{r}
# library(GGally)
# ggpairs(l2fc)
```

```{r}
# ggcorr(l2fc)
# ```
# 
# ```{r}
# library(CCA)
```
```{r}
mc <- matcor(t(l2fctable_InfectedvsControl), t(effector))

```

```{r}
mc$XYcor

```


```{r}
correlate <- mc$XYcor
```

```{r}
library(corrplot)
```

```{r fig.height=15, fig.width=15}
corrplot(correlate)
```


R Canonical Correlation Analysis


```{r}
# library(corrr)
# cc1 <- cc(l2fc, effector)

```

```{r}
# compute canonical loadings
# cc2 <- comput(l2fc, effector, cc1)
```