---
title: "Colourblind friendly plots"
author: "Luise A. Seeker"
date: "11/05/2021"
output: html_document
---
# Introduction
It is challenging to plot data of whole single cell RNA seq experiments, because
often many clusters are found and it is challenging to find colour palettes
that offer enough vlaues to allow separation of all clusters. Even more 
challenging is it to stick to colourblind-friendly palettes, because they 
usually only contain a few colours. 

Below, I am attempting to re-plot some of the Feature and umap plots in a 
colourblind-friendly way.



# Preparations
## Load libraries
```{r}
library(Seurat)
library(ggplot2)
library(colorBlindness)
library(reshape2)
library(here)

```


## Loaad data
```{r}
nad_ol <- readRDS(here("data",
                       "single_nuc_data",
                       "oligodendroglia",
                       "srt_oligos_and_opcs_LS.RDS"))

```


## Prepare colour palette
```{r, fig.width = 8, fig.height = 6}
mat <- matrix(1:81, nrow = 9, ncol = 9)
mat1 <- melt(t(mat[9:1, ]))
len <- length(Blue2DarkRed12Steps)-1
mat1$v2 <- cut(mat1$value,
              breaks = seq(0,ceiling(81/len)*len, 
                           length.out = len+1))
ht <- ggplot(mat1) + 
  geom_tile(aes(x=Var1, y=Var2, fill=v2)) + 
  scale_fill_manual(values=Blue2DarkRed12Steps) + 
  theme_bw()
# check the plot by CVD simulator
cvdPlot(ht)


```

```{r}
Blue2DarkRed12Steps

```


# Plot Feature Plots
```{r, fig.width = 8, fig.height = 2}

FeaturePlot(nad_ol, features = c("RBFOX1", "OPALIN"), order = T, 
            min.cutoff = "q1",
            max.cutoff = "q99", blend = T, 
            cols = c("#AAF7FF", "#A50021", "#290AD8"), 
            blend.threshold = 0)  &NoAxes()
```


```{r, fig.width = 8, fig.height = 2}

FeaturePlot(nad_ol, features = c("RBFOX1", "SPARC"), order = T, 
            min.cutoff = "q1",
            max.cutoff = "q99", blend = T, 
            cols = c("#AAF7FF", "#A50021", "#290AD8"), 
            blend.threshold = 0)  &NoAxes()
```


```{r, fig.width = 8, fig.height = 2}

FeaturePlot(nad_ol, features = c("OPALIN", "SPARC"), order = T, 
            min.cutoff = "q1",
            max.cutoff = "q99", blend = T, 
            cols = c("#AAF7FF", "#A50021", "#290AD8"), 
            blend.threshold = 0)  &NoAxes()
```


```{r, fig.width = 8, fig.height = 2}

FeaturePlot(nad_ol, features = c("AFF3", "LGALS1"), order = T, 
            min.cutoff = "q1",
            max.cutoff = "q99", blend = T, 
            cols = c("#AAF7FF", "#A50021", "#290AD8"), 
            blend.threshold = 0)  &NoAxes()
```


```{r, fig.width = 8, fig.height = 2}


FeaturePlot(nad_ol, features = c("AFF3", "FMN1"), order = T, 
            min.cutoff = "q1",
            max.cutoff = "q99", blend = T, 
            cols = c("#AAF7FF", "#A50021", "#290AD8"), 
            blend.threshold = 0)  &NoAxes()
```


```{r, fig.width = 8, fig.height = 2}
FeaturePlot(nad_ol, features = c("RBFOX1", "FMN1"), order = T, 
            min.cutoff = "q1",
            max.cutoff = "q99", blend = T, 
            cols = c("#AAF7FF", "#A50021", "#290AD8"), 
            blend.threshold = 0)  &NoAxes()
```


```{r, fig.width = 8, fig.height = 2}
FeaturePlot(nad_ol, features = c("SPARC", "FMN1"), order = T, 
            min.cutoff = "q1",
            max.cutoff = "q99", blend = T, 
            cols = c("#AAF7FF", "#A50021", "#290AD8"), 
            blend.threshold = 0)  &NoAxes()


```

```{r, fig.width = 8, fig.height = 2}
FeaturePlot(nad_ol, features = c("SLC22A3", "PAX3"), order = T, 
            min.cutoff = "q1",
            max.cutoff = "q99", blend = T, 
            cols = c("#AAF7FF", "#A50021", "#290AD8"), 
            blend.threshold = 0)  &NoAxes()


```
```{r, fig.width = 8, fig.height = 2}
FeaturePlot(nad_ol, features = c( "MPZ","PMP2"), order = T, 
            min.cutoff = "q1",
            max.cutoff = "q99", blend = T, 
            cols = c("#AAF7FF", "#A50021", "#290AD8"), 
            blend.threshold = 0)  &NoAxes()


```
The base colours above should be colour-blind, friendly, but I am not so sure
about the blended colours.

```{r}

DimPlot(nad_ol, group.by = "ol_clusters_named", cols = Blue2DarkRed12Steps,
        label = T)



```

```{r, fig.width = 10, fig.height = 3}

DimPlot(nad_ol, group.by = "ol_clusters_named", cols = Blue2DarkRed12Steps,
        split.by = "Tissue",
        label = T,
        pt.size = 1)



```

# Session info

```{r}

sessionInfo()
```