---
title: "Working with matrix output by sourmash compare"
output: html_document
---

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

### Load a comparison matrix into R


```{r chunk1}

sourmash_comp_matrix <- read.csv("ecoli.cmp.csv")

# Label the rows
rownames(sourmash_comp_matrix) <- colnames(sourmash_comp_matrix)

# Transform for plotting
sourmash_comp_matrix <- as.matrix(sourmash_comp_matrix)

```

### Make an MDS plot

```{r mds}
fit <- dist(sourmash_comp_matrix)
fit <- cmdscale(fit)
x <- fit[, 1]
y <- fit[, 2]
plot(fit[ , 1], fit[ , 2], xlab = "Dimension 1", ylab = "Dimension 2")
```

### Make a tSNE plot

```{r tSNE}
library(Rtsne)
tsne_model <- Rtsne(sourmash_comp_matrix, check_duplicates=FALSE, pca=TRUE, perplexity=5, theta=0.5, dims=2)
d_tsne = as.data.frame(tsne_model$Y) 
plot(d_tsne$V1, d_tsne$V2)
```

### Make an unclustered heatmap

```{r heatmap}
heatmap(sourmash_comp_matrix, Colv=F, scale='none')
```

### Make a clustered heatmap

```{r clustered-heatmap}
hc.rows <- hclust(dist(sourmash_comp_matrix))
hc.cols <- hclust(dist(t(sourmash_comp_matrix)))
heatmap(sourmash_comp_matrix[cutree(hc.rows,k=2)==1,], Colv=as.dendrogram(hc.cols), scale='none')
```