analyse.Rmd

# Pathogen detection in metagenomic datasets

Plot the hashed K-mer matches of the datasets you have found.

```{r mash}
library(ggplot2)
mash = read.table("/vol/data/output.tsv")

publications = read.table("/vol/data/publications.tsv")

ggplot(mash) + geom_histogram(aes(x=V3)) +
  xlab("Matched K-mer hashes") +
  ylab("Found datasets")
```

It is also important to check the mash identity.
Plot the number of matched k-mer hashes against the mash identity.

```{r scatter}
ggplot(mash) + geom_point(aes(x=V2, y=V3))
```

We are interested in matches with at least 90% mash identity and
700 matched k-mer hashes. Let's filter for those matches

```{r filter}
mash_filtered <- mash[mash$V2 >= 0.90 & mash$V3 >= 100,]
mash_filtered$V1
```

It would be good to know which environment your hits belong to.
We could join both datasets and plot the environments.

```{r join}
mash_publications <- merge(mash_filtered, publications, by="V1")
ggplot(mash_publications) + geom_bar(aes(x=V2.y)) +
  xlab("Environment") + ylab("Number of Hits") +
  theme(axis.text.x = element_text(angle = 60, vjust = 0.5, hjust=1))
```

Finally we should plot which pathogens we have found in which environment

```{r join_patho}
ggplot(mash_publications) +
  geom_bar(aes(x=V2.y, fill=V7.x)) +
  xlab("Environment") +
  ylab("Number of Hits") +
  theme(axis.text.x = element_text(angle = 60, vjust = 0.5, hjust=1)) +
  scale_fill_brewer(palette="Paired")
```