05_iSEE_intro.Rmd

# iSEE introduction

Instructor: Leo

## Toy RSE data

```{r}
## Lets build a simple SummarizedExperiment object following information
## from the documentation
library("SummarizedExperiment")
## ?SummarizedExperiment

## Adapted from the official documentation:

## First we create the data pieces that we'll use to build our
## SummarizedExperiment object. In this case, we'll have 200 genes
## measured in 6 samples.
nrows <- 200
ncols <- 6

## Let's make up some count numbers at random
set.seed(20210223)
counts <- matrix(runif(nrows * ncols, 1, 1e4), nrows)

## Then some basic infomratino for our genes
rowRanges <- GRanges(
    rep(c("chr1", "chr2"), c(50, 150)),
    IRanges(floor(runif(200, 1e5, 1e6)), width = 100),
    strand = sample(c("+", "-"), 200, TRUE),
    feature_id = sprintf("ID%03d", 1:200)
)
names(rowRanges) <- paste0("gene_", seq_len(length(rowRanges)))

## Next, we create some information about samples
colData <- DataFrame(
    Treatment = rep(c("ChIP", "Input"), 3),
    row.names = LETTERS[1:6]
)

## Finally we put all these pieces together in a single R object
rse <- SummarizedExperiment(
    assays = SimpleList(counts = counts),
    rowRanges = rowRanges,
    colData = colData
)

## Overview
rse
```

## iSEE

How can you make plots from `SummarizedExperiment` objects without having to
write any code? The answer is with `iSEE` 🎨

* http://bioconductor.org/packages/iSEE
* http://bioconductor.org/packages/release/bioc/vignettes/iSEE/inst/doc/basic.html


```{r isee_basic, eval = FALSE}
## Let's explore the `rse` object interactively
library("iSEE")
iSEE::iSEE(rse)
```

## Exercise with data from spatialLIBD

* We'll download a `SingleCellExperiment` object, which is similar to `SummarizedExperiment` as it extends it.
    - http://bioconductor.org/packages/SingleCellExperiment
    - http://bioconductor.org/packages/spatialLIBD
    - https://doi.org/10.1038/s41593-020-00787-0
    - https://osca.bioconductor.org/
    - https://www.nature.com/articles/s41592-019-0654-x Figures 2 and 3

```{r download_sce_layer}
## Lets get some data using spatialLIBD
sce_layer <- spatialLIBD::fetch_data("sce_layer")
sce_layer

## We can check how big the object is with lobstr
lobstr::obj_size(sce_layer)
```

* Just like with our `rse` object, we can use `iSEE::iSEE()` to explore the data.

```{r explore_sce_layer, eval = FALSE}
iSEE::iSEE(sce_layer)
```

<style>
p.exercise  {
background-color: #E4EDE2;
padding: 9px;
border: 1px solid black;
border-radius: 10px;
font-family: sans-serif;
}
</style>

<p class="exercise">
**Exercise 1**:
Create a plot and download a PDF that reproduces as closely as possible the plot on the right side of the following slide. 
</p>

<script async class="speakerdeck-embed" data-slide="10" data-id="329db23f5f17460da31f45c7695a9f06" data-ratio="1.33333333333333" src="//speakerdeck.com/assets/embed.js"></script>

<p class="exercise">
**Exercise 2**:
Explore with a _heatmap_ the expression of the genes `MOBP`, `MBP`, and `PCP4`. If we use _clustering_ (group genes based on similar expression patterns), which two genes are most similar to each other?
</p>

<p class="exercise">
**Exercise 3**:
In which dorsolateral prefrontal cortex (DLPFC) layers (`L1`, `L2`, ..., `L6` grey matter layers, and `WM` for _white matter_) do we see the highest expression for the genes _MOBP_ and _MBP_?
</p>

This list of ENSEMBL IDs will be useful:

```
ENSG00000168314
ENSG00000183036
ENSG00000197971
```

## Community

`iSEE` authors:

* Kévin Rue-Albrecht https://twitter.com/KevinRUE67
* Federico Marini https://twitter.com/FedeBioinfo
* Charlotte Soneson https://twitter.com/CSoneson
* Aaron Lun https://twitter.com/realAaronLun
* Another example exploring data with `SummarizedExperiment` and `iSEE`:

<blockquote class="twitter-tweet"><p lang="en" dir="ltr">Today we explored RNA-seq data from <a href="https://twitter.com/StefanoBerto83?ref_src=twsrc%5Etfw">@StefanoBerto83</a> et al who made it easy to re-use. Thank you! ^^<a href="https://twitter.com/lcolladotor?ref_src=twsrc%5Etfw">@lcolladotor</a> used <a href="https://twitter.com/hashtag/shiny?src=hash&amp;ref_src=twsrc%5Etfw">#shiny</a> + <a href="https://twitter.com/hashtag/ggpubr?src=hash&amp;ref_src=twsrc%5Etfw">#ggpubr</a> as well as <a href="https://twitter.com/hashtag/iSEE?src=hash&amp;ref_src=twsrc%5Etfw">#iSEE</a><br><br>📔 <a href="https://t.co/iUQHE0xqRc">https://t.co/iUQHE0xqRc</a><br>🗞️ <a href="https://t.co/qhAdXbhY9c">https://t.co/qhAdXbhY9c</a><a href="https://twitter.com/hashtag/rstats?src=hash&amp;ref_src=twsrc%5Etfw">#rstats</a> <a href="https://twitter.com/Bioconductor?ref_src=twsrc%5Etfw">@Bioconductor</a><a href="https://t.co/OXTukByhoo">https://t.co/OXTukByhoo</a></p>&mdash; LIBD rstats club (@LIBDrstats) <a href="https://twitter.com/LIBDrstats/status/1360349352250789890?ref_src=twsrc%5Etfw">February 12, 2021</a></blockquote> <script async src="https://platform.twitter.com/widgets.js" charset="utf-8"></script>

<blockquote class="twitter-tweet"><p lang="en" dir="ltr">Are you making a heatmap to 👀 gene expression? Have you wondered whether to center &amp;scale the data?<br><br>I made this 5 min video to help answer these ❓<br><br>Shoutout to <a href="https://twitter.com/hashtag/iSEE?src=hash&amp;ref_src=twsrc%5Etfw">#iSEE</a> by <a href="https://twitter.com/KevinRUE67?ref_src=twsrc%5Etfw">@KevinRUE67</a> <a href="https://twitter.com/FedeBioinfo?ref_src=twsrc%5Etfw">@FedeBioinfo</a> <a href="https://twitter.com/CSoneson?ref_src=twsrc%5Etfw">@CSoneson</a> et al <a href="https://twitter.com/hashtag/rstats?src=hash&amp;ref_src=twsrc%5Etfw">#rstats</a> <a href="https://twitter.com/Bioconductor?ref_src=twsrc%5Etfw">@Bioconductor</a><a href="https://twitter.com/LieberInstitute?ref_src=twsrc%5Etfw">@LieberInstitute</a> <a href="https://t.co/KwQHLODTQV">https://t.co/KwQHLODTQV</a></p>&mdash; 🇲🇽 Leonardo Collado-Torres (@lcolladotor) <a href="https://twitter.com/lcolladotor/status/1598456517216595968?ref_src=twsrc%5Etfw">December 1, 2022</a></blockquote> <script async src="https://platform.twitter.com/widgets.js" charset="utf-8"></script>