Skip to content

Commit

Permalink
Reformat Exercise 1
Browse files Browse the repository at this point in the history
  • Loading branch information
@AshKernow
AshKernow committed Sep 28, 2023
1 parent 5a3bc6f commit 2ecc3d7
Show file tree
Hide file tree
Showing 2 changed files with 442 additions and 412 deletions.
204 changes: 114 additions & 90 deletions Markdowns/06_FeatureSelectionAndDimensionalityReduction.Rmd
View file
Original file line number Diff line number Diff line change
Expand Up @@ -15,6 +15,7 @@ library(scater)
library(scran)
library(PCAtools)
library(tidyverse)
library(patchwork)
knitr::opts_chunk$set(error=FALSE,
message=FALSE,
Expand Down Expand Up @@ -50,6 +51,7 @@ library(scater)
library(scran)
library(PCAtools)
library(tidyverse)
library(patchwork)
```

We will load the _SingleCellExperiment_ object generated in the
Expand Down Expand Up @@ -624,128 +626,144 @@ for the exercise. We want to achieve the following:
- Explore how the main tuneable parameter of the algorithm - perplexity -
affects the results

```{r exercise1, eval=FALSE}
# Exercise 1: t-SNE ----
##### Run t-SNE {-}

1. Add the t-SNE result to the reducedDim slot of the SCE object.
2. Name this reducedDim "TSNE_perplex50".
3. Set perplexity = 50 (which is the default if we don't specify it).
4. Run t-SNE based on the PCA we ran previously using the first 10 principal
components.
5. Plot the PCA results ussing `ggcells` and colour the points by *SampleName*.

# add the t-SNE result to the reducedDim slot of the SCE object
# we name this reducedDim "TSNE_perplex50"
# we set perplexity = 50 (which is the default if we don't specify it)
# we run t-SNE based on the PCA we ran previously
# we will use the first 10 principal components
```{r exercise1, eval=FALSE}
set.seed(123) # set a random seed to ensure reproducibility
sce <- runTSNE(sce,
name = "TSNE_perplex50",
perplexity = 50,
dimred = "PCA",
n_dimred = 10)
# Make a custom visualisation using ggcells
ggcells(sce, aes(x = TSNE_perplex50.1, y = TSNE_perplex50.2,
ggcells(sce, aes(x = TSNE_perplex50.1,
y = TSNE_perplex50.2,
colour = SampleName)) +
geom_point()
```

# Part A
# Re-run the algorithm but change the random seed number.
# Do the results change dramatically between runs?
FIXME

# Part B
# Instead of colouring by SampleName, colour by expression of known cell markers
# CD79A (B cells)
# CST3 (monocytes)
# CD3D (T cells)
# HBA1 (erythrocytes)
FIXME
##### Part A {-}

# Part C
# Facet these plots by SampleName to better understand where each marker is mostly expressed
FIXME
* Re-run the algorithm but change the random seed number and generate a plot of
the new reduction.
* Do the results change dramatically between runs?

# Part D
# Explore different perplexity values (for example 5 and 500)
# Do you get tighter or looser clusters?
FIXME
```{r exercise1a, eval=FALSE}
YOUR CODE HERE
```


<details><summary>Hint A</summary>
<details><summary>Hint</summary>

Look at the set.seed() function

</details>

<details><summary>Hint B</summary>
<details><summary>Answer</summary>

You can replace what we colour by with any of the gene names in our dataset as they are stored as the rownames in our object.
```{r answer1a, eval=FALSE}
set.seed(321)
sce <- runTSNE(sce,
name = "TSNE_perplex50_seed321",
perplexity = 50,
dimred = "PCA",
n_dimred = 10)
ggcells(sce, aes(x = TSNE_perplex50_seed321.1, y = TSNE_perplex50_seed321.2,
colour = SampleName)) +
geom_point()
```

</details>

<details><summary>Hint C</summary>
##### Part B {-}

The function facet_wrap() can be used to modify ggplots as we did earlier.
Instead of colouring by *SampleName* colour by expression of known cell markers:

</details>
* CD79A (B cells)
* CST3 (monocytes)
* CD3D (T cells)
* HBA1 (erythrocytes)

<details><summary>Hint D</summary>
```{r exercise1b, eval=FALSE}
YOUR CODE HERE
```

You can replace values in the perplexity argument of the runTSNE() function.
<details><summary>Hint</summary>

You can replace what we colour by with any of the gene names in our dataset
as they are stored as the rownames in our object.

</details>

<details><summary>Answer</summary>

Here is the complete script:
e.g for CD79A:

```{r exercise1_solution, purl=FALSE, eval=FALSE}
# Run t-SNE ----
```{r answer1b, eval=FALSE}
ggcells(sce, aes(x = TSNE_perplex50_seed321.1,
y = TSNE_perplex50_seed321.2,
colour = CD79A)) +
geom_point()
```

# add the t-SNE result to the reducedDim slot of the SCE object
# we name this reducedDim "TSNE_perplex50"
# we set perplexity = 50 (which is the default if we don't specify it)
# we run t-SNE based on the PCA we ran previously
set.seed(123) # set a random seed to ensure reproducibility
sce <- runTSNE(sce,
name = "TSNE_perplex50",
perplexity = 50,
dimred = "PCA",
n_dimred = 10)
</details>

# Make a custom visualisation using ggcells
ggcells(sce, aes(x = TSNE_perplex50.1, y = TSNE_perplex50.2,
colour = SampleName)) +
geom_point()

# Re-run the algorithm but change the random seed number.
# Do the results change dramatically between runs?
set.seed(321)
sce <- runTSNE(sce,
name = "TSNE_perplex50_seed321",
perplexity = 50,
dimred = "PCA",
n_dimred = 10)
##### Part C {-}

ggcells(sce, aes(x = TSNE_perplex50_seed321.1, y = TSNE_perplex50_seed321.2,
colour = SampleName)) +
geom_point()
Facet these plots by SampleName to better understand where each marker is mostly
expressed

```{r exercise1c, eval=FALSE}
YOUR CODE HERE
```

# Modify the visualisation to colour the points based on logcounts of known cell markers
# CD79A (B cells)
# CST3 (monocytes)
# CD3D (T cells)
# HBA1 (erythrocytes)
ggcells(sce, aes(x = TSNE_perplex50_seed321.1, y = TSNE_perplex50_seed321.2,
colour = CD79A)) +
geom_point()
<details><summary>Hint</summary>

The function facet_wrap() can be used to modify ggplots as we did earlier.

</details>

# Facet these plots by SampleName to better understand where each marker is mostly expressed
<details><summary>Answer</summary>

```{r answer1c, eval=FALSE}
ggcells(sce, aes(x = TSNE_perplex50_seed321.1, y = TSNE_perplex50_seed321.2,
colour = CD79A)) +
geom_point() +
facet_wrap(~ SampleName)
```

</details>

# Explore different perplexity values (for example 5 and 500)
# Do you get tighter or looser clusters?
##### Part D {-}

Rerun the t-SNE using different perplexity values (for example 5 and 500)

* Do you get tighter or looser clusters?

```{r exercise1d, eval=FALSE}
YOUR CODE HERE
```

<details><summary>Hint</summary>

You can replace values in the perplexity argument of the runTSNE() function.

</details>

<details><summary>Answer</summary>

First re-run the t-SNE with different perplexity levels

```{r answer1d1, eval=FALSE}
set.seed(321)
sce <- runTSNE(sce,
name = "TSNE_perplex5",
Expand All @@ -757,22 +775,28 @@ sce <- runTSNE(sce,
perplexity = 500,
dimred = "PCA",
n_dimred = 10)
```

# visualise
ggcells(sce, aes(x = TSNE_perplex5.1, y = TSNE_perplex5.2,
colour = SampleName)) +
geom_point() +
labs(title = "Perplexity 5")
ggcells(sce, aes(x = TSNE_perplex50.1, y = TSNE_perplex50.2,
colour = SampleName)) +
geom_point() +
labs(title = "Perplexity 50")
ggcells(sce, aes(x = TSNE_perplex500.1, y = TSNE_perplex500.2,
colour = SampleName)) +
geom_point() +
labs(title = "Perplexity 500")
Then visualise using `ggcells`

```{r answer1d2, eval=FALSE}
tsne_5 <- ggcells(sce, aes(x = TSNE_perplex5.1,
y = TSNE_perplex5.2,
colour = SampleName)) +
geom_point() +
labs(title = "Perplexity 5")
tsne_50 <- ggcells(sce, aes(x = TSNE_perplex50.1,
y = TSNE_perplex50.2,
colour = SampleName)) +
geom_point() +
labs(title = "Perplexity 50")
tsne_500 <- ggcells(sce, aes(x = TSNE_perplex500.1,
y = TSNE_perplex500.2,
colour = SampleName)) +
geom_point() +
labs(title = "Perplexity 500")
```

Some things to note from our data exploration:
Expand Down
650 changes: 328 additions & 322 deletions Markdowns/06_FeatureSelectionAndDimensionalityReduction.html
View file

Large diffs are not rendered by default.

0 comments on commit 2ecc3d7

Please sign in to comment.