HC 6 - Analysis of Transcriptomics Data - Part 2: Single-cell RNA-seq

Question 1

Experimental design and data collection in scRNA-seq

Answer 1

Frame the biological question: characterization of unknown samples of e.g. intestinal epithelial cells and their gene signatures
> 3’-droplet-based scRNA-seq

Question 2

Quality control of scRNA-seq raw data

Answer 2

Phred scores

Question 3

During read mapping and expression quantification, it is important to deal with barcodes and UMIs. What are those?

Answer 3

-Barcodes: reads unique for which cell/sample
-UMI: reads unique for each molecule (is the read unique or PCR duplicate) > count the UMIs per gene for quantification, these are different unique reads not due to PCR errors

Question 4

Fast mapping procedures

Answer 4

-Mapping on a transcriptome downloaded from database
-Unspliced alignment: Bowtie2

Question 5

After Phred quality control and read mapping and expression quantification you end up with the expression matrix which needs specific quality control: remove low quality cells. On which four QC metrics is the cell QC performed?

Answer 5

-Number of counts per barcode (cell) (count depth) > set a cutoff value for maintaining cell data
-Number of genes per barcode/cell
-Fraction of counts from mitochondrial genes per barcode/cell: the cells are damaged when low mitochondrial genes
-Fraction of ERCC spike-ins, if present

Question 6

The expression matrix often needs batch correction after cell QC. How?

Answer 6

Using PCA > plotting all cells as dots in PCA plot
-batch effect is not very large if the set PCs describe low percentage of variance
-the biological variation across groups is confounded with technical variation from processing cells in different batches
-batch effect screws up the result

Question 7

After the corrections, normalization of the scRNA-seq expression data is needed. What are the problems?

Answer 7

-many zero’s
-high variability
-often Funky Genes

Question 8

Normalization by deconvolution workflow scRNA-seq

Answer 8

-Cluster cells together
-Pool the cells per cluster to increase counts and reduce zero’s
-Robust estimate of each pool size factor
-Repeat for multiple pools
-Solve linear system of equations to obtain per-cell size factor

Question 9

Possible exam question: what is the difference in normalization with bulk RNAseq and scRNA seq?

Answer 9

In scRNA-seq: not a general scaling factor is used, but clusters of cells are taken and pooled to obtain estimate cell-specific normalization factors

Question 10

Imputation is performed after normalization. What is the problem?

Answer 10

Too many zero’s

Question 11

What are dropout events and where are they found in the plot of log(RPM)

Answer 11

Dropout event occurs when transcript is expressed but is entirely undetected in its mRNA profile
> 0-value in logRPM plot

Question 12

Why does a dropout event occur?

Answer 12

Due to low amounts of mRNA in individual cells, and the low sequencing depths typical for scSeq experiments

Question 13

The frequency of dropout events depends on the scRNA-seq …

Answer 13

protocols

Question 14

10X genomics (droplet based) has got generally many dropouts. What is the trade-off?

Answer 14

For the same budget, it measures more cells, but with less sequencing depths and more dropouts

Question 15

What is imputation?

Answer 15

The process of filling in the zero’s with expression values with the information from other cells that are ‘similar’

Question 16

What is a statistical artifact?

Answer 16

An interference which causes bias/manipulation of data

Question 17

Imputation: zero inflation

Answer 17

Log(count+1) transformation
> creates bias in analysis
> needed because zero’s in data cannot be log transformed

Question 18

Why shouldn’t all zeros be imputed?

Answer 18

A zero can mean biological variance

Question 19

There are different approaches for imputation. Why is it not ideal to impute all gene expressions? Give 2 reasons.

Answer 19

-Imputing expressions unaffected by dropout would introduce new bias
-Could also eliminate meaningful biological variation

Question 20

Why is it appropriate to treat all zero expressions as missing values?

Answer 20

-Some zero expressions may reflect true biological non-expression
-Zero expression can be resulted from gene expression stochasticity (fluctuation)

Question 21

What do the imputation methods search for?

Answer 21

Comparable genes which make it possible to fill in the missing values

Question 22

scImpute

Answer 22

-For each gene: to determine which expression values are most likely affected by dropouts
-For each cell: to impute the highly likely dropout values by borrowing information from the same genes’ expression in similar cells

Question 23

Procedures for imputation are based on … from similar cells/genes

Answer 23

borrowing

Question 24

Goals scRNA-seq data analysis

Answer 24

Identify clusters of cells with similar expression patterns, characterize these clusters with marker genes

Question 25

Feature selection: what is it?

Answer 25

What are the relevant genes for clustering cells as subpopulations?

Question 26

What do feature selected genes have to be?

Answer 26

Biologically active

Question 27

Assumptions feature selection

Answer 27

-Variable genes are biologically active
-Variation in expression for most genes is driven by uninteresting processes like sampling noise

Question 28

How can the technical component of variation be found under the assumption of variation in most genes is noise?

Answer 28

> the fitted value of the trend represents an estimate of its uninteresting variation (technical component)

Question 29

How to find biological component variation

Answer 29

For each gene the difference between its total variance and technical component
(loess)

Question 30

It is assumed that all genes are somewhat variable, but for most genes because of …

Answer 30

Randomness > Poisson

Question 31

Clustering

Answer 31

Identifying cells which are somewhat similar in gene expression > subpopulation
- Because, you have analyzed many different cells individually but don’t know which sample is what

Question 32

Unsupervised clustering : Hierarchical clustering is based on

Answer 32

Samples or genes

Question 33

Euclidian distance

Answer 33

How dissimilar genes/samples 1 and 3 for example
> calculate distance across all samples (for gene) or vice versa
> square those
> sum all these and put in sqrt

Question 34

Similarity and distance in comparing genes

Answer 34

Correlation: simialrity
Distance: 1 - correlation

Question 35

Positive and negative correlation lines

Answer 35

+ /
- \

Question 36

Workflow Hierarchical clustering

Answer 36

1. Set up a matrix with all possible distances between all data points
2. e.g. 5 genes in one dimension
3. Agglomerative method: start with each datapoint in one cluster
4. What is the smallest smallest distance
5. Attach these points and put the distance on the vertical (boxes with connecting lines plot)
6. Recalculate distance matrix (linkage: distance of newly formed cluster to other data points)
7. What is the shortest distance
8. Attach these points and put on vertical
9. Recalculate distance matrix
10. Shortest distance, attachment, vertical
11. etc.

Question 37

The height of the branches of the hierarchical clustering are proportional to the…

Answer 37

distance between genes

Question 38

Distance measures

Answer 38

-Euclidean distance
>Always => 0
>Zero for identical profiles
>high for profiles of little similarity
-City Block distance
>large effects in single dimension are dampened
-Pearson Correlation
>for centered (equal mean and sd) data
>correlation coefficients from -1 to 1
>clustering on |r| will put anti-correlated and correlated in one cluster

Question 39

Spearman correlation is more robust against .. than Pearson

Answer 39

outliers

Question 40

Single linkage vs complete linkage for recalculating distance matrix in HC

Answer 40

-Single linkage: distance between two clusters is that between the nearest points > result: chaining (genes added to clusters one at a time)
-Complete linkage: based on furthest points > result: small compact clusters, not suited for fuzzy (vaag) data
-Many other methods like average linkage exist

Question 41

Hierarchical clustering: isomorphism

Answer 41

-Horizontal order of genes is non-informative
-Each time a node is drawn, a decision is made where to put it
(vertical distances are informative! > distances)

Question 42

Unsupervised clustering: K-means: why

Answer 42

it is interesting to know which genes have similar expression profiles, these are maybe involved in similar biological processes

Question 43

K-means clustering groups genes by similarity in expression patterns. Name the algorithm steps

Answer 43

1. Choose K initial cluster centers at random
2. Partition objects (genes) into k clusters by assigning objects to the closest centroid
3. Calculate the centroid of each of the k clusters.
4. Assign each object to cluster i, by first calculating the distance from each object to all cluster centers, choose closest.
5. If object changes clusters, recalculate the centroids
6. Repeat until objects not moving anymore.

Question 44

In K-means clustering, the dots are:

Answer 44

genes

Question 45

For K-means clustering, genes are plotted for expression in samples 1 and 2 for example. Per cluster: there is a ….

Answer 45

difference between sample 1 and 2

Question 46

We want to see structure or clusters in the data set: why might PCA not work well on very complex data sets?

Answer 46

PCA works if the first 2 PCs account for most of the variation and clustering in the data

Question 47

tSNE solution for PCA problem

Answer 47

tSNE takes high dimensional dataset and reduces it to low dimensional graph, that retains a lot of the original information

Question 48

tSNE is a … method

Answer 48

deduction-reduction method

Question 49

Steps tSNE visualization

Answer 49

1. Determine the similarity between all the points in scatter plot
2. Randomly project the data in the low dimensional space
3. Determine similarity between all the points on the line
4. Similarity determination > clustering on similarity matrix (scores)
5. Move the points (iterations) that the similarity matrix after randomness are similar to the first similarity matrix > adjusted points
6. tSNE moves points a little it at a time and takes direction that makes the random matrix more like the similarity matrix.

Question 50

A method that works similar like tSNE is

Answer 50

UMAP

Question 51

Differences tSNE and UMAP

Answer 51

o Interpretation of the distance between objects or clusters
o tSNE preserves local structure in the data
> when different cells lay close to each other the distance is correct
> large distances are not accurate in the plot
o UMAP claims to preserve local and most of the global structure in the data

Question 52

With tSNE you cannot interpret the distance between clusters A and B at different ends of the plot: this means

Answer 52

You cannot infer that these clusters are more dissimilar than A and C if C is closer than A in the plot.
> But: within A, the points closer to each other are more similar objects than those at different ends of cluster A.

Question 53

With UMAP you can interpret

Answer 53

-Distances between points within clusters
-Distances between points/clusters between clusters