computational precision medicine

Question 1

What can PCA be used for?

Answer 1

Visualization of multidimensional sample differences, Dimensionality reduction, Visualization of batch effects

Question 2

What can the units of the axes in a PCA plot be?

Answer 2

Raw component scores

Question 3

What are the units of gene quantification in RNA sequencing?

Answer 3

Transcripts per million, Fragments per kilobase per million, Number of reads mapped to a given transcript

Question 4

What is the distribution of RNA sequencing counts?

Answer 4

Negative binomial distribution

Question 5

What is the purpose of computational precision medicine?

Answer 5

Finding and evaluating new therapeutic targets, Supporting the practices of precision medicine with data analysis, Supporting of precision diagnostics by large-scale data analysis

Question 6

What is the difference between Pearson’s and Spearman’s correlation coefficients?

Answer 6

Pearson’s correlation evaluates the linear relationship between two continuous variables, whereas Spearman’s correlation evaluates the monotonic relationship

Question 7

What are batch effects?

Answer 7

Technical variance that we want to remove, Batch effects can introduce unwanted systematic variation in the data, which is unrelated to the biological factors of interest.

Question 8

How many dimensions are produced using PCA?

Answer 8

same as the number of features in the orignal space

Question 9

What is the purpose of the UCSC Xena project?

Answer 9

Enabling visual data analysis, A standardized pre-processing of mulitple omics data sets, A repository of gene expression data

Question 10

What is NOT a hallmark of cancer

Answer 10

Resistance to immunotherapy

Question 11

Why do you log2 transform data and what is meant by “count+1”?

Answer 11

when the numbers are too big, we transform it to see everything, and count+1 is just because we can’t log2 to zero

Question 12

What is RPKM?

Answer 12

reads per kilobase per million reads mapped, normalizes for gene length and sequencing depth, if high RPKM then we have high expression

Question 13

What is FPKM?

Answer 13

Fragments per kilo base of transcript per million mapped fragments, analouges to RPKM and used for paired end data, if high FPKM we have high expression

Question 14

What is TPM?

Answer 14

Transcripts per million fragments, also normalizes for gene lenght and sequencing depth and is better suited to compare expression between samples

Question 15

What can you not do for either RPKM, FPKM and TPM?

Answer 15

You cannot use for differential gene expression (DESEeq2/edgeR)

Question 16

What is the purpose of molecular subtyping cancer samples?

Answer 16

To classify cancer samples into distinct subgroups based on molecular characteristics

Question 17

In K-nearest neighbors (KNN) classification, what does the value of K represent?

Answer 17

The number of neighbors considered for classification

Question 18

In distance to centroid classification, how is the class label of a new data point determined?

Answer 18

By assigning the class label of the centroid closest to the data point

Question 19

Which methods is commonly used approaches for molecular subtyping of cancer?

Answer 19

Immunohistochemistry (IHC), Next-generation sequencing (NGS), Gene expression profiling

Question 20

What is the primary objective of single sample gene set enrichment analysis (ssGSEA)?

Answer 20

To determine the functional enrichment of gene sets in a single sample

Question 21

What are batch effects in gene expression data?

Answer 21

Systematic variations in gene expression attributed to technical factors

Question 22

When comparing gene expression data of cohorts of patients, why is it important to consider biological confounders?

Answer 22

Biological confounders can affect the reliability of outcome, lead to biased interpretation and affect reliability of gene expression measurements

Question 23

How is the distance between two ranked vectors typically measured?

Answer 23

Kendall’s tau distance

Question 24

Why am I teaching you to work with microarray data when RNA seq provides a more comprehensive gene expression profiling?

Answer 24

A lot of cancer subtyping schemes are defined on microarray data