5. Genomic Data Analysis

Question 1

how can cell types be identified

Answer 1

• physical appearance
• presence/absence of surface proteins
• isolation of cells and profiling of individual characteristics using sequencing technologies

Question 2

how is single-cell RNA sequencing conducted

Answer 2

quality control (outlier removal) –> normalisation –> feature selection –> dim. reduction –> cell-cell distance –> unsupervised clustering

Question 3

what is the purpose of quality control in single-cell rna sequencing

Answer 3

find unreliable cells, possible doublets

Question 4

what is the purpose of feature selection & dim reduction in single-cell rna sequencing

Answer 4

find the most informative genes and strongest signals from background noise & decreasing processing time

Question 5

what is the purpose of cell-cell distance in single-cell rna sequencing

Answer 5

to assist with clustering algos

Question 6

what is the process of clustering

Answer 6

feature selection/dim reduction (optional) –> clustering algo design/selection –> cluster validation –> results interpretation

Question 7

how are clusters validated

Answer 7

adjusted rand index - measures the similarity between clusters

visual inspection

cluster magnitude/cardinality

downstream performance

Question 8

how can we improve clustering

Answer 8

• try scaling data
• try another similarity measure
• check assumptions of algo match data distribution

Question 9

how can we check the similarity measure

Answer 9

manually select known distant examples, and similar examples, and determine whether the distance metric is conducive

Question 10

how can we check that we have the optimum number of clusters

Answer 10

plot loss v number of clusters
trial & error

Question 11

what are some clustering algo categories

Answer 11

• centroid based (kmeans)
• connectively based
• density based (GMM)
• hierarchical
• distribution based (DBSCAN)

Question 12

what is centroid based

Answer 12

• fast and efficient
• separate datapoints by multiple centroids and squared distance of data points from them

Question 13

what is the kmeans algo

Answer 13

• pick k, number of clusters
• pick centroids (random points in space)
• assign each data point to centroid
• update centroids by mean loc of their points
• stop when centroids don’t move much else repeat

Question 14

what are the clustering distances

Answer 14

euclidean, manhattan, minowski, hamming

Question 15

what is euclidean

Answer 15

straight line, numeric data only

Question 16

what is manhattan

Answer 16

distance along each axis/feature e.g. walking around city blocks, numeric data only

Question 17

what is minowski

Answer 17

generalised form of manhattan & euclidean
uses higher orders to further exacerbate dissimilarities
only for numeric data

Question 18

what is hamming

Answer 18

count similarities in each feature
accommodates categorical

Question 19

what are the pros & cons of k-means

Answer 19

• can get stuck in local optimum
• clusters are spherical
• can be tripped by outliers
• numerical only
• simple
• quick

Question 20

what are the two approaches to hierarchical clustering and how do they work

Answer 20

agglomerative - bottom up, start with many clusters & merge together in a tree based hierarchy

divisive - start from the top and break into smaller clusters, same tree based hierarchy

Question 21

what are the pros and cons of hierarchical clustering

Answer 21

• no assumptions on cluster number
• can correspond to meaningful taxonomies
• once a decision is made to combine two clusters, it cant be reversed
• slow on large datasets

Question 22

what is DBSCAN

Answer 22

density based spatial clustering of applications with noise

radius of original points (similar to centroids are user defined)

Question 23

what is the DBSCAN algorithm

Answer 23

• pick the radius that will define core points
• pick a random core point and find points that fall within it’s radius, these are core points
• continue to iterate across the data points finding core points that fall within the radius of other core points to finish the cluster
• then find non-core points which are close to core points. don’t use non-core points to further extend the cluster
• first cluster is finished. repeat for other clusters

Question 24

what is the silhouette coefficient

Answer 24

an alternative to the elbow method

pick a range of values for K (e.g. 1-10)

for each point calculate silhouette CE
a(i) = distance from point i to every other point in cluster
b(i) = distance from point i to every other point in data

goal is a(i) < b(i) to
s(i) = b(i)-a(i)/ larger of b(i) and a(i)

worst silhouette CE = -1, best = 1

every cluster will have it’s own silhouette plot, and the average is taken of the SCI over each cluster