W3 - Using The Result Of Genetic Studies

Question 1

What are the two large data analysis techniques?

Answer 1

 GWAS
– Identify loci in the genome associated with a phenotype/disease - identifies millions of SNPs then use stats for identification of disease to pinpoint loci.
 RNAseq
– Take a sample tissue and sequence all the RNA to provide a snapshot of gene expression levels

Question 2

What are the difficulties of using GWAS data?

Answer 2

 Large GWAS for complex diseases detect many loci
– Prioritisation - so thinking which area we need to focus on.
 90% of GWAS SNPs are in non-coding regions of the genome
– Causal genes?
 What is the mechanism of action explaining the association?
– Tissue/cell type?
– Molecular mechanism?

Question 3

What is Linkage Disequilibrium?

Answer 3

Where alleles (DNA markers) occur together more often than can be accounted for by chance because of their physical proximity on a chromosome.

This is to do with the crossing over and recombination. The further away the loci, the more chance of cross-overs and if close together, more chance of being inherited togethe. If it were in linkage equilibrium, there would be a 50% change of two loci being inherited at the same time.

Question 4

What is RNAseq data?

Answer 4

Looking to see which genes have more or fewer mRNA copies present.
 Relative expression data for every gene
 Need to set significance threshold
– P-value
– Fold change
 Novel, allele specific expression, and
alternative transcripts may be identified

Question 5

What is the osteocyte transcriptome?

Answer 5

 Same cell type at different locations show different transcriptomes
 >100 novel transcripts identified
 Distinct pattern of different expressed genes in different tissues/cell types

Question 6

What are the other applications of RNAseq?

Answer 6

 Cell populations response to treatments
 How gene expression changes through development or under disease conditions
 Single cell transcriptome analysis

Question 7

What are the difficulties of using RNAseq data?

Answer 7

 Many expression changes likely to be found
– Difficult to differentiate real from methodological artifacts
 Transcriptome is a snapshot of expression in a specific cell/tissue and at a specific time
 Identification of differential expression does not provide biological reasoning

Question 8

How useful are GWAS and RNA sequencing in when it comes to complex diseases and large data sets?

Answer 8

GWAS
 Identifies associations across
whole genome
 Large number of loci
 Doesn’t identify causal variants or
genes
 Doesn’t identify cell
type/tissue/developmental stage

RNA Sequencing
 Transcriptome of single cell/tissue
type
 Large number of differentially
expressed genes
 Misses changes in other cell types
or stages of development
 Doesn’t identify reason for
differential gene expression

Question 9

What is a pathway analysis?

Answer 9

You can take a list from a large data set and see which genes are enriched and related.

Question 10

What is personalised medicine?

Answer 10

 Applying the results of genetic studies to the healthcare management of an individual
– Predict and prevent disease
– Diagnosis
– Personalised interventions

Question 11

What is GWAS loci relating to biological reasons?

Answer 11

Links loci to disease traits.
-Causal mutation/gene for each loci
-Genes or pathways identifying with disease
-Prioritising what to investigate further
- Validating findings.

Question 12

What are difficulties linking loci to gene?

Answer 12

 Linkage Disequilibrium makes it difficult to distinguish causal variant
 90% of GWAS SNPs are in non-coding regions
– Regulatory elements
* Promoters, Enhancers, TF binding sites
 May act at a distance from effected gene(s)
 Need to determine relevant cells/tissues

Question 13

What is fine mapping?

Answer 13

 High resolution study of loci attempting to pinpoint individual variants directly effecting trait
 Statistical and probabalistic methods
or comparison to a SNP correlation reference panel

Question 14

How are causal genes assigned?

Answer 14

 Proximity
 Non-synonymous exonic change
 Chromatin conformation capture

Question 15

What are the relevant cell types for fine mapping?

Answer 15

 Often unclear what are the causal cells
 SNP enrichment analysis
– Gene expression
– Regulatory elements
– Open chromatin - likely to indicate genes actively expressed. So you can assign each gene a regulatory activity score.

Question 16

What are the basic stages of analysing GWAS?

Answer 16

 100s of GWAS loci, 90% in non-coding regions
 Fine mapping to attempt to define causal variants at loci
 Analysis of causal SNP location to predict causal gene
 Cell type SNP enrichment analysis to determine relevant cell types

Question 17

How can we combine genotype and expression data?

Answer 17

 Can we use gene expression data to help annotate our GWAS loci?
 eQTL - A locus that explains a fraction of the genetic variance of a gene
expression phenotype
 Colocalisation analysis – Compare the GWAS and eQTL at a locus to determine if they are due to the same causal variants
 TWAS -Transcriptome wide association studies

Question 18

What is co-localisation analysis?

Answer 18

GWAS and eQTL can be combined to pinpoint which gene and SNP is likely to associate with what you are looking for. You can compare if they are shown by the same signal or if they are two different signals that happens to be together.

Question 19

What is an explanation for locus overlap?

Answer 19

 GWAS and eQTL loci can overlap for 3 reasons:
 Independent causal variants
 A single causal SNP
 Pleiotropy

Question 20

What is a Transcriptome wide association study?

Answer 20

 Directly test for associations between gene expression levels and phenotypes. So this is basically combining GWAS and RNAseq at once.
– Overcomes most issues with GWAS and RNAseq
– Currently not generally feasible
 Gene expression is highly heritable
– Tissue specific eQTL maps
 You can use genotype to predict gene expression levels
– GWAS studies give us the SNP genotypes

Question 21

How do we prioritise results from GWAS?

Answer 21

We can use our knowledge of the genome and gene expression to prioritise loci/genes for further investigation
– eQTLs – expression quantative trait loci
– Colocalisation analysis to annotate causal genes
 TWAS to directly associate gene expression to trait phenotype

Question 22

How do you validate your results?

Answer 22

 Once you have prioritized genes for investigations what do you do?
– Cell Studies
– Functional Phenotyping
– High throughput screens

Question 23

How do you narrow in on interesting genes?

Answer 23

 GWAS looking at BMD
– 518 loci
– Fine mapping
– SNP annotation to assign causal genes
– Gene annotation to identify interesting results
– Dishevelled Associated Activator Of
Morphogenesis 2 (DAAM2)

Question 24

What is the functional analysis animal models?

Answer 24

 Knockout animals
– Total knockout
– Cell specific
– Inducible
– Gene editing

Question 25

What is in vivo functional analysis?

Answer 25

 Weaker bones are the relevant functional phenotype to osteoporosis

Question 26

What are reverse genetic screens?

Question 27

What are reverse genetics with modern tools?

Answer 27

 Classical reverse genetics
– Cloning to introduce mutations in target gene
– Observe phenotypic outcome
 Targeted genetics
– Gene targeting and trapping
– Gene editing

Question 28

What is the International Mouse Phenotyping Consortium?

Answer 28

 98% of human genes have a homolog in the
mouse genome
 IMPC is generating knockout mice strains for
each mouse gene
– ~20,000 genes
 All mice are same background, kept under the
same conditions
– Basic phenotyping
– Results freely available