Session 4 - Screening and biochemical genetics Flashcards
Define Linkage
A statistical method to determine the link between a region of the genome and a given genotype in a family or group of affected patients.
It is performed on small sets of data, often large pedigrees, but can also be used for sib-pair analysis.
Used to identify genes with large effects, often Mendelian.
Define an Association study
A statistical method of determining association between SNPs and disease in large populations.
Performed on large cohorts in a case-control fashion.
Can identify genes with small effects
What are the advantages and disadvantages of Linkage analysis?
Advantages:
Localisation of disease region on a chromosome.
Multiple markers can be studied at once
Disadvantages:
Need multigenerational cases
Difficult to study complex traits, however data from multiple families can be combined to increase the power of the study
What is used to measure Linkage? What is the cut off?
LOD scores
no linkage -2 < indeterminate < 3 Linkage
Define linkage disequilibrium
The likelihood that two independent markers will be inherited together at a frequency greater than chance. This is different from Linkage.
What can cause linkage disequilibrium?
Natural selection
Chance
How is linkage disequilibrium expressed? How is it calculated?
D
D(ab) = P(ab) - P(a)P(b)
Alleles in complete LD have a score of 1.
What factors affect LD?
Recombination Gene conversion Selection Population structure New mutations Genetic drift Gene flow Population history.
What is the Transmission Disequilibrium Test (TDT)?
A method of linkage analysis that can be performed in trios or between sib-pairs. It involves calculating the likelihood of two affected individuals inheriting the same allele from parents. It is used to study complex disease traits in families. Overtransmission of an allele suggests disease susceptibility
Name a successful Linkage study
Identification of NOD2 in Crohn’s disease
Name a successful GWAS study
Identification of associations between genotypes and age-related macular degeneration
Name the 4 possible explanations for finding a positive association between variant and disease.
- There is a true association
- variant is in LD with the true variant.
- Association by chance
- Case-controls my be selected from different populations
Give some advantages and disadvantages of GWAS
Advantages:
No pre-hypothesis
Good for studying complex disease
Tend to have better genomic resolution that linkage studies.
Disadvantages:
Need large cohort of ethnically matched cases and controls
Cannot test causality
Can be expensive
Doesn’t test for CNVs
Large number of statistical tests means a very low P value is required.
What assumptions are made in order for GWAS results to be valid?
Case-controls are ethnically matched
Case participants reflect disease phenotype
Genomic and clinical data are collected in the same way in cases and controls.
How can study power be defined? What is it determined by?
The ability of a study design to pick up associations accurately.
The MAF of the risk allele, samples size, LD between true risk allele and another, genetic heterogeneity of the sample population, relative risk of the risk allele.
What are the two types of linkage analysis? What information do each of them require?
Parametric: Gene information, MOI, penetrance, allele frequency
Non-parametric: no data.
What type of linkage analysis can be used in consanguineous populations
Autozygosity mapping.
What are the problems encountered during LOD score analysis?
Vulnerable to lots of sampling and experiemntal errors
Computational limitations
Locus heterogeneity
Low resolution
Parameters need to be correctly spec’d
Low penetrance and phenocopies in a family.
What is the principle of non-parametric linkage analysis?
That the region of a genome in which disease-causing variants are located is more likely to be shared by affected individuals than would be expected by chance.
What must be distinguished in non-parametric linkage analysis?
Whether regions are identical by descent (the patients are related and the variant has been inherited from an ancestor), or if they are identical by state (the allele has entered the family twice - not identical due to ancestral inheritance)
The frequency of the allele in the population must be taken into account.
What can Affected Sib Pair analysis be used for?
To identify genes that cause multifactorial disorders in affected siblings.
Explain the process behind ASP analysis
Genotype affected siblings
Look for chromosomal regions where sharing of allele is above the expected mendelian values
Identify regions of the genome that are IBD
Re-examine region to narrow down candidate area
List some population genome studies. State their purpose
1000G - catalogue of human genetic variation - 26 populations ~2600 individuals
100KG - stimulate UK genomics, introduce genomics into healthcare, identify new disease, produce ethical programme based on consent.
COSMIC - Catalogue of Somatic mutations in Cancer
HapMap - Identify linked haplotypes (international)
ExAC - Aggregation of data from multiple large scale genome studies. 60706 individuals from multiple populations. Now GnomAD
EVS - combination of multiple US studies
The Cancer Genome Atlas - somatic mutations identified in different cancers. Tumour profiling and clinical information.
CHIP - identification of mutations normally associated with haem-onc that don’t have disease
PAGE - 1000 foetus/parent trios with abscan of NT>4mm, looking to develop good exome test for prenatal testing
DGV - database of CNVs and SVs in the normal population
DDD - 12,000 trios with ID
CR-UK stratified medicine programme - aim to combine genetic testing with trial enrollment. 2 stages. First stage looked at 10 genes in multiple tumours to ID markers for disease/therapy/prognosis. Stage 2 now looking at the Lung Cancer Matrix Trial of nsc lung cancers to identify genetic faults driving their disease.
ENCODE
What types of DTC testing are there?
Testing for predisposition to disease Dispositional health tests (warfarin dose, lifetime risk of dementia) Nutrigenomics Compatibility tests Paternity tests Related-ness tests Ancestry testing