20.02.24 Calculations in Genetics/Risk assessment

Question 1

Who published Bayes theorem

Answer 1

-Reverend Thomas Bayes, in 1763

Question 2

When is bayesian analysis applied in Genetics

Answer 2

To calculate genetic risks in complex pedigrees and to calculate the probability of having or lacking a disease-causing mutation after a negative result is obtained.

Question 3

How is bayesian analysis performed

Answer 3

• A prior probability of an event, using new information, provides a revised posterior probability.
• Prior probability is based on pedigree information
• New information is the result of genetic testing
• New information can also include biochemical measurements (e.g. creatine kinase levels in Duchene muscular dystrophy)

Question 4

How is a bayesian calculation table set out

Answer 4

• Two columns for the two hypotheses being considered (carrier vs non-carrier). Mutually exclusive
• Prior probabilities when combined equal 1. Based on pedigree information
• Conditional probability. The probability of seeing the new information (negative result i.e. test sensitivity) if event 1 (carrier) or 2 (non-carrier) is true.
• Joint probability= multiply the prior and conditional.
• Posterior probability= divide the joint probability for event 1 by the total joint probability of event 1 and 2.

Question 5

How would you calculate someones risk, if their father is affected by an autosomal dominant condition that has reduced penetrance of 0.8

Answer 5

• Prior prob= 1/2
• Cond= 2/10
• Therefore final risk is 1/6

-Risk to his offspring is then= his risk x penetrance x 50% risk offspring will inherit. So 1/6 x 8/10 x 1/2= 1/15

Question 6

What are confidence intervals

Answer 6

• Used when testing a sample of a population, CIs give an indication of how uncertain we are about that measurement with regards to the true population value.
• CIs give a range of values in which we can be fairly confident (95% confident) that the true value lies.
• A smaller CI means the sample statistic represents the data well (often set at 95%).
• If the experiment was to be repeated 100 times and calculate the 95% CI each time, then 95% of the intervals would contain the population mean.

Question 7

How do you calculate a 95% confidence interval of a mean value

Answer 7

• sample mean +/- 1.96 x standard error

- Standard error= standard deviation/ square root of number of samples

Question 8

What are odds ratios

Answer 8

• Odds ratios are used to assess an association between exposure (aspect of medical history) and an outcome (disease/disorder), generally used in case-control studies.
• OR represents the relative odds than an outcome will occur given a particular exposed, compared to the odds of the outcome occurring in the absence of that exposure.
• If OR is 1 then exposure does not affect odds of outcome
• If OR is >1, exposure is associated with a higher odds of outcome.
• If OR is <1, exposure is associated with a lower odds of outcome.

Question 9

What is Sensitivity of a test

Answer 9

• the ability of a test to correctly identify individuals who are affected by a disease, (the true positive rate).
• number of true positives / (number of true positives+ false negatives)

Question 10

What is Specificity of a test

Answer 10

• the ability of a test to correctly identify individuals who are not affected by a disease (the true negative rate)
• number of true negatives/ (number of true negatives+false positives)

Question 11

What is Positive predictive value (PPV)

Answer 11

• The proportion of positive tests that are true positives

- number of true positives/ (number of true positives+false positives)

Question 12

What is negative predictive value (PPV)

Answer 12

• The proportion of negative tests that are true negatives

- number of true negatives (number of true negatives+false negatives).