Executive Certificate in Applied Statistics for Precision Medicine

Programme Highlights

Precision medicine leverages on insights derived from the analysis of high-throughput datasets containing multiple parameters.

The Executive Certificate in Applied Statistics for Precision Medicine aims to provide foundational knowledge and skills for learners to apply appropriate statistical approaches in evaluating and analysing high-throughput datasets in precision medicine. Learners will develop the skills in designing experiments based on the statistical principles, evaluating data quality, analysing and interpreting high throughput data for precision medicine.

Benefits of Attending

Learners should able to:

• Explain principles of statistical modelling of data for precision medicine
• Explain the challenges of analyzing high throughput datasets for precision medicine
• Design high throughput experiments in precision medicine to minimize confounders and increase statistical power to detect differences
• Evaluate data quality (batch effects, missing data, noise) and perform preprocessing where appropriate (batch correction, imputation, normalization)
• Analyze high throughput data to find and evaluate differences in samples using appropriate statistical tools.

Course Agenda

1. Overview of R/Bioconductor
2. Overview of basic statistics
   a) Probability distributions (count and continuous) for biological data
   b) Hypothesis testing (frequentist and Bayesian approaches)
3. Challenges with datasets in precision medicine
   a) Noise/batch effect
   b) Multiple comparisons
   c) n << p
   d) Integration of different datasets
4. Statistical principles for high throughput data in precision medicine
   a) Design of high throughput experiments for precision medicine and sample size calculation
   b) Evaluation of data quality
        – Batch effect detection and correction
        – Missing data and imputation
        – Normalization approaches (e.g., quantile normalization/variance stabilization)
   c) Approaches for handling m << p datasets
        – Dimensionality reduction
        – Feature selection
        – Regularization
   d) Approaches for controlling for multiple hypothesis tests (e.g., FWER/FDR)
5. Case studies of analysis in precision medicine
   a) Variant calling analysis and prioritization
   b) RNA Seq analysis and gene set enrichment

Pre-Requisites and Assessment

Prerequisites:

• Applicants must have a Bachelor’s degree and may be required to show relevant work experience. Applicants with other qualifications and experience may be considered on a case by case basis, subject to approval by the school.
• Basic knowledge of R language.

Assessment:

• Project work

Course Fee