Molecular Profiling and Biomarker Discovery
Microarrays are routinely used in genomic studies to detect changes in mRNA expression levels and have been key in developing biomarkers for several diseases. These experiments have fundamental statistical and data processing challenges associated with them. This course covers the statistical aspects of experimental design, biological and technical replicates, preprocessing, quality assessment, parametric and non-parametric statistical tests, multiple-hypothesis testing, P-value correction and false discovery rates, visualization techniques (e.g. heatmaps, volcano plots), and biological significance (e.g. functional annotation, pathways, hypergeometric tests, gene set enrichment). The course also covers the increasing role of molecular profiling in disease treatment, particularly in oncology.
At the end of this course, students will be able to:
Explain -omics technologies and biomarker study design.
Describe the different transcriptomic technologies and data availability.
Describe the role of mass spectrometry in omics analysis.
Design a data workflow to use raw or processed transcriptomic data using R/Bioconductor.
Apply inferential, exploratory and predictive statistical analysis of microarray data.
Apply functional annotation to transcriptomics data, understand the statistical procedures involved and the limitations.
Perform network analysis of -omics technology using R and Cytoscape.
Create a data workflow for non-coding RNASeq data.
Identify microRNA biomarkers from microarrays and RNASeq data.