Overview
This lecture introduces heat maps as a key visualization tool in differential gene expression analysis, explaining their function, interpretation steps, and practical examples.
Gene Expression Data Basics
- Gene expression data tables have genes as rows and samples (e.g., patients or cells) as columns.
- Each cell in the table typically contains a value reflecting gene expression (e.g., log2 fold change).
What is a Heat Map?
- A heat map visually represents gene expression data using colored tiles instead of numbers.
- Color intensity and hue indicate the level and direction of gene expression changes (e.g., red for upregulation, blue for downregulation, white for little change).
Heat Map Clustering and Interpretation
- Clustering reorders rows (genes) and columns (samples) to group similar expression patterns together.
- Dendrograms on the sides of the heat map show hierarchical relationships among genes and samples.
- Clustered heat maps can reveal which samples or genes form distinct groups, often reflecting biological categories (e.g., cancer vs. healthy tissue).
Steps to Interpret a Heat Map
- Step 1: Check the x-axis to identify what each column (sample) represents.
- Step 2: Check the y-axis to see what each row (gene) represents.
- Step 3: Understand the color scheme (e.g., red = upregulated, blue = downregulated).
- Step 4: Look for patterns or clusters that indicate biologically relevant groupings or trends.
Example: Real-Life Heat Map Analysis
- Sample heat map columns may represent individual cells, clustered by cell type.
- Rows may represent a subset of genes, such as cell type–specific markers.
- Clear patterns (e.g., clusters of red or blue) indicate groups of upregulated or downregulated genes corresponding to particular sample types.
Key Terms & Definitions
- Heat Map — A graphical representation of data where values are depicted by color.
- Gene Expression — The amount a gene is transcribed and translated in a cell or tissue.
- Log2 Fold Change — A way to quantify gene expression changes between conditions on a log2 scale.
- Clustering — Grouping similar genes or samples together based on expression patterns.
- Dendrogram — A tree diagram used to show hierarchical relationships from clustering.
Action Items / Next Steps
- Review heat map figures in recent RNA-seq publications.
- Practice interpreting heat maps using the four-step approach described.
- Prepare questions or topics for future lectures.