Hierarchical Clustering
By casual inspection, we could surmise that:
- Gene C’s behavior is opposite that of Genes A, B, and D.
- Gene B and Gene D have the most similar behaviors.
| Gene A | Gene B | Gene C | Gene D | |
|---|---|---|---|---|
| Gene A | 1 | 0.450 | −0.633 | 0.597 |
| Gene B | 0.450 | 1 | 0.107 | 0.729 |
| Gene C | −0.633 | −0.107 | 1 | −0.454 |
| Gene D | 0.597 | 0.729 | 0.454 | 1 |
To analyze thousands of genes, scientists use a technique called hierarchical clustering.
Hierarchical clustering works by taking the most similar genes and joining them in a cluster. Genes B and D are the most similar at 0.729, so they are joined to become [BD].