Lecture20 - Other Clustering Methods Lecture 20: Data...

Info iconThis preview shows pages 1–3. Sign up to view the full content.

View Full Document Right Arrow Icon
Lecture 20: Data Visualization § Other cluster methods k -means clustering • Self-organizing Maps (SOMs) § Visualization methods § Boolean queries § Principal Component Analysis Some slides adapted from notes created by Dr. Jaideep Chaudhary Other Clustering Methods § k -means Clustering § Self-Organizing Maps (SOMs) k -means SOMs Image from Patrik D'haeseleer (2005) Nature Biotechnology 23, 1499-1501 k -means Clustering Procedure § 1. Determine the number (k) of distinct clusters to use for clustering the gene expression data sets • The clustering will change depending on choice of the number of clusters § 2. Assign cluster centroids (mean or center of cluster) arbitrary positions in the expression data space (alternative : assign genes to arbitrary clusters) § 3. Calculate the distance of each gene to each cluster centroid (based on expression data) § 4. Assign genes to the closest (most similar) cluster centroid • e.g., Find cluster centroid with smallest distance to gene § 5. Recalculate cluster centroids (mean of cluster) based on the genes assigned to that cluster § 6. Repeat steps 3 - 5 until clusters converge (or for a certain number of iterations) • Genes no longer reassigned to new cluster • Cluster centroids (means) remain constant k -means clustering k = 3 Arbitrarily choose k positions as initial cluster centroids (centers) Continue to reassign genes and recalculate centroids until genes are no longer reassigned to new cluster Images generated using k -means online demo at: http://www.rob.cs.tu-bs.de/content/04-teaching/06-interactive/Kmeans/Kmeans.html Assign each gene to the nearest cluster centroid Experiment 1 Experiment 2 Experiment 1 Experiment 2 Experiment 1 Recalculate the cluster centroid (mean)
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
k -means clustering example (continued) Experiment 1 Experiment 2 End of Iteration 1 Experiment 2 Experiment 1 Iter. 2 reassign Experiment 1 Iter. 2 recalculate Iter. 3 reassign Experiment 1 Iter. 3 recalculate Experiment 1 Experiment 1 Iter. 4 reassign Images generated using k -means online demo at: http://www.rob.cs.tu-bs.de/content/04-teaching/06-interactive/Kmeans/Kmeans.html k -means clustering example (continued) Images generated using k -means online demo at: http://www.rob.cs.tu-bs.de/content/04-teaching/06-interactive/Kmeans/Kmeans.html Experiment 1 Iter. 4 recalculate Experiment 1 Iter. 5 reassign Iter. 6 recalculate Experiment 1 Experiment 1 Iter. 7 END Experiment 1 Iter. 5 recalculate Iter. 6 reassign Experiment 1 Experiment 1 Initialization k -means clustering Experiment 1 Final Clustering k -means clustering example: Summary
Background image of page 2
Image of page 3
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 01/20/2012 for the course MBIOS 478 taught by Professor Staff during the Fall '11 term at Washington State University .

Page1 / 6

Lecture20 - Other Clustering Methods Lecture 20: Data...

This preview shows document pages 1 - 3. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online