Non-linear dimensionality reduction of signaling networks
2007

Analyzing Signaling Networks with Isomap

Sample size: 1006 publication 10 minutes Evidence: moderate

Author Information

Author(s): Ivakhno Sergii, Armstrong J Douglas

Primary Institution: Massachusetts Institute of Technology

Hypothesis

Can non-linear dimensionality reduction techniques like Isomap effectively identify clusters in signaling networks?

Conclusion

The extended Isomap approach successfully identified clusters in signaling networks corresponding to different treatment conditions.

Supporting Evidence

  • Isomap was able to reconstruct clusters corresponding to different cytokine treatments.
  • Isomap outperformed PCA in identifying functionally coherent clusters.
  • The first three Isomap components captured 71% of the variance in the data.

Takeaway

This study used a special math tool called Isomap to help understand how different signals in cells can lead to different responses, like whether a cell lives or dies.

Methodology

The study applied an unsupervised non-linear dimensionality reduction approach, Isomap, to analyze two cell signaling networks.

Limitations

Isomap struggled to identify clusters for low-dose treatments, which formed a supercluster.

Participant Demographics

The study involved human epithelial cancer cells and RAW 264.7 macrophages.

Statistical Information

P-Value

p<0.05

Statistical Significance

p<0.05

Digital Object Identifier (DOI)

10.1186/1752-0509-1-27

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication