Resolving the structure of interactomes with hierarchical agglomerative clustering
2011

New Algorithm for Clustering Biological Interaction Networks

Sample size: 5000 publication Evidence: high

Author Information

Author(s): Park Yongjin, Bader Joel S

Primary Institution: Johns Hopkins University

Hypothesis

Can a new clustering algorithm improve the prediction of biological interactions in networks?

Conclusion

The Hierarchical Agglomerative Clustering (HAC) algorithm outperforms existing methods in predicting links in biological networks.

Supporting Evidence

  • HAC provides the best performance in link prediction compared to other clustering methods.
  • Link prediction based on joint clustering of physical and genetic interactions performed worse than predictions based on individual data sets.
  • The algorithm effectively identifies both top-level and bottom-level clusters in biological networks.

Takeaway

Scientists created a new way to group similar proteins together, which helps predict how they interact with each other better than before.

Methodology

The study developed a new algorithm called Hierarchical Agglomerative Clustering (HAC) for clustering biological networks and predicting links.

Limitations

The algorithm may become trapped in local minima and may not generalize well across different types of networks.

Digital Object Identifier (DOI)

10.1186/1471-2105-12-S1-S44

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication