Semi-supervised gene shaving method for predicting low variation biological pathways from genome-wide data
2009

New Method for Identifying Low Variation Biological Pathways

publication Evidence: moderate

Author Information

Author(s): Zhu Dongxiao, Michael Q Zhang, Michael S Waterman, Xuegong Zhang

Primary Institution: University of New Orleans

Hypothesis

Can a semi-supervised gene clustering algorithm effectively identify gene clusters with modest expression variation using prior knowledge of signaling pathways?

Conclusion

The proposed algorithm successfully identifies tightly regulated gene clusters with low and concordant expression variation.

Supporting Evidence

  • The algorithm demonstrated advantages over the original gene shaving algorithm using two microarray data sets.
  • It effectively identified gene clusters showing concerted and modest expression variation.
  • The method allows for the incorporation of incomplete prior knowledge to predict new pathway genes.

Takeaway

This study created a new way to group genes that change together, even if those changes are small, by using what we already know about how genes work together.

Methodology

The study used a semi-supervised gene clustering algorithm that incorporates prior knowledge of signaling pathways to identify gene clusters with modest expression variation.

Limitations

The algorithm's effectiveness depends on the availability and accuracy of prior knowledge, which may not always be complete.

Digital Object Identifier (DOI)

10.1186/1471-2105-10-S1-S54

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication