A general strategy to determine the congruence between a hierarchical and a non-hierarchical classification
2007
A New Method for Comparing Hierarchical and Non-Hierarchical Classifications
Sample size: 308
publication
Evidence: high
Author Information
Author(s): Marco Antonio, MarĂn Ignacio
Primary Institution: Universidad de Valencia
Hypothesis
How can we effectively compare hierarchical classifications with non-hierarchical classifications?
Conclusion
The proposed method is highly sensitive and outperforms previously described strategies.
Supporting Evidence
- Our method detected 28 significant clusters for 7 GO classes, while previous methods detected at most 4.
- The coverage for the significant clusters was 85.4%, indicating high sensitivity.
- The method was applied to analyze the yeast mitochondrial interactome, showing a 75% detection rate for protein complexes.
Takeaway
This study presents a new way to compare two types of data classifications, helping scientists understand complex biological data better.
Methodology
The method uses permutation analyses of ranked clusters to determine significant enrichments in hierarchical classifications.
Limitations
The method may be computationally intensive, especially with larger datasets.
Statistical Information
P-Value
p' < 0.01
Statistical Significance
p<0.01
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website