New Method for Finding Gene Patterns in Data
Author Information
Author(s): Gan Xiangchao, Liew Alan Wee-Chung, Yan Hong
Primary Institution: King's College London
Hypothesis
Can a geometric perspective improve the detection of biclusters in gene expression data?
Conclusion
The proposed geometric framework effectively identifies biologically significant gene subsets in microarray data.
Supporting Evidence
- The algorithm can detect overlapping biclusters effectively.
- It outperforms existing methods in identifying biologically significant gene groups.
- The method is robust against noise in gene expression data.
Takeaway
This study introduces a new way to find groups of genes that behave similarly under certain conditions, which helps scientists understand how genes work together.
Methodology
The study uses a geometric interpretation of biclustering and implements a Hough transform-based hyperplane detection algorithm.
Limitations
The algorithm's computational demands increase significantly with larger datasets.
Participant Demographics
The study focuses on gene expression data from human lymphoma samples.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website