Related Experiment Videos
Combinatorial motif analysis and hypothesis generation on a genomic scale.
Y J Hu1, S Sandmeyer, C McLaughlin
1Information and Computer Science Department, Department of Biological Chemistry, College of Medicine, University of California, Irvine, USA.
Bioinformatics (Oxford, England)
|June 27, 2000
Summary
This study introduces novel algorithms for identifying regulatory motifs and predicting their combined activity in gene regulation. The new methods effectively identify sequence motifs and analyze their combinations, advancing biosequence analysis.
Area of Science:
- Bioinformatics
- Computational Biology
- Genomics
Background:
- Computer-assisted methods are crucial for biosequence analysis.
- Gene activity relies on transcription factors binding to regulatory motifs.
- Identifying these motifs and their combinations is key to understanding gene regulation.
Purpose of the Study:
- Develop algorithms for identifying regulatory motifs.
- Predict the activity of combined regulatory motifs.
- Enhance the analysis of biosequences for gene regulation studies.
Main Methods:
- Employed a novel motif-finding method with multiple objective functions and improved stochastic iterative sampling.
- Utilized constructive induction for combinatorial motif analysis.
- Applied standard inductive learning algorithms for hypothesis generation.
Main Results:
- Validated the motif-finding method on 10 yeast regulons and 14 artificial sequence families.
- Demonstrated the effectiveness of motif combination and classification approaches.
- Applied methods to analyze DNA array data from genome-wide gene expression studies.
Conclusions:
- The developed algorithms are effective for identifying regulatory motifs and predicting their combinatorial activity.
- The new methods advance the computational analysis of gene regulation.
- Software will be available upon request.