DNA sequence recognition by hybridization to short oligomers

Area of Science:

• Genomics
• Bioinformatics
• Molecular Biology

Background:

• Massive hybridization experiments are crucial for DNA analysis.
• Sequencing by hybridization (SBH) relies on interrogating clone arrays with oligomer probes.
• Current methods often involve DNA sequence reconstruction for similarity searches.

Purpose of the Study:

• To develop a direct similarity search method using hybridization data, bypassing sequence reconstruction.
• To propose an information-theoretic approach for analyzing oligomer lists from hybridization experiments.
• To create a computationally efficient and error-tolerant method for DNA clone identification.

Main Methods:

• Developed a format 1 hybridization technology for large-scale clone interrogation.
• Proposed a similarity search method leveraging subword structure of identified oligomers.
• Applied minimal length encoding and algorithmic mutual information for sequence comparison.
• Tested the method on real and simulated hybridization data.

Main Results:

• Demonstrated that direct analysis of oligomer lists is more effective than reconstructed sequences for similarity searches.
• The proposed method correctly identified DNA clones using 109 short oligomer probes.
• The method is robust to hybridization errors and requires fewer probes than traditional sequencing.
• Applicable to both format 1 and format 2 hybridization experiments.

Conclusions:

• Directly utilizing oligomer lists from hybridization experiments is superior for DNA similarity searches.
• The developed method offers computational efficiency and error tolerance.
• This approach can guide large-scale DNA sequencing efforts using various methods.