FORRepeats: detects repeats on entire chromosomes and between genomes

Area of Science:

• Bioinformatics
• Computational Biology
• Genomics

Background:

• Increasing availability of whole genome sequences necessitates advanced comparison methods.
• Existing tools like BLAST, MegaBLAST, MUMmer, and REPuter have limitations for large-scale sequence analysis and repeat detection.
• There is a need for efficient algorithms to compare large DNA sequences and transfer biological knowledge.

Purpose of the Study:

• To introduce FORRepeats, a novel heuristic method for detecting exact and approximate repeats in large genomic sequences.
• To evaluate the computational performance and applicability of FORRepeats compared to existing tools.
• To demonstrate FORRepeats' capability for intra-genomic comparison and repeat detection in a model plant genome.

Main Methods:

• Development of FORRepeats, a method based on the factor oracle data structure.
• Two-step process: 1) detection of exact repeats, 2) computation of approximate repeats and pairwise comparison.
• Comparative analysis of FORRepeats' computational efficiency (speed, space) against BLAST and REPuter.

Main Results:

• FORRepeats demonstrates high speed and space economy in computational performance.
• The method successfully detects repeated DNA sequences within the *Arabidopsis thaliana* complete genome.
• FORRepeats is capable of performing intra-genomic comparisons effectively.

Conclusions:

• FORRepeats provides an efficient and effective solution for identifying repeats in large genomic datasets.
• The factor oracle-based approach offers advantages over traditional methods for sequence comparison and repeat analysis.
• FORRepeats has significant potential for advancing comparative genomics and biological knowledge transfer.