Search for SINE repeats in the rice genome using correlation-based position weight matrices

Area of Science:

• Genomics
• Bioinformatics
• Molecular Biology

Background:

• Transposable elements (TEs), including Short Interspersed Nuclear Elements (SINEs), comprise a substantial portion of eukaryotic genomes.
• SINEs are widespread in mammalian genomes and also present in plants.
• TEs accumulate mutations post-insertion, complicating identification and annotation via current bioinformatics approaches.

Purpose of the Study:

• To identify highly divergent SINE copies within the rice (Oryza sativa subsp. japonica) genome.
• To evaluate the efficacy of the Highly Divergent Repeat Search Method (HDRSM) for SINE discovery.

Main Methods:

• The HDRSM utilizes correlations of neighboring symbols to build position weight matrices (PWMs) for SINE families.
• A simulated dataset with SINE copies featuring substitutions and indels was used to assess HDRSM accuracy against RepeatMasker.
• The HDRSM was applied to search for 39 SINE families in the rice genome.

Main Results:

• The HDRSM demonstrated superior performance in identifying inserted repeats and accurately defining their boundaries compared to RepeatMasker.
• The HDRSM identified 14,030 SINE copies in the rice genome, with 5,704 copies not detected by RepeatMasker.
• HDRSM successfully located divergent SINE copies with high statistical significance.

Conclusions:

• The HDRSM is effective for finding divergent SINE copies and accurately determining their genomic boundaries.
• RepeatMasker excels at identifying shorter, more similar SINE copies, while HDRSM is better suited for more diverged elements.
• A combined approach using both HDRSM and RepeatMasker is recommended for a comprehensive analysis of SINE distribution in genomes.