Dot2dot: accurate whole-genome tandem repeats discovery

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• Eukaryotic DNA contains numerous repetitive sequences, particularly tandem repeats (TRs), whose functions are not fully understood.
• TRs are implicated in various cellular processes and genetic disorders, necessitating accurate discovery methods.
• Existing computational models struggle to capture the variability of TRs, hindering association studies.

Purpose of the Study:

• To develop an accurate and efficient computational algorithm for discovering tandem repeats (TRs) in large-scale genomic datasets.
• To address the limitations of current methods in capturing TR variability for association studies and whole-genome analysis.

Main Methods:

• Formalized the structure of TRs in self-compared sequence dot-plot matrices.
• Developed a compact representation of these matrices searchable in linear time.
• Implemented the Dot2dot algorithm for TR discovery.

Main Results:

• Dot2dot demonstrates higher accuracy compared to established TR discovery methods.
• The algorithm successfully analyzed the largest known reference genome in approximately one day on a standard PC.
• Experiments on five curated TR collections validated the algorithm's performance.

Conclusions:

• Dot2dot provides an accurate and computationally efficient solution for whole-genome tandem repeat discovery.
• The algorithm's speed and accuracy facilitate large-scale genomic studies involving TRs.
• This tool can advance research into the functional roles of TRs and their association with genetic disorders.