New Method for Potential Fusions Detection in Protein-Coding Sequences.

Area of Science:

• Bioinformatics
• Genomics
• Molecular Evolution

Background:

• Gene fusion is a key mechanism for novel gene formation.
• Traditional bioinformatics methods rely on sequence similarity, failing to detect fusions with divergent ancestral sequences.
• Previous work identified triplet periodicity (TP) change points as potential indicators of gene fusion.

Purpose of the Study:

• To improve the triplet periodicity (TP) change point detection method.
• To investigate the prevalence and significance of TP change points in eukaryotic genomes.
• To identify novel fused genes, particularly those with lost or highly diverged ancestral sequences.

Main Methods:

• Developed and refined a method for detecting triplet periodicity (TP) change points in protein-coding sequences (CDS).
• Applied the improved method to analyze genes across six eukaryotic genomes.
• Utilized BLAST for sequence alignment to validate potential fused genes.

Main Results:

• TP change points were detected in 20%-40% of genes across six eukaryotic genomes at a 2%-3% type I error rate.
• Approximately 30% of detected TP change points were attributable to amino acid repeats.
• Another 30% corresponded to potential fused genes identified by BLAST alignment.
• The remaining TP change points likely represent fused genes with undetectable ancestral sequences.

Conclusions:

• The improved TP change point detection method is more sensitive and identifies 2-3 times more significant TP change points than previous methods.
• This approach enhances the discovery of fused genes, including those with highly diverged or lost ancestral sequences.
• TP change point analysis offers a valuable complementary strategy to sequence similarity searches in gene fusion studies.