tRNAfeature: An algorithm for tRNA features to identify tRNA genes in DNA sequences

Area of Science:

• Genomics
• Molecular Biology
• Bioinformatics

Background:

• Accurate identification of transfer RNA (tRNA) genes is essential for understanding biological evolution.
• Unusual tRNA substructures pose challenges for conventional detection methods, potentially leading to incorrect anticodon identification.
• Developing robust methods for detecting tRNA genes with atypical structures is an ongoing research need.

Purpose of the Study:

• To propose and evaluate a novel method, tRNAfeature, for identifying tRNA genes, with a focus on those possessing unusual substructures.
• To enhance the accuracy and reliability of tRNA gene detection in genomic data.

Main Methods:

• The tRNAfeature method refolds sequences, specifically targeting single-stranded regions that deviate from canonical tRNA structural models.
• The method was applied to predict 53,926 archaeal, eubacterial, and eukaryotic tRNA genes from the tRNADB-CE database.
• tRNA genes were subsequently scanned across whole-genome sequencing data.

Main Results:

• tRNAfeature demonstrated superior performance compared to existing tRNA identification methods.
• The method successfully identified a large set of tRNA genes across diverse domains of life.
• The approach is effective in detecting tRNAs with non-canonical structural features.

Conclusions:

• The proposed tRNAfeature method offers a more powerful and accurate approach for identifying tRNA genes, including those with unusual structures.
• This advancement aids in a more comprehensive understanding of tRNA gene evolution and function across various organisms.
• tRNAfeature represents a significant improvement for tRNA gene annotation in large-scale genomic datasets.