Overlapping translation of nucleic acid sequences for bioinformatics applications

Area of Science:

• Bioinformatics
• Computational Biology
• Genomics

Background:

• Traditional sequence alignment tools like blastN and TblastX compare nucleic acid sequences directly or translate them into proteins in three reading frames.
• These methods can be computationally intensive and may miss subtle similarities due to frame shifts or errors.

Purpose of the Study:

• To develop and evaluate an alternative method for nucleic acid sequence similarity searching.
• To improve the speed, sensitivity, and accuracy of detecting sequence similarities compared to existing tools.

Main Methods:

• Nucleic acid sequences were translated into overlapping protein-like sequences (OTSS).
• These OTSS were then searched using the BlastP algorithm.
• Performance was compared against blastN and TblastX using established datasets and metrics.

Main Results:

• BlastP searches using OTSS identified approximately two-thirds of matches found by blastN and TblastX, but also revealed additional significant similarities.
• Matches discovered by BlastP with OTSS were generally longer, had higher scores, and lower E-values, indicating more robust similarity detection.
• Visualization of extensive sequence similarities, such as between prion protein and human insulin gene, was improved using OTSS.

Conclusions:

• The developed method using overlapping translated sequences (OTSS) offers a faster and more sensitive alternative to TblastX for nucleic acid similarity searches.
• This approach enhances the detection of weak sequence similarities, reduces sensitivity to sequencing errors and mutations, and eliminates frame shift issues.
• OTSS-based BlastP searches provide a valuable improvement for sequence analysis in bioinformatics.