Optimization of ribosomal RNA profile alignments

Area of Science:

• Bioinformatics
• Computational Biology
• Molecular Evolution

Background:

• Maintaining large alignments of ribosomal RNA (rRNA) sequences is crucial for biological research.
• Current methods for updating rRNA alignments involve manual and automatic approaches.
• Profile alignment methods offer potential for optimizing rRNA sequence alignment.

Purpose of the Study:

• To evaluate profile alignment methods for ribosomal RNA (rRNA) sequence alignment.
• To optimize parameter choices and sequence weighting schemes for improved alignment accuracy.
• To develop and test a new sequence weighting strategy for rRNA alignments.

Main Methods:

• Empirical comparison of various sequence weighting schemes on a large eukaryotic SSU rRNA alignment.
• Systematic evaluation across a range of gap penalties.
• Development of a novel weighting scheme prioritizing sequence similarity within profiles.

Main Results:

• The new weighting scheme, which emphasizes sequences most similar to the new sequence, demonstrated superior performance.
• This novel scheme, when integrated with traditional weighting methods, yielded the most accurate alignments.
• Performance was assessed using a comprehensive dataset of eukaryotic small subunit rRNA sequences.

Conclusions:

• Profile alignment methods, particularly with optimized weighting, significantly enhance rRNA sequence alignment.
• The developed weighting scheme offers a more accurate approach for incorporating new sequences into existing alignments.
• This work provides a refined methodology for large-scale rRNA sequence analysis.