VLF: An R package for the analysis of very low frequency variants in DNA sequences

Area of Science:

• Bioinformatics
• Genomics
• Molecular Evolution

Background:

• DNA sequence data is rapidly accumulating in public databases.
• Previous methods for analyzing very low frequency variants (VLFs) were slow and error-prone.
• Accurate sequence data is crucial for population genetics and phylogeography.

Purpose of the Study:

• Introduce VLF, an R package for VLF analysis in nucleotide and amino acid sequences.
• Automate the detection of errors in DNA sequence records.
• Provide a faster, more scalable, and user-friendly alternative to existing VLF analysis methods.

Main Methods:

• Developed the VLF R package to calculate nucleotide or amino acid frequencies at each sequence position.
• Implemented a user-specified frequency threshold (default p=0.001) to identify VLFs.
• Applied the VLF package to avian and fish DNA barcode datasets.

Main Results:

• VLF package demonstrated high consistency with previous Excel-based VLF analysis on a bird DNA barcode dataset.
• Identified biological artifacts and taxonomic issues (species synonymy) in sequence data.
• Successfully applied VLF to assess sequence data quality in avian and fish datasets.

Conclusions:

• The VLF R package offers significant improvements in automation, speed, scalability, and ease-of-use for VLF analysis.
• VLF is a valuable tool for ensuring the accuracy of large genetic databases like BOLD and GenBank.
• The package aids in exploring microevolutionary patterns and sequence conservation.