RAPSearch

RAPSearch performs fast protein similarity searches to annotate protein-coding sequences translated from short DNA reads generated by next-generation sequencing.

Key Features:

• Reduced Amino Acid Alphabet: Uses a reduced amino acid alphabet to streamline similarity detection and improve search speed.
• Suffix Array-Based Seed Detection: Leverages suffix arrays to detect seeds of flexible length for identifying protein similarities.
• Six-Frame Translation: Translates short DNA reads across six reading frames prior to protein similarity search.
• Speed and Efficiency: Achieves approximately 20–90 times faster processing than BLASTX for large NGS datasets.
• Sensitivity and Specificity: Misses about 1.3–3.2% of similarity hits compared to BLASTX while identifying approximately 0.3–2.1% additional homologous proteins that BLASTX does not detect.

Scientific Applications:

• Metagenomics: Annotation of protein-coding genes from short reads generated by metagenomic sequencing.
• Functional Annotation: Rapid identification of homologous proteins to support inference of biological functions and characterization of microbial communities.

Methodology:

Short DNA reads are translated in six reading frames, converted to a reduced amino acid alphabet, and suffix arrays are used to detect flexible-length seeds for protein similarity search.