Skip to main content
Fig. 1 | BMC Bioinformatics

Fig. 1

From: GRASPx: efficient homolog-search of short peptide metagenome database through simultaneous alignment and assembly

Fig. 1

Summary of the major improvements of GRASPx. Q is the reference/query sequence. a Prebuilt Extension Link: The original GRASP algorithm first searches the suffix x (green) of the current contig P against the suffix array SA (which is built on the sequencing read set R), and then identifies the MESs \( \overrightarrow{p}* \) (pink). GRASPx simply follows the pre-built extension links (black dashed arrows) to identify all MESs. b Local Assembly: The original GRASP algorithm extends the contig until significant score drop-off is observed, which might lead to unnecessary extensions of the random seeds (orange). GRASPx extends each seed with a pre-set maximal depth. Homologous contigs (with corresponding blue seeds) that are broken by this strategy are re-assembled through a recalibration step based on sequence overlap (highlighted by the vertical double-sided reverse arrows). c Query-level Parallelism: GRASP spawns a thread (gray box) for each seed extension. GRASPx spawns a thread for each query sequence

Back to article page