Database search and clustering performance

USEARCH features home page

Database search

 

Speed

Sensitivity

Benchmark

click for details

Nucleotide

30-200 x BLASTN

70 x MEGABLAST

100% x BLASTN (all %ids)
150% x MEGABLAST (high %id)
200-500% x MEGABLAST (med.-low %id)

RFAM

Protein

20-250 x BLASTP

100% x BLASTP (≥50%id)
86% x BLASTP (35-50%id)
29% x BLASTP (25-35%id)

PDB90-R

Translated

20-250 x BLASTX

99% x BLASTX (≥50%id)
84% x BLASTX (35-50%id)
25% x BLASTX (25-35%id)

PDB90-X

 
Clustering

 

Speed

Sensitivity

Benchmark

click for details

Nucleotide

30-50 x CD-HIT

110-160% x CD-HIT

COSTELLO

Protein

0.8-70 x CD-HIT  

115-135% x CD-HIT

ORTHO

454 reads

30 x CD-HIT-454

110% x CD-HIT-454

454READS

OTUs

(Benchmarks are controversial, see Discussion).

MOCK

Comments on comparing CD-HIT and the UCLUST algorithm in USEARCH.
 

Chimera filtering
See the UCHIME paper for benchmark comparisons to ChimeraSlayer and Perseus.