This set of benchmarks highlights the ability of PLAST to perform both fast and sensitive search on large data banks.
PLAST, BLAST, UBLAST and DIAMOND are evaluated using the following data sets:
- nucleic/protein: Marine metagenome / Uniprot and RNA-seq / Human Proteome
- protein/protein: E.Coli proteome / Uniprot and Yeast proteome – Uniprot
- nucleic/nucleic: Marine metagenome / Genbank (virus) and Zair Ebola virus / Genbank (virus)
Tools name are specified according to their parameter setting:
- default parameters: plast, blast, ublast, diamond
- Priority to Quality: plastQ, blastQ, ublastQ, diamondQ
- Priority to Speed: plastS, blastS, ublastS, diamondS
In addition, for protein/protein benchmarks, SSearch and SWIPE are used as reference tools.
Benchmarks have been done using the CAST tool (*).
You can also have a look at the original PLAST benchmark, conducted by the end of software development.
(*) Drezen E. and Lavenier D. (2014) Quality metrics for benchmarking sequences comparison tools. Advances in Bioinformatics and Computational Biology Lecture Notes in Computer Science, Springer, 8826, pp.144-153. [PDF]
