soedinglab/MMseqs2

Health v2 · maintenance only

Adoption not part of the score

2131 stars · 299 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

MMseqs2 is an open-source C++ suite for ultra-fast, sensitive search and clustering of huge protein and nucleotide sequence sets, running orders of magnitude faster than BLAST with comparable sensitivity. It supports profile searches, taxonomic assignment of metagenomic contigs, and GPU-accelerated homology search across multi-core and server environments.

Use cases

• search large protein sequence databases faster than BLAST
• cluster millions of protein sequences in linear time
• run PSI-BLAST-like profile searches at high speed
• assign taxonomy to metagenomic contigs
• cluster nucleotide sequence sets with linclust
• run interactive local sequence searches via a web server app
• accelerate homology search on GPUs

When to choose

• you need to search or cluster very large sequence datasets where BLAST is too slow
• you want near-BLAST sensitivity at a fraction of the runtime
• you need scalable multi-core or GPU-accelerated sequence search
• you are doing metagenomic analysis or massive protein set clustering

When to avoid

• you only need to search a handful of sequences and prefer BLAST's familiar output
• you need pairwise alignment scoring details beyond MMseqs2's approximations
• you require a Windows-native build without cygwin
• you need structural or RNA-specific alignment tools instead of sequence search

Facets

application · maturity active

search-engine cli gpu-computing bioinformatics big-data windows cpp cli sequence-search sequence-clustering blast-alternative protein-sequences nucleotide-sequences profile-search taxonomy-assignment metagenomics high-performance-computing search linux macos gpu docker

Member repositories

For agents

markdown · JSON · MCP: product_card(name="soedinglab/MMseqs2")