Thal-Kak Multiple Sequence Alignment Databases (prebuilt MMseqs2 and HH-suite databases for structure prediction)

Nine prebuilt protein sequence databases for multiple sequence alignment search, published by CSSB at Seoul National University in the native on-disk formats of MMseqs2 and HH-suite. MSAs are the primary input to protein structure prediction (AlphaFold, Protenix, Boltz, RoseTTAFold) and to conservation and coevolution analysis, and building these indexes from raw FASTA is the expensive, awkward step; this archive skips it. Sources are UniProt UniRef100 release 2026_01, the reduced BFD, MGnify release 2024_04, EnVhogDB and the Logan protein catalogue derived from the NCBI Sequence Read Archive. UniRef100, MGnify and the reduced BFD were re-clustered; EnVhogDB was re-clustered at its upstream standard level. Two top-level prefixes hold the two formats, mmseqs/ and hhblits/, with LICENSE.txt and README.md at the bucket root (the archive is mounted whole, not per prefix, so those stay visible). Measured 2026-10-27: 11 objects, 398.9 GiB, as nine .zst-compressed tar archives (four under mmseqs/, five under hhblits/). The bucket README states 1.77 TiB after extraction plus about 0.70 TiB of MMseqs2 search indexes built locally, so plan for roughly 2.5 TiB of local disk, not 399 GiB. They are individually enormous: hhblits/uniref100_2026_01.tar.zst is 157.2 GiB, mmseqs/mgnify_clusters.tar.zst 127.0 GiB, hhblits/mgnify_clusters.tar.zst 33.7 GiB. Each database carries the licence of the source it derives from and no database mixes sources: CC0 1.0, CC BY 4.0 and CC BY-SA 4.0 across most of them, but the two Logan databases carry no licence grant at all. LICENSE.txt at the root is the authoritative per-database record and also documents what was changed relative to each upstream distribution.

Category
Structural Biology
Provider
Center for Structural Systems Biology (CSSB), Seoul National University
License
Per-database, inherited from each upstream source (see LICENSE.txt at the bucket root)
Hosted on
Amazon S3, bucket thal-kak-msa, us-east-1
Access
Free and public. No account with the data provider is needed.
Source
https://registry.opendata.aws/thal-kak-msa/

Analyze this dataset without downloading it

On Scigantic, Thal-Kak Multiple Sequence Alignment Databases (prebuilt MMseqs2 and HH-suite databases for structure prediction) opens in a hosted JupyterLab notebook with the data already mounted, so you can read files where they are stored instead of copying them first. Each notebook includes an AI assistant that can see the mounted files and write and run analysis code. Open the dataset page to launch a notebook, or browse the full catalog.

Related Structural Biology datasets