Release 3.0.0
Summary
This release adds RNA Multiple Sequence Alignment search using Riboseek, enabling nucleotide sequence homology search against RNA databases such as RNAcentral.
New Features
RNA MSA Search
A new endpoint /biology/riboseek/msa-search/predict enables RNA sequence alignment search:
- Nucleotide input: Submits RNA/DNA sequences using the standard IUPAC nucleotide alphabet (A, C, G, T, U and ambiguity codes)
- Riboseek-powered: Uses Riboseek, which encodes RNA dinucleotides for sensitive homology search
- GPU-accelerated: Shares the same GPU Server infrastructure as protein search for low-latency queries
- Auto-detection: The NIM automatically distinguishes RNA databases (built with Riboseek) from protein databases (built with MMseqs2) by reading the scoring matrix embedded in each index file. No manual configuration is required.
RNA databases are not included in the default protein profile. Select the databases:rna model profile to download the RNA databases (refer to Task-Specific Profiles and Custom Databases for details). For usage examples, refer to RNA MSA Search in the Getting Started guide.
Deprecation Updates
- The
/biology/colabfold/msa-search/config/msa-database-configsendpoint, previously announced for removal in 3.0.0, remains available in this release. Removal is deferred to a future release. - The
NIM_MSA_DB_NAME_MAPPINGSenvironment variable, previously announced for removal in 3.0.0, remains available in this release.
Migration Notes
Model profile hashes may have changed in this release. If you pin a specific profile using NIM_MODEL_PROFILE, update the hash. Run list-model-profiles to obtain the new hashes (refer to Listing Available Profiles for more information).