Feature Context Search
Feature Context Search
Search our database of 130,000+ microbial genomes for genomic neighborhoods where a combination of features occurs, without a query sequence. Features are Pfam domains, Rfam families and intergenic SAE features (interpretable features learned from the gLM2 genomic language model), combined with and, or and not. A neighborhood matches when the required features are found within window genes of each other. Returns the proteins in each matching window, with taxonomy and functional annotations.
Feature Context Search › Request Body
The features that must occur together: Pfam, Rfam and SAE feature literals combined with and, or and not. At most 10 literals. The expression must always require at least one feature to be present, so an expression that can be satisfied by not terms alone is rejected. Valid identifiers are listed by /api/v1/pfam-families, /api/v1/rfam-families and /api/v1/sae-features.
Only return matches from genomes matching this taxonomy expression. See /api/v1/taxa.
windowWindow radius in genes (not base pairs) around each feature hit. Required features must fall within this window, and the proteins in it are returned. At most 20.
Return at most one match per taxon at this rank, so results are not dominated by over-represented lineages. Set to null to return every match.
maxResultsMaximum number of results to return, up to 100.
includeOtherSaesAlso return SAE features in the window that are not named in featureFilter. Makes the response much larger.
includeOtherRfamsAlso return Rfam hits in the window for families not named in featureFilter.
Feature Context Search › Responses
Successful Response
Genomic neighborhoods where the feature expression is satisfied.