Genomics & Single-Cell · DeepMind
Sequence-to-expression across a 200kb window covering distal enhancers.
30s–5min
Runs on On-demand GPU
Runs on our servers with on-demand compute. A first run needs time to load the model; active capacity can be reused and scales down when idle.Run this model
About this model
Enformer's long receptive field is the point: it captures enhancer–promoter interactions up to 100kb away, which short-context models structurally cannot. The standard tool for asking what a non-coding variant does to expression.
Standardized I/O contract
Every model in the Hub speaks the same contract, which is what lets the Router and the agent call any of them without special-casing.
Inputs
Genomic sequence
Variant sequence
Organism
Outputs
CAGE, DNase and histone mark predictions per bin.
