Foundation model
Unsupervised
Public
The model learns what a plausible protein looks like. Still, this data covers less than 0.01% of all natural sequences.
Target agnosticProject agnostic
Our wet lab in Amsterdam measures how sequence changes affect expression, stability and solubility. Public databases do not have this data.
Public
The model trains again on sequences from your protein family. It learns which positions you can change, and which positions you must keep.
Target agnosticProject specific
Assay data from your company and this project. The model learns all your properties and their trade-offs.
Target specificProject specific
The model designs the next variants. The lab tests them, and the model is retrained on all results so far. A few rounds are usually enough.
