PoET-2 Introduces Retrieval-Augmented Multimodal Protein Foundation Model
On Feb 26, 2026, researchers published PoET-2, a multimodal, retrieval-augmented protein foundation model that integrates family-specific in-context evolutionary retrieval with optional structure conditioning. PoET-2 employs a hierarchical equivariant transformer encoder and dual decoders supporting generative and bidirectional modes, achieving state-of-the-art zero-shot variant effect prediction and improved supervised embeddings, notably for multi-mutation and indel scoring on small datasets.
Key Points
- 1Demonstrates state-of-the-art zero-shot variant effect prediction, including multi-mutation and indel scoring.
- 2Combines retrieval augmentation with multimodal, family-centric modeling to capture evolutionary constraints more effectively.
- 3Enables better supervised sequence-function learning, particularly for small datasets and challenging mutation types.
Scoring Rationale
Strong methodological novelty and improved zero-shot performance, limited by single-source arXiv preprint lacking peer review.
Sources
Public references used for this report.
Practice interview problems based on real data
1,625 SQL & Python problems across 15 industry datasets — the exact type of data you work with.
Try 250 free problems
