An Embedding Is Not a Mutation-Effect Score: A Protein LM Primer
Fair-Rain3366 · reddit · 2026-10-14
If you're using protein language models to rank enzyme mutations, beware a common trap: extracting an embedding doesn't by itself predict activity.
- Embeddings (per-residue or averaged ESM vectors) are just representations — you still need a predictor trained on labeled examples to predict a property
- Mutation scores are a different output: a masked model compares the likelihood of the replacement vs. original amino acid, ranking substitutions without your labels — but it's not an activity measurement or a probability of success
- Practical rule: feed embeddings into a trained predictor, or use the model's documented mutation-scoring procedure, and evaluate against the property you actually measured; the two routes are not interchangeable
Links included to the ESM repo, the NeurIPS 2021 mutation-scoring paper, and a rewirebio.io guide (AI-assisted, disclosed).
More from Research
- 1-gram wireless mesoscope images cortex-wide activity in freely moving mice — KordingLab · 2026-10-23
- Reversible computing veteran poses open problem: universal BARCS primitives from unbiased Josephson junctions — MikePFrank · 2026-10-23
- Arc Institute uses genomic language model to map 376M prokaryotic noncoding sequences into 5.14M clusters — tedgreenwald · 2026-10-23
- The rise of AI-assisted peer review: reviewers are making it into the appendix — abursuc · 2026-10-23
- Anthropic uses Claude to formally verify security properties in 100+ open-source libraries at RISC-V assembly level — AnthropicAI · 2026-10-23
- Researchers test 9 strategies to pick LLM teams, beating random selection by up to 36.1% — shangbinfeng · 2026-10-23