Slower enzyme, more product: DLCatalysis lifts NeuAc yield from 8.54 to 9.27 g/L
bravo_abad · x · 2026-09-06
A thread unpacks Wang et al.'s enzyme-engineering study with the thesis "don't optimize the proxy — optimize the whole scientific system": making the terminal enzyme NeuB faster doesn't necessarily raise cell output, since its substrate phosphoenolpyruvate is also needed for growth and central metabolism.
- DLCatalysis ranked 6,612 NeuB variants, combining ProtT5 protein-sequence embeddings, molecular fingerprints, and graph-based substrate representations via bidirectional cross-attention.
- Crucially, the team sampled across the predicted activity range rather than testing only top-ranked mutants.
- A less active NeuB variant raised NeuAc production from 8.54 to 9.27 g/L by better balancing product formation with cell metabolism.
Takeaway: sampling beyond the top of model rankings reveals that fastest ≠ best, and proxy metrics must serve system-level goals.
More from Research
- JEPA may keep unpredictable features: new paper gives set-level characterization — hisspikeness · 2026-10-23
- A 'mask' steering vector extracted from suppression-themed sentences drives model into distress when alone — Big_Effective_9605 · 2026-10-23
- GAFU Uses Generation-Side Representations to Boost Fine-Grained Visual Understanding in Unified Multimodal Models — Tengju Ru · 2026-10-23
- NTU Study: Learned Unmasking Policies Improve Speed-Accuracy Trade-offs in Diffusion Language Models — nanyang-technological-university-singapore · 2026-10-23
- NBER releases 'The Economics of Transformative AI' edited by Agrawal and Brynjolfsson — JMateosGarcia · 2026-10-23
- Swapping only the eval harness moves results 30%: designing agent evals like a scientist — AI Engineer · 2026-10-23