Summary
Protein structure diffusion produces fewer β-sheets than helices (1), and design success rates are generally lower (2). (3) nonetheless confirmed by circular dichroism that sheet designs express and fold correctly. This tendency against sheets can be ameliorated by fine-tuning (4). In contrast, language-model-designed proteins do not have this bias (5).
Details
(6) do not mention this explicitly but their experimentally characterized designs are all exclusively alpha helical. In general designs that aren’t fully α-helical these tend to have lower predicted TM-score when refolding (middle panel below).
Figures
Figure 3 from (1)
Figure 5h from (2)
Figure S14 from (2)
Figure 3 from (4)
Ref (5)
See also
- Protein backbones designed by diffusion, but not by language models, have more secondary structure
- De novo designed proteins with alpha helices are easier to predict than those with other secondary structures
1.
Lin Y, Alquraishi M. Generating Novel, Designable, and Diverse Protein Structures by Equivariantly Diffusing Oriented Residue Clouds. In: International Conference on Machine Learning. PMLR; 2023. p. 20978–1002. Available from: https://proceedings.mlr.press/v202/lin23a.html
2.
Ingraham JB, Baranov M, Costello Z, Barber KW, Wang W, Ismail A, et al. Illuminating protein space with a programmable generative model. Nature. 2023;623(7989):1070–8. Available from: https://doi.org/10.1038/s41586-023-06728-8
3.
Lee JS, Kim J, Kim PM. Score-based generative modeling for de novo protein design. Nature Computational Science. 2023;3(5):382–92. Available from: https://doi.org/10.1038/s43588-023-00440-3
4.
Huguet G, Vuckovic J, Fatras K, Thibodeau-Laufer E, Lemos P, Islam R, et al. Sequence-Augmented SE(3)-Flow Matching For Conditional Protein Backbone Generation. 2024; Available from: https://arxiv.org/abs/2405.20313
5.
Alamdari S, Thakkar N, van den Berg R, Tenenholtz N, Strome R, Moses AM, et al. Protein generation with evolutionary diffusion: sequence is all you need. openRxiv; 2023. Available from: https://doi.org/10.1101/2023.09.11.556673
6.
Watson JL, Juergens D, Bennett NR, Trippe BL, Yim J, Eisenach HE, et al. De novo design of protein structure and function with RFdiffusion. Nature. 2023;620(7976):1089–100. Available from: https://doi.org/10.1038/s41586-023-06415-8