Lu Tianyu, Shuai Richard, Kouba Petr, Li Zhaoyang, Chen Yilin, Shirali Akio, Kim Jinho, Huang Po-Ssu
Department of Bioengineering, Stanford University.
Department of Biophysics, Stanford University.
bioRxiv. 2025 Aug 18:2025.08.18.670959. doi: 10.1101/2025.08.18.670959.
We present Protpardelle-1c, a collection of protein structure generative models with robust motif scaffolding and support for multi-chain complex generation under hotspot-conditioning. Enabling sidechain-conditioning to a backbone-only model increased Protpardelle-1c's MotifBench score from 4.97 to 28.16, outperforming RFdiffusion's 21.27. The crop-conditional all-atom model achieved 208 unique solutions on the La-Proteina all-atom motif scaffolding benchmark, on par with La-Proteina while having ~10 times fewer parameters. At 22M parameters, Protpardelle-1c enables rapid sampling, taking 40 minutes to sample all 3000 MotifBench backbones on an NVIDIA A100-80GB, compared to 31 hours for RFdiffusion.
我们展示了Protpardelle-1c,这是一组蛋白质结构生成模型,具有强大的基序支架功能,并支持在热点条件下生成多链复合物。对仅主干模型启用侧链条件后,Protpardelle-1c的MotifBench分数从4.97提高到28.16,超过了RFdiffusion的21.27。作物条件全原子模型在La-Proteina全原子基序支架基准测试中获得了208个独特的解决方案,与La-Proteina相当,但参数数量减少了约10倍。Protpardelle-1c有2200万个参数,能够快速采样,在NVIDIA A100-80GB上对所有3000个MotifBench主干进行采样需要40分钟,而RFdiffusion则需要31小时。