arXiv cs.AI· Lorenz Vaitl, Ivan Tanasijevic, Arian Rokkum Jamasb, L. Santiago Mille-Fragoso, Pan Kessel·· 1 天前AI 评分31
研究：AlphaFold 3 类模型无需完整循环即可预测置信度，按需回收计算预算
Recycle only when you have to!
AI 导读
针对 AlphaFold 3 类模型（如 Boltz、Protenix）的蛋白质结合体设计，研究发现其 ipSAE、ipTM 等置信度指标与结构一致性指标可在早期 recycling 阶段直接从 trunk 预测。该预测器计算成本低、可微分且每个循环都可用，既可作为筛选时分配 recycling 预算的决策信号，也可作为幻觉式设计的目标函数。两种应用均在显著降低计算成本的同时提升了性能。
正文 · 原文
Abstract:AlphaFold 3-like models (e.g., Boltz, Protenix, etc.) have pushed the frontier of protein binder design. This architecture is used both for the generation of binder candidates (e.g., by hallucination or converting these models into backbone or full atom generative models) as well as for the screening of these candidates for valid binders. In this paper, we start from the observation that AlphaFold 3's confidence metrics (ipSAE, ipTM) as well as structural consistency metrics can be predicted directly from the trunk even at early recycling. Because our proposed predictor is cheap, differentiable, and available at every cycle, it can be used in two distinct ways: as a decision signal to wisely spend the recycling budget for screening and as an objective for hallucination-based design. We show that both applications lead to increased performance at significantly reduced computational cost.
| Subjects: | Quantitative Methods (q-bio.QM); Artificial Intelligence (cs.AI); Machine Learning (cs.LG) |
| Cite as: | arXiv:2610.40665 [q-bio.QM] |
| (or arXiv:2610.40665v1 [q-bio.QM] for this version) | |
| https://doi.org/10.48550/arXiv.2610.40665 arXiv-issued DOI via DataCite |
Submission history
From: Lorenz Vaitl [view email]
[v1]
Fri, 9 Oct 2026 18:42:17 UTC (3,714 KB)
来源：arXiv cs.AI · arxiv.org