{"slug": "prohiflo-hierarchical-flow-matching-with-functional-guidance-for-de-novo-protein", "title": "ProHiFlo: Hierarchical Flow Matching with Functional Guidance for De Novo Protein Generation", "summary": "Researchers have developed ProHiFlo, a hierarchical flow matching framework for de novo protein generation that models backbone geometry before refining to all-atom coordinates, reducing computational cost while maintaining accuracy. The framework incorporates functional guidance from pretrained predictors to steer generation toward desired properties without retraining, and uses an adaptive SE(3)-equivariant architecture for efficient multi-scale processing. In enzyme active site scaffolding, ProHiFlo achieved a 58.9% success rate compared to 41.2% for RFDiffusion, requiring four fewer sampling steps.", "body_md": "arXiv:2606.11243v1 Announce Type: new\nAbstract: De novo protein generation has transformative potential in therapeutic design, enzyme engineering, and synthetic biology. While diffusion-based and flow matching approaches have achieved progress, they typically operate at single resolution and lack mechanisms for incorporating functional constraints. We introduce ProHiFlo, a hierarchical flow matching framework with three innovations: (1) coarse-to-fine generation that models backbone geometry before refining to all-atom coordinates, reducing computational cost while maintaining accuracy; (2) functional guidance leveraging pretrained predictors to steer generation toward desired properties without retraining; (3) adaptive SE(3)-equivariant architecture for efficient multi-scale processing. Experiments on unconditional generation, motif scaffolding, and functional design demonstrate state-ofthe-art performance while requiring 4 fewer sampling steps. On enzyme active site scaffolding, ProHiFlo achieves 58.9% success rate compared to 41.2% for RFDiffusion.", "url": "https://wpnews.pro/news/prohiflo-hierarchical-flow-matching-with-functional-guidance-for-de-novo-protein", "canonical_source": "https://arxiv.org/abs/2606.11243", "published_at": "2026-06-12 04:00:00+00:00", "updated_at": "2026-06-12 04:02:46.187504+00:00", "lang": "en", "topics": ["generative-ai", "machine-learning", "artificial-intelligence", "ai-research"], "entities": ["ProHiFlo", "RFDiffusion"], "alternates": {"html": "https://wpnews.pro/news/prohiflo-hierarchical-flow-matching-with-functional-guidance-for-de-novo-protein", "markdown": "https://wpnews.pro/news/prohiflo-hierarchical-flow-matching-with-functional-guidance-for-de-novo-protein.md", "text": "https://wpnews.pro/news/prohiflo-hierarchical-flow-matching-with-functional-guidance-for-de-novo-protein.txt", "jsonld": "https://wpnews.pro/news/prohiflo-hierarchical-flow-matching-with-functional-guidance-for-de-novo-protein.jsonld"}}