ConforNets: Latents-Based Conformational Control in OpenFold3
Minji Lee, Colin Kalicki, Minkyu Jeon, Aymen Qabel, Alisia Fadini + 1 more
TLDR
ConforNets introduce a novel method for controlling protein conformations in AlphaFold3 models, achieving state-of-the-art success in generating alternate states.
Key contributions
- Addresses AlphaFold's limitation in capturing diverse protein conformational states.
- Introduces ConforNets, using channel-wise affine transforms on AF3's pre-Pairformer latents.
- Achieves state-of-the-art success in unsupervised generation of alternate protein states.
- Enables novel supervised conformational transfer across protein families.
Why it matters
AlphaFold models struggle with diverse protein conformations, limiting their biological utility. ConforNets provide a robust, reusable mechanism to control these states, crucial for understanding protein function and drug design. This advancement opens new avenues for exploring protein dynamics.
Original Abstract
Models from the AlphaFold (AF) family reliably predict one dominant conformation for most well-ordered proteins but struggle to capture biologically relevant alternate states. Several efforts have focused on eliciting greater conformational variability through ad hoc inference-time perturbations of AF models or their inputs. Despite their progress, these approaches remain inefficient and fail to consistently recover major conformational modes. Here, we investigate both the optimal location and manner-of-operation for perturbing latent representations in the AF3 architecture. We distill our findings in ConforNets: channel-wise affine transforms of the pre-Pairformer pair latents. Unlike previous methods, ConforNets globally modulate AF3 representations, making them reusable across proteins. On unsupervised generation of alternate states, ConforNets achieve state-of-the-art success rates on all existing multi-state benchmarks. On the novel supervised task of conformational transfer, ConforNets trained on one source protein can induce a conserved conformational change across a protein family. Collectively, these results introduce a mechanism for conformational control in AF3-based models.
📬 Weekly AI Paper Digest
Get the top 10 AI/ML arXiv papers from the week — summarized, scored, and delivered to your inbox every Monday.