Efficient Generation of Protein and Protein–Protein Complex Dynamics via SE(3)-Parameterized Diffusion Models
Bibliographic record
Abstract
Protein and protein-protein complex conformations play a critical role in biological functions, while exploring these via traditional molecular dynamics (MD) simulation is computationally expensive. Enhanced sampling methods offer improvements but remain limited by vast conformational spaces. Recently, advances in generative deep learning have provided new avenues for protein conformational sampling. To address this challenge, we propose protein trajectory diffusion (PTraj-Diff), a geometric diffusion framework designed for generating protein and protein-protein complex trajectories. PTraj-Diff simulates protein dynamics through a denoising process that iteratively reconstructs stable conformations from random noise. By parametrizing protein structures using residue-level SE(3) transformations, the model effectively captures geometric constraints and structural relationships inherent in natural proteins while introducing tensor product attention to reduce computational overhead and lower requirements for data and hardware resources. Simultaneously, we integrate a power Bert Encoder to achieve precise long-range temporal dependency. Experimental results demonstrate that PTraj-Diff efficiently explores the conformational trajectories of protein monomers and protein-protein complexes. Moreover, it is compatible with diverse conformations generated by AlphaFold3, enabling the prediction of high-quality trajectories. As deep generative modeling continues to integrate with MD simulations, this emerging approach is poised to become a powerful tool for investigating protein conformational dynamics and elucidating biological functions.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".