bio.rodeo
ModelsOrganizationsProvidersLeaderboardAboutSign in
bio.rodeo

The authoritative source for evaluating biological foundation models. No hype, just honest analysis.

Categories
  • DNA & Gene
  • RNA
  • Protein
  • Small molecule
  • Single-cell
  • Spatial omics
  • Pathology
  • Imaging
  • Metabolomics
  • Biosignals
  • Language model
bio.rodeoModelsOrganizationsProvidersLeaderboardAboutFAQSubmit a modelContact
© 2026 Pulsatance. All rights reserved. ~
Built by Pulsatance
models / protein / asam
Protein
Michigan State UniversityReleased March 2025

aSAM

Protein conformational ensemble generator that samples heavy-atom structures in a latent space, with a variant conditioned on temperature.

The short version

  • —Generates all-atom MD-like ensembles from one input PDB structure
  • —Takes temperature as a continuous input, producing melting-curve-like folded fractions
  • —Diffuses in an SE(3)-invariant latent space, so side chain chi angles are learned directly
  • —Compresses GPU-months of simulation sampling into a single generation pass
85Openness3Citations
41GitHub stars
Apache-2.0License

Where to run it

No providers recorded yet. Browse all providers

Structure predictors return one conformation, but proteins exist as populations whose relative weights determine function. Molecular dynamics (MD) simulation samples those populations from physics, at a cost that makes exhaustive sampling impractical for most systems. A growing class of deep generative models is trained on MD trajectories to reproduce ensembles in seconds instead of GPU-months, with AlphaFlow and BioEmu-1 as prominent examples.

aSAM — atomistic Structural Autoencoder Model — is the entry in that class from Giacomo Janson, Alexander Jussupow, and Michael Feig at Michigan State University. It differs from its peers in two ways. First, it is a latent diffusion model: an autoencoder compresses heavy-atom coordinates into an SE(3)-invariant per-residue latent representation, and the diffusion process runs entirely in that latent space, with an AlphaFold 2 Structure Module serving as the decoder. Because every heavy atom is represented in the latent code, the model learns joint distributions of backbone and side chain torsions directly, rather than generating a backbone and appending side chains afterwards.

Second, one variant takes temperature as an explicit input. That variant, aSAMt, is trained on the mdCATH dataset of simulations at five temperatures and generates ensembles conditioned on a continuous temperature value, including values it never saw in training. A companion constant-temperature model, aSAMc, is trained on ATLAS at 300 K and serves as the head-to-head comparison against AlphaFlow. (The reference implementation lives in a repository named sam2, unrelated to the Segment Anything family of image models.)

#Key Features

  • Latent-space atomistic generation: The autoencoder reconstructs encoded MD snapshots to 0.3–0.4 Å heavy-atom RMSD, and diffusing in that space lets the model reproduce backbone φ/ψ and side chain χ distributions that Cβ-based generators cannot learn.
  • Continuous temperature conditioning: aSAMt accepts any temperature as a scalar input, interpolating smoothly between the five mdCATH training temperatures and extrapolating to values outside them with physically plausible folded-state trends.
  • Thermal unfolding behavior: Sweeping temperature yields sigmoid folded-fraction profiles resembling experimental melting curves, from which an apparent melting temperature can be fitted.
  • Efficient sampling: Generation is far cheaper than the simulations it emulates, and inference cost can be halved by reducing the number of diffusion steps with negligible loss of accuracy.
  • Two released checkpoints: Separate models trained on ATLAS and on mdCATH ship as a GitHub release, download automatically at first use, and take a single-chain PDB file as input.

#Technical Details

The encoder is a four-block transformer fed radial basis expansions of Cα–Cα, N–O, and side chain centroid distance matrices plus φ, ψ, ω, and χ1–χ4 torsion features, producing a latent of dimension 32 per residue; the decoder is a five-block AlphaFold 2 Structure Module. A denoising diffusion probabilistic model with a Diffusion Transformer backbone, a maximum of 1,000 timesteps, and a sigmoid noise schedule learns the latent distribution, conditioned on sequence, an initial 3D structure, and — for aSAMt — temperature. aSAMc was trained on 1,174 ATLAS chains (0.35 ms cumulative simulation time) and aSAMt on 5,268 mdCATH domains at 320, 348, 379, 413, and 450 K (55.0 ms cumulative), with 90 held-out test domains sharing at most 20% sequence identity with training. On the ATLAS test set, AlphaFlow holds a small edge in Cα RMSF correlation (0.904 versus 0.886), while aSAMc is better on backbone and side chain torsion distributions and runs 17.4× to 28.0× faster at the default 100 diffusion steps. Against MD run under a matched protocol, aSAMt was 1,000× to 2,317× faster. On 62 monomeric proteins with measured melting temperatures, fitted values correlated with experiment at r = 0.510 while sitting systematically about 130 K high.

#Applications

aSAM suits any workflow that needs a conformational ensemble rather than a single structure and can tolerate MD-quality statistics: estimating residue flexibility, generating starting structures for targeted simulations, examining side chain rotamer populations at a binding site, and screening designed proteins for conformational stability. Temperature conditioning opens a use case its peers do not cover — comparing the thermal behavior of homologs or variants in silico, which the authors demonstrate on mesophile/hyperthermophile protein pairs and on de novo designs from RFdiffusion characterized experimentally as highly thermostable.

#Impact

aSAM was published in Communications Chemistry in November 2025, following a March 2025 bioRxiv preprint; the version of record carries CC BY-NC-ND 4.0 terms while the preprint is CC BY. Its main contribution is showing that environmental conditions can be folded into a transferable ensemble generator at all, and that latent-space diffusion is a practical route to atomistic detail without an all-atom denoiser. Real limitations remain and are documented by the authors: relative free energies between metastable states are not quantitatively reproduced, since mdCATH trajectories are too short to observe many transitions; accuracy degrades at 450 K, where the folded-fraction error rises to 0.141; and the model handles only single-chain proteins of standard amino acids with no ligands or cofactors. Code and both checkpoints are released under Apache 2.0.

At a glance

Released
March 2025
Category
Protein
License
Apache-2.0
Organization
Michigan State University

Related models

  • DynaFold

    Shanghai Jiao Tong University

  • All-Atom Protein Sequence Diffusion

    Delft University of Technology

  • Ambient Protein Diffusion

    University of Texas at Austin / MIT CSAIL

  • UniSim

    Tsinghua University / Renmin University of China

  • AlphaFlow

    MIT

Links

GitHub RepositoryResearch PaperbioRxiv PreprintDataset

Tags

autoencoderconformational_ensemble_generationdiffusiongenerativemolecular_dynamics

Something wrong?

Much of this page is generated or calculated automatically. Flag anything that looks off and we will re-run it.