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 / chief-inverse-folding
Protein
Shanghai Jiao Tong UniversityXinhua Hospital, Shanghai Jiao Tong University School of MedicineWenzhou Medical UniversityPeking UniversityTsinghua UniversityReleased March 2025

CHIEF (Chimera Ensemble Inverse Folding)

Protein inverse folding ensemble that fuses five pretrained sequence designers through a self-attention encoder, reaching 63.1% recovery on CATH4.2.

The short version

  • —Designs sequences for a fixed backbone by blending five inverse folding models per residue
  • —Self-attention reweights each base model by sequence context instead of fixed voting weights
  • —Holds above 55% recovery on backbones past 800 residues, where single models fall off
  • —Recovers 72.4% of residues at the closest protein-protein interface contacts
  • —Redesigned malate dehydrogenase sequences all expressed and showed enzymatic activity
13Openness0Citations

Where to run it

No providers recorded yet. Browse all providers

Inverse folding — writing an amino acid sequence that will fold into a given backbone — is now served by several strong models, and they disagree with each other in instructive ways. Message-passing designers are trained on multi-chain assemblies and handle interfaces well; autoregressive sequence-structure transformers capture evolutionary signal but were trained on proteins shorter than 500 residues; models optimized on soluble subsets favor hydrophilic surface residues. Each therefore carries a different bias, and none dominates across protein length, solvent exposure, or conservation.

CHIEF (Chimera Ensemble Inverse Folding) treats that heterogeneity as a resource rather than a problem. Developed by Zilong Geng, Yuze Wang, Tingting Liu and colleagues at Shanghai Jiao Tong University with collaborators at several Chinese hospitals and universities, it runs five frozen pretrained designers — ProteinMPNN-vanilla, ProteinMPNN-soluble, ESM-IF, Frame2seq and PiFold — over the same backbone, then learns a transformer encoder that combines their per-residue probability distributions. Unlike averaging or weighted voting, which assign each base model a fixed weight, the attention layer reweights them according to sequence context, so a model can dominate exactly where it is strong. Two of the five base models are variants of ProteinMPNN.

This entry describes a protein design model and is unrelated to CHIEF, the histopathology foundation model that shares the acronym.

#Key Features

  • Context-dependent fusion: A self-attention encoder weighs the five base models per residue rather than assigning them fixed global weights, letting the ensemble exceed any linear combination of its members.
  • Robust to missing members: Ablating any single base model changes sequence recovery and perplexity only slightly, indicating the ensemble draws on all five rather than proxying for one.
  • Scales to long chains: Recovery stays above 55% for backbones beyond 800 residues, a regime where the autoregressive base model drops below 40%.
  • Strong at interfaces: On a test set of 586 homomers and 423 heteromers, recovery at the closest interface contacts (Cα distance 4 Å) reaches 72.4%.
  • Low compositional bias: Amino acid usage in generated sequences tracks the natural distribution, avoiding the missing- and over-used residue types seen in several base models.

#Technical Details

Each base model receives backbone N, Cα, C and O atoms and emits a 21-dimensional log-probability per residue (20 amino acids plus a missing token). The five outputs are concatenated into an [N, 105] tensor, augmented with sinusoidal positional encodings, and passed through a three-layer transformer encoder with four attention heads and a hidden dimension of 128, trained with a masked negative log-likelihood objective using Adam at a learning rate of 0.001 for more than 200 epochs on a single A100. Training used PDB structures clustered at 30% sequence identity, split into 23,358 training, 1,464 validation and 1,539 test clusters. Sequence recovery reaches 61.8% on the PDB test set (perplexity 3.4), 63.1% on single-chain CATH4.2, 70.6% on TS50 and 71.2% on TS500 — 16.6–28.0% above the best base model, with perplexity 22.7–34.6% lower. Averaging and weighted-voting ensembles of the same five models reach 56.3% and 59.2%. Designed sequences refolded with ESMFold match or beat the base models on RMSD, TM-score, pLDDT and secondary structure recovery. On ProteinGym deep mutational scanning, CHIEF ranks mid-pack on Spearman correlation, AUC and MCC — below ESM-IF — but leads on NDCG (0.75) and top-10% recall (0.22), the metrics that matter for prioritizing high-fitness variants.

#Applications

CHIEF is aimed at enzyme engineering and functional protein design, where the goal is a sequence that folds and works rather than one that merely matches a reference. The authors redesigned malate dehydrogenase from an AlphaFold-predicted backbone (UniProt A0A319AA41), sampling 1,000 sequences at temperature 0.15 and synthesizing ten; all ten expressed in E. coli BL21, purified, and showed measurable enzymatic activity, with one comparable to the wild-type enzyme. A second demonstration redesigned the 954-residue nuclease Cas13d, reaching a TM-score of 0.92 against the AlphaFold 3 prediction of the wild type, versus 0.72 for the best single base model.

#Impact

The result argues that ensembling frozen inverse folding models is worth more than it sounds: the gain over the strongest member is large, and the least-squares attribution analysis recovers the expected division of labor, with the autoregressive model carrying early residues and conserved positions while the multi-chain designers carry interfaces and surfaces. Practical uptake is limited for now. The work is a preprint that has not been peer reviewed, the wet-lab validation rests on ten designed sequences of a single enzyme, and no code or weights have been released — the paper states that the repository is being prepared for public release. Until that lands, CHIEF is a reproducible recipe rather than a runnable tool.

At a glance

Released
March 2025
Category
Protein
Organizations
Shanghai Jiao Tong University / Xinhua Hospital, Shanghai Jiao Tong University School of Medicine / Wenzhou Medical University / Peking University / Tsinghua University

Related models

  • BC-Design

    Gerstein Lab / Yale University

  • ProteinZero

    University of Illinois Urbana-Champaign

  • Caliby

    Stanford University

  • ADFLIP

    MRC Laboratory of Molecular Biology

  • ProteinMPNN

    Institute for Protein Design

Links

Research Paper

Tags

de_novo_designinverse_foldingprotein_designproteomicsrepresentation_learningsupervisedtransformer

Something wrong?

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