eSFM — Enzyme ↔ Substrate Specificity Foundation Model

Paper: Vibe Coding Specificity Foundation Models · doi: 10.64898/2026.06.04.730134 All VC-SFM models: huggingface.co/SFM-BIIE-ETHZ Code: github.com/SFM-BIIE-ETHZ/Vibe-Coding-SFMs


What it does

This SFM learns a joint embedding space for enzyme protein sequences and substrate small molecules (SMILES) via contrastive learning on reaction data. Given an enzyme, retrieve its most likely substrate — or score enzyme–substrate pairs directly.

Component Model
Agent encoder ESM-2 (protein language model)
Target encoder MoLFormer-XL (SMILES)
Training data ReactZyme (177,389 enzymes / 2,646 substrates; 177,442 pairs)
Split Sequence-identity 100% holdout · fold 0

Performance — pool-512 retrieval (from the paper)

Evaluated by pool-512 retrieval: each test pair's true target is placed in a pool of 512 candidates (1 positive + 511 random negatives), scored by cosine similarity, over 100 random trials at the best-validation checkpoint. Random baseline = 0.2%. Values are the 5-fold cross-validated mean ± SD (folds 0–3; fold 4 excluded for split degeneracy) reported in the paper for this SFM. The released checkpoint is the fold-0, identity-100 model trained with the identical configuration, data, and split.

Direction R@1 (%) R@5 (%) R@10 (%)
enzyme → substrate 86.3 ± 1.5 95.3 ± 1.0 97.1 ± 0.8
substrate → enzyme 61.8 ± 0.4 88.2 ± 0.2 93.5 ± 0.5

Released checkpoint = fold 0; per-fold values are not separately tabulated in the paper (mean reported).


Quick start

from huggingface_hub import hf_hub_download
import torch, torch.nn.functional as F

ckpt_path = hf_hub_download("SFM-BIIE-ETHZ/eSFM_VC-SFM", "model.pth")

# Load with the Vibe-Coding-SFMs codebase
# (https://github.com/SFM-BIIE-ETHZ/Vibe-Coding-SFMs)
from calm.encoder.model import CALMEncoder
model = CALMEncoder.from_pretrained(ckpt_path)
model.eval()

agent_emb  = model.encode_query("MKALLIVLGLVSSVSQASST...")   # enzyme protein
target_emb = model.encode_target("OC(=O)c1ccccc1")            # substrate SMILES

score = F.cosine_similarity(agent_emb, target_emb, dim=-1)

Files in this repo

File Description
model.pth Released checkpoint · fold 0 · identity-100 split
results_train_val_test.csv Per-epoch training/validation/test logs (training-time batch metrics, not the pool-512 numbers above)

Citation

@article{reddy2026vcsfm,
  title   = {Vibe Coding Specificity Foundation Models},
  author  = {Reddy, Sai T.},
  journal = {bioRxiv},
  year    = {2026},
  doi     = {10.64898/2026.06.04.730134}
}

License

Released under the SFM Research Preview License v1.0-preview (see LICENSE.md). Free for research use — academic, non-profit, government, and industry research. The specific molecules disclosed in the accompanying preprints are dedicated to the public. Commercial-use and patent-licensing terms are deferred and being arranged with ETH Zürich / BIIE; the SFM architectures and training methods are the subject of pending patent applications. For commercial enquiries: sai.reddy@ethz.ch

Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support