SFM-BIIE-ETHZ/crisprSFM_VC-SFM
crisprSFM — CRISPR gRNA ↔ Off-Target DNA 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 CRISPR guide RNAs (gRNAs) and candidate off-target genomic DNA sites via contrastive learning. Given a gRNA, retrieve the most likely off-target loci for rapid off-target risk assessment.
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.
Paper fold-0 pool-512 R@1 (gRNA→off-target) = 98.6%.
Quick start
from huggingface_hub import hf_hub_download
import torch, torch.nn.functional as F
ckpt_path = hf_hub_download("SFM-BIIE-ETHZ/crisprSFM_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("GAGTCCGAGCAGAAGAAGAA") # 20-nt gRNA (no PAM)
target_emb = model.encode_target("GAGTCCGAGCAGAAGAAGAANGG") # 23-nt off-target + PAM
score = F.cosine_similarity(agent_emb, target_emb, dim=-1)Files in this repo
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
