---
license: other
license_name: sfm-research-preview-license-v1.0
license_link: LICENSE.md
tags:
  - mirna
  - rna
  - gene-regulation
  - post-transcriptional
  - vc-sfm
  - biology
  - bioinformatics
pipeline_tag: feature-extraction
---

# mir-SFM — miRNA ↔ mRNA Target Specificity Foundation Model

**Paper:** Vibe Coding Specificity Foundation Models · doi: [10.64898/2026.06.04.730134](https://doi.org/10.64898/2026.06.04.730134)
**All VC-SFM models:** [huggingface.co/SFM-BIIE-ETHZ](https://huggingface.co/SFM-BIIE-ETHZ)
**Code:** [github.com/SFM-BIIE-ETHZ/Vibe-Coding-SFMs](https://github.com/SFM-BIIE-ETHZ/Vibe-Coding-SFMs)


---

## What it does

This SFM learns a joint embedding space for **microRNAs (miRNAs)** and **target mRNA sequences** via contrastive learning on experimentally validated interactions. Given a miRNA, retrieve the most likely target mRNAs.

| Component | Model |
|-----------|-------|
| Agent encoder | DNABERT-2 |
| Target encoder | DNABERT-2 |
| Training data | ENCORI CLIP-seq (135 miRNAs / 110,221 targets; 206,692 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 (%) |
|-----------|---------|---------|----------|
| miRNA → mRNA | 98.0 ± 0.6 | 100.0 ± 0.0 | 100.0 ± 0.0 |
| mRNA → miRNA | 25.4 ± 4.2 | 53.4 ± 7.4 | 67.5 ± 7.2 |

*Paper fold-0 pool-512 R@1 (miRNA→mRNA) = **98.4%**.*

---

## Quick start

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

ckpt_path = hf_hub_download("SFM-BIIE-ETHZ/mirSFM_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("UAGCUUAUCAGACUGAUGUUGA")             # mature miRNA
target_emb = model.encode_target("GCAUGUUUUCAAAGAUGAGAGGACGCAUAUAAUUU")  # mRNA target

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

```bibtex
@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