README.md

---
license: other
license_name: om-lula-community-license-1.1
license_link: LICENSE
pipeline_tag: other
tags:
  - biology
  - drug-discovery
  - small-molecule-discovery
  - protein-ligand
  - binding-affinity
  - open-weight
  - lula-1
---

> **License notice.** By downloading, accessing, or using LULA-1, you agree to
> the [Om LULA Community License 1.1](LICENSE). LULA-1 is licensed for research,
> evaluation, benchmarking, teaching, and other non-commercial research uses
> only. **Any commercial work requires a separate Om commercial license**,
> including internal commercial discovery, commercial drug discovery, screening,
> hit finding, lead optimization, portfolio decisions, production R&D, product
> candidate identification, patent or therapeutic program work, hosted
> inference, paid API/SaaS access, resale, support/deployment, product
> bundling, and competing model services. For commercial licensing, contact
> dmc@omtx.ai.

<style>
  @font-face {
    font-family: 'OmMontserrat';
    font-style: normal;
    font-weight: 100 900;
    font-display: swap;
    src: url('https://huggingface.co/spaces/omtx/README/resolve/main/assets/Montserrat-Latin-Variable.woff2') format('woff2');
  }
  .om-card { background:#0A0A0A; border:1px solid #2A2A2A; border-radius:14px; overflow:hidden;
    font-family:'OmMontserrat','Montserrat','Helvetica Neue',Arial,sans-serif; color:#fff;
    margin-bottom:26px; }
  .om-card p, .om-card h1 { margin-top:0; margin-left:0; margin-right:0; }
  .om-hero { position:relative; overflow:hidden; background:#0A0A0A; }
  .om-hero-img { position:absolute; inset:0; width:100%; height:100%; object-fit:cover;
    object-position:72% 50%; margin:0; }
  .om-scrim { position:absolute; inset:0;
    background:linear-gradient(90deg,rgba(10,10,10,0.97) 0%,rgba(10,10,10,0.90) 44%,rgba(10,10,10,0.30) 100%); }
  .om-cmark { position:absolute; color:#3B3636; font-size:19px; line-height:1; font-weight:300; z-index:5; }
  .om-copy { position:relative; z-index:2; padding:36px 34px 38px; max-width:660px; }
  .om-mono { color:#71717A; font-size:11px; font-weight:500; letter-spacing:0.26em;
    text-transform:uppercase; margin-bottom:14px; }
  .om-eyebrow { color:#E2E756; font-size:12px; font-weight:600; letter-spacing:0.32em;
    text-transform:uppercase; margin-bottom:18px; }
  .om-h1 { font-size:40px; font-weight:700; letter-spacing:-0.03em; line-height:1.0;
    margin-bottom:18px; color:#fff; }
  .om-accent { color:#E2E756; }
  .om-lede { color:#D4D4D8; font-size:16px; line-height:1.5; margin-bottom:22px; max-width:54ch; }
  .om-pill { display:inline-block; border:1px solid #2A2A2A; color:#D4D4D8; font-size:11px;
    font-weight:700; letter-spacing:0.08em; text-transform:uppercase; padding:8px 15px;
    border-radius:100px; margin:0 7px 7px 0; }
  .om-status { border-top:1px solid #2A2A2A; background:#0D0C0C; padding:26px 34px 28px; }
  .om-chip { display:inline-block; border:1px solid #3B3636; color:#71717A; font-size:11px;
    font-weight:700; letter-spacing:0.08em; text-transform:uppercase; padding:7px 13px;
    border-radius:6px; margin:0 7px 7px 0; }
  .om-chip-live { border-color:#E2E756; color:#E2E756; }
</style>

<div class="om-card">
  <div class="om-hero">
    <img class="om-hero-img" src="assets/lula1-hero.jpg" alt="Protein scaffold with a ligand bound in a highlighted pocket" />
    <div class="om-scrim"></div>
    <span class="om-cmark" style="top:15px; left:15px;">+</span>
    <span class="om-cmark" style="top:15px; right:15px;">+</span>
    <span class="om-cmark" style="bottom:15px; right:15px;">+</span>
    <div class="om-copy">
      <p class="om-mono">omtx.ai</p>
      <p class="om-eyebrow">Open-weight release track</p>
      <h1 class="om-h1">LULA-1 — <span class="om-accent">sequence-only</span> protein–ligand scoring.</h1>
      <p class="om-lede">
        Protein amino-acid sequence plus ligand SMILES in, binding score out. No structure input,
        no docking, no folding step.
      </p>
      <p>
        <span class="om-pill">Sequence-only</span>
        <span class="om-pill">Local inference</span>
        <span class="om-pill">Open weights</span>
      </p>
    </div>
  </div>
  <div class="om-status">
    <p class="om-eyebrow">Model at a glance</p>
    <span class="om-chip om-chip-live">1.7M parameters</span>
    <span class="om-chip om-chip-live">6.8 MB</span>
    <span class="om-chip om-chip-live">Sequence-only</span>
    <span class="om-chip om-chip-live">Local inference</span>
    <span class="om-chip om-chip-live">Open weights</span>
    <span class="om-chip">Public AUROC 0.7615</span>
    <span class="om-chip">EF@1000 54.8×</span>
  </div>
</div>

# LULA-1

LULA-1 is a lightweight, fast, sequence-only protein–ligand binding scorer, trained on ~500M data
points from Om and public sources. It takes a protein
amino-acid sequence and a ligand SMILES string and returns a binding score. There is no structure
input, no docking, and no folding step.

It is fast because of how it is built: the protein is embedded once, and every additional ligand
costs only a projection and a cosine similarity. That is what makes screening a million-molecule
library practical.

The intended public interface is deliberately small:

```python
score(protein_sequence: str, smiles: str | list[str]) -> record | list[record]
```

## What this release contains

LULA-1 ships as a compact scoring head that runs on top of two public pretrained encoders. The
checkpoint is **6.8 MB — 1,705,984 parameters**. This makes the released Om weights easy to verify,
cache, and move between local environments while using widely available protein and molecule
representations at inference time.

### Parameters

| Component | Parameters | Source |
|---|---:|---|
| **LULA-1 scoring head** | **1,705,984** | this repository |
| ESM-2 650M — protein encoder (1280-dim) | 652,358,616 | `facebook/esm2_t33_650M_UR50D` |
| ChemBERTa-77M-MTR — ligand encoder (384-dim) | ~3,500,000 | `DeepChem/ChemBERTa-77M-MTR` |
| **Total at inference** | **~657.6M** | |

The scoring head is ~0.26% of the parameters in play at inference. Breakdown of the checkpoint:

| Tensor | Shape | Parameters |
|---|---|---:|
| `protein_projector.weight` | 1024 × 1280 | 1,310,720 |
| `protein_projector.bias` | 1024 | 1,024 |
| `ligand_projector.weight` | 1024 × 384 | 393,216 |
| `ligand_projector.bias` | 1024 | 1,024 |
| **Total** | | **1,705,984** |

The checkpoint also carries `score_logit_scale` (10.0), a scalar loaded as a Python float rather
than a tensor — it is a persisted hyperparameter, not a trainable parameter, and is excluded from
the count above.

Architecturally this is a ConPLex-style contrastive scorer: each encoder output is passed through a
single `Linear → GELU → Dropout` projection to a shared 1024-dim space, L2-normalized, compared by
cosine similarity, scaled by the logit scale persisted in the checkpoint, and passed through a
sigmoid. The output is a probability-like score in `[0, 1]`.

The `omtx lula download` command downloads the LULA-1 weights and the required public encoder assets
into the local cache for scoring. Om distributes the LULA-1 scoring head in this repository; the
third-party encoders remain governed by their own upstream terms.

## Usage

**Quickstart notebook:** [`notebooks/lula1_quickstart.ipynb`](notebooks/lula1_quickstart.ipynb) —
install, download, verify, and score a real target end to end. CPU works; GPU is faster.

```bash
pip install "omtx[lula]>=2.0.12"

hf auth login         # required for gated LULA-1 weights; say no to adding the token as a git credential
omtx lula download    # LULA-1 weights + both required encoders
omtx lula verify      # SHA256 against the release manifest
```

```python
from omtx.lula import load_model

CA2 = (
    "MSHHWGYGKHNGPEHWHKDFPIAKGERQSPVDIDTHTAKYDPSLKPLSVSYDQATSLRIL"
    "NNGHAFNVEFDDSQDKAVLKGGPLDGTYRLIQFHFHWGSLDGQGSEHTVDKKKYAAELHL"
    "VHWNTKYGDFGKAVQQPDGLAVLGIFLKVGSAKPGLQKVVDVLDSIKTKGKSADFTNFDP"
    "RGLLPESLDYWTYPGSLTTPPLLECVTWIVLKEPISVSSEQVLKFRKLNFNGEGEPEELM"
    "VDNWRPAQPLKNRQIKASFK"
)
mols = [
    "CC(=O)Nc1nnc(s1)S(N)(=O)=O",
    "Cc1ccc(cc1)S(=O)(=O)N",
    "CC(C)Cc1ccc(cc1)C(C)C(=O)O",
    "CCN(CC)CCNC(=O)c1ccc(N)cc1",
    "c1ccc(cc1)C(=O)O",
    "CCO",
]

model = load_model()
for row in sorted(model.score(protein_sequence=CA2, smiles=mols), key=lambda r: r["rank"]):
    print(row["rank"], round(row["score"], 4), row["smiles"])
```

Verified in Google Colab on 2026-07-29 with `omtx[lula]>=2.0.12`; LULA-1 is
now gated on Hugging Face, so run `hf auth login` or pass a Hugging Face token
before `omtx lula download`. Expected CA2 ranking:

```text
1 0.9995 CC(=O)Nc1nnc(s1)S(N)(=O)=O
2 0.9991 Cc1ccc(cc1)S(=O)(=O)N
3 0.2331 CCO
4 0.1643 CCN(CC)CCNC(=O)c1ccc(N)cc1
5 0.1518 CC(C)Cc1ccc(cc1)C(C)C(=O)O
6 0.0984 c1ccc(cc1)C(=O)O
```

Batch scoring returns, per molecule: `score`, `rank`, and `top_percentile_in_batch`.
`rank` and `top_percentile_in_batch` are computed within the batch you submit, making the output
directly useful for prioritizing a candidate set.

The same scoring is available hosted, with no local setup, via `client.lula1.score(...)`.

## Intended use

Ranking and triage of compound sets against a protein target — particularly where no structure is
available, or where a structure-based method would be too slow across the full set. LULA-1 returns a
relative score for prioritization.

## Your data stays local

Scoring and fine-tuning run entirely on your machine. Nothing about your targets or compounds is
transmitted to Om.

- `omtx lula score` and local fine-tuning make **no network calls**. Protein sequences, SMILES,
  labels, scores, and checkpoints never leave your environment.
- There is **no telemetry** in the local path — no usage pings, no analytics, no phone-home.
- `omtx lula download` is the only command that uses the network. It fetches the weights and the two
  public encoders, and sends nothing about your data.
- After that download, the whole path runs **offline**. Point `OMTX_LULA_HOME` at a shared cache and
  scoring works on an air-gapped machine.

This is a property of the code, not a policy promise: the scoring module contains no HTTP client and
no network imports. It is verifiable by inspection of the published package.

For teams that cannot send proprietary targets or compound libraries to a third-party API, this is
the point of an open-weight release.

## Hosted models and fulfillment

LULA-1's weights are open-weight so you can screen locally with no dependency on Om. Three things are
available from Om when you want more than that:

**Hosted LULA-1** — the same model, no local setup, no encoder downloads, no GPU. Useful for large
screens and for teams that want managed runs and tracking.
[omtx.ai/models](https://omtx.ai/models)

**LULA-2** — Om's private hosted successor for advanced target classes, including protein–protein
interfaces and flat or shallow pockets. It reads the protein and ligand together with
cross-attention rather than scoring from independent embeddings and is available through Om's hosted
API.
[omtx.ai/models](https://omtx.ai/models)

**Molecule Fulfillment** — once you have a ranked shortlist, order the physical compounds through
Om: search, quote, checkout, and order status in one flow.
[omtx.ai](https://omtx.ai)

Pricing for hosted scoring is at [omtx.ai/pricing](https://omtx.ai/pricing).

A practical pattern: screen wide and cheap with open-weight LULA-1, then use Om's hosted models and
fulfillment tools for managed runs, ordering, and final prioritization.

## What ships

| Path | |
|---|---|
| `model/best.pt` | Scoring-head checkpoint |
| `model/model_config.json` | Model configuration |
| `release_manifest.json` | Machine-readable hashes used by `omtx lula verify` |
| `checksums.txt` | Human-readable SHA256 list |
| `examples/` | Runnable scoring examples |
| `LICENSE`, `NOTICE` | Terms and third-party components |

Use is governed by `LICENSE`. Third-party pretrained encoders are governed by their own terms. This
release package focuses on the LULA-1 scoring head, verification assets, examples, and
documentation.

## Training data

LULA-1 was trained on **~500 million data points** from **Om and public sources**.

## Evaluation

LULA-1 is evaluated as a ranking model for hit triage. Scores are optimized for prioritizing
candidate molecules within a target-specific screening set.

### Public metrics

The LULA-1 release checkpoint is `lula1_v6`.

| Metric | LULA-1 v6 |
|---|---:|
| Public validation AUROC | **0.7615** |
| Proteome-scale macro AUROC | **0.633** |
| Proteome-scale aggregate EF@1000 | **54.8×** |

### Proteome-scale sweep

LULA-1 was also evaluated in a large proteome-scale ranking sweep.

| | |
|---|---|
| Proteins scored | **10,569** |
| Protein–ligand pairs scored | **10,575,246,882** |
| Known binders | 5,025,463 |
| Macro AUROC | 0.633 |
| Median AUROC | 0.678 |
| Aggregate EF@1000 | **54.8×** |

EF@1000 measures top-of-list enrichment: across the sweep, LULA-1 found 267,324 known binders in
the top-1000 ranked molecules per target, versus 4,877.6 expected under random ranking.

### Enrichment by target evidence

LULA-1 enrichment improves as more target evidence is available:

| Known binders | Proteins | Median AUROC | Mean P@1000 | Expected P@1000 | Implied EF |
|---|---:|---:|---:|---:|---:|
| 1-9 | 4,910 | 0.648 | 0.0092% | 0.0003% | ~31× |
| 10-49 | 2,230 | 0.610 | 0.1016% | 0.0024% | ~42× |
| 50-199 | 1,528 | 0.686 | 0.6811% | 0.0103% | ~66× |
| 200-999 | 1,102 | 0.738 | 3.8868% | 0.0470% | ~83× |
| >=1000 | 799 | 0.810 | 26.4541% | 0.5175% | ~51× |

### Target-level results

Per-protein metrics from the proteome-scale sweep, for well-known drug targets.

EF is AUPR divided by the target's observed positive rate in the ranked sweep.

| Target | UniProt | Known binders | AUROC | AUPR | EF |
|---|---|---:|---:|---:|---:|
| KIT | P10721 | 5,354 | 0.891 | 0.1005 | 19× |
| PIM1 | P11309 | 9,127 | 0.887 | 0.3810 | 42× |
| PARP1 | P09874 | 6,982 | 0.878 | 0.3436 | 50× |
| ALK | Q9UM73 | 4,265 | 0.876 | 0.1802 | 42× |
| BACE1 | P56817 | 15,149 | 0.876 | 0.4582 | 31× |
| BTK | Q06187 | 14,504 | 0.873 | 0.1465 | 10× |
| CA2 | P00918 | 10,901 | 0.870 | 0.5495 | 51× |
| HDAC1 | Q13547 | 12,855 | 0.853 | 0.5510 | 43× |
| CDK2 | P24941 | 16,693 | 0.845 | 0.2298 | 14× |
| MTOR | P42345 | 10,258 | 0.824 | 0.1394 | 14× |
| ABL1 | P00519 | 5,808 | 0.808 | 0.0540 | 9× |
| EGFR | P00533 | 18,363 | 0.788 | 0.1895 | 11× |
| ESR1 | P03372 | 7,815 | 0.757 | 0.2192 | 28× |
| AR | P10275 | 5,551 | 0.724 | 0.0799 | 14× |

Kinases and well-characterized enzyme families rank strongly. Target-level metrics and batch
rankings help prioritize each campaign.

## License

Weights are released under the **Om LULA Community License 1.1**. By downloading, accessing, or
using LULA-1, you agree to that license. See `LICENSE` for the full terms and `NOTICE` for
third-party components.

Summary (the `LICENSE` file governs):

- **Permitted without a commercial license** — non-commercial research, evaluation,
  benchmarking, teaching, security testing, local inference, local fine-tuning for non-commercial
  research, non-commercial demos, publishing non-commercial scientific results, and sharing derived
  checkpoints where terms allow.
- **Requires a separate Om commercial license** — any commercial work or commercialization,
  including internal commercial discovery, commercial drug discovery, screening, hit finding, lead
  optimization, portfolio decisions, production R&D, product candidate identification, patent or
  therapeutic program work, monetized hosting, paid API/SaaS access, reselling model access,
  support/deployment, bundling LULA-1 into a paid product, or building a competing model API around
  Om weights.
- **Commercial licensing contact** — email dmc@omtx.ai.
- **Publication attribution required** — public papers, preprints, benchmarks, reports, model cards,
  or presentations using LULA-1 must cite Om Therapeutics Inc. LULA-1, version `lula1_v6`, and the
  Hugging Face model page.

## Attribution

LULA-1's architecture is derived from **ConPLex**, developed at MIT.

> Singh, R., Sledzieski, S., Bryson, B., Cowen, L., & Berger, B. (2023).
> *Contrastive learning in protein language space predicts interactions between drugs and protein
> targets.* **Proceedings of the National Academy of Sciences**, 120(24).
> [doi:10.1073/pnas.2220778120](https://doi.org/10.1073/pnas.2220778120)

Reference implementation: [github.com/samsledje/ConPLex](https://github.com/samsledje/ConPLex)
(MIT License).

If you use LULA-1 in published work, please cite the ConPLex paper alongside Om.

## Links

- Om — <https://omtx.ai>
- Hosted models — <https://omtx.ai/models>
- Pricing — <https://omtx.ai/pricing>
- GitHub — <https://github.com/omtx-ai>