CEMBRA OA pseudo-SNP SVD64 development archive

This private repository is a curated model-and-evidence export from the oa_tl research codebase. It scores paired 3,000 bp reference/alternate DNA windows for osteoarthritis (OA) variant prioritization using a 9,216-feature hashed DNA representation, a 64-dimensional encoder initialized by label-free TruncatedSVD pretraining, and a tanh/linear ranker.

Research archive only. This is a development model, not an externally confirmed, clinically validated, calibrated, or deployment-ready predictor. It must not be used for diagnosis, treatment, or patient-level decisions.

Why this model was selected

The repository contains many historical and experimental checkpoints. This export selects the compact pseudo-SNP SVD64 arm because it is the only model that passed its locked development gate and it remained the best native single arm in the later five-model NT/Borzoi comparison.

On the governed OA2025 PIP >= 0.30 development cohort (168 rows, 56 positives, 49 guarded components), its pooled out-of-fold result was:

Metric	pseudo-SNP SVD64 transfer	independently tuned scratch	Delta
AUROC	0.611288	0.593750	+0.017538
Average precision	0.433655	0.417225	+0.016429
Component-weighted concordance	0.620748	0.600340	+0.020408
Component-weighted top-1	0.455782	0.425170	+0.030612

The gain was modest and not uniform across folds. External confirmation was not opened or scored. See results/pseudosnp/AUDIT.md and results/pseudosnp/model_selection_gate.json for the exact claim boundary.

Important negative evidence

The formal leakage-resistant NT-v2 comparison was a development NO-GO: pretrained-minus-random matched concordance was -0.1136 on 22 matched groups. No NT-v2 model lock was justified.
The later fixed 1:1 NT+Borzoi fusion was also a NO-GO. Its AUROC/AP were 0.500638/0.356623, below pseudo-SNP SVD64 by 0.110651/0.077032.
July 2026 NT-v2 metrics and eight approximately 374 MiB checkpoints are retained locally as historical artifacts but are superseded for model selection and intentionally not uploaded here.

Files

Path	Contents
`weights/pretrained_svd_encoder.npz`	Label-free 9,216 x 64 SVD initialization and spectrum
`weights/folds/fold_*_transfer_fixed_model.npz`	Five outer-fold tanh/linear rankers
`inference.py`	Standalone NumPy feature extraction and raw scoring
`verify_export.py`	Recomputes one held-out score per fold from the uploaded weights
`results/pseudosnp/`	Gate, receipts, OOF predictions, metrics, plot, and audit
`results/v2_formal/`	Formal leakage-resistant NT-v2 NO-GO evidence
`results/nt_borzoi/`	Latest fusion NO-GO evidence
`results/legacy/`	Superseded historical metric tables only
`REPO_SUMMARY.md`	Whole-repository technical and scientific summary
`SHA256SUMS`	Integrity hashes for uploaded artifacts

Weight format

Every fold file is a safe NumPy .npz archive containing only float32 arrays:

encoder_weight: [9216, 64]
encoder_bias: [64]
scorer_weight: [64]
scorer_bias: scalar

The raw fold score is

tanh(features @ encoder_weight + encoder_bias) @ scorer_weight + scorer_bias.

The uploaded artifacts are cross-validation fold models, not a single full-development refit. The convenience mean across the five fold scores in inference.py is useful for research inspection but was not the evaluated OOF quantity and is not a validated deployment ensemble.

Usage

from inference import PseudoSnpSVD64

model = PseudoSnpSVD64(".")
fold_scores = model.score_pair_by_fold(reference_3000bp, alternate_3000bp)
research_only_mean = model.score_pair(reference_3000bp, alternate_3000bp)

The two sequences must be exactly 3,000 bases long, contain the variant at zero-based index 1,500, and differ at that position only. Scores are raw ranking logits, not probabilities. Do not apply a clinical threshold.

To verify the package against the stored OOF evidence:

python verify_export.py

Provenance

Source repository: git@github.com:zibin-zhao/cembra-bio.git
Exported from commit: 76a214e44d83d7577d7d9bc7a67796240e970d25
Source branch at export: codex/nt-borzoi-oof-fusion
Selected run: 20260808T003427Z_multiscale_pseudosnp_svd_344bbf0cca5b
Selected model schema: oa-tl-tanh-encoder-linear-triplet-ranker.v1
Feature schema: oa-tl-multiscale-pseudosnp-hashed-dna.v1
Original audited run-manifest SHA-256: ad72b7a4f5585dd6ebd3c9639d0ff48d7a69d4d124472e204c2b2edd8533925a

The 2026-08-22 fusion audit and runner existed as uncommitted worktree artifacts at export time; its own audit records result-affecting source and artifact hashes. It is included as later negative context, not as selected model provenance.

License

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