datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
clinvar-acmg-distillation
medtech-genomics-distillation (v2)
Distillation dataset for 5-class germline variant classification (Pathogenic / Likely Pathogenic / VUS / Likely Benign / Benign) under ACMG/AMP guidelines. Used to train the medtech-8b and medtech-4b LoRA adapters.
Each row is a chat-format example: a hint-free prompt built from a ClinVar variant trace (gene, HGVS, consequence, gnomAD allele frequency, review status) plus a DeepSeek-R1 teacher completion containing classification… See the full description on the dataset page: https://huggingface.co/datasets/Rsan0948/clinvar-acmg-distillation.acmg-variant
AmberTrace — ACMG Variant Classification
Clinical variant classification over ACMG evidence criteria. Features and prompts only — classifications are obtained live from AmberTrace.
AT = gold — this dataset ships no answers
The AmberTrace verifier is the answer. Every certified-answer column
(gold / oracle / decision / triage_reason / undecidable) has been
stripped from these files: a public (features → certified decision) map
would give the decision function away… See the full description on the dataset page: https://huggingface.co/datasets/AmberTraceLabs/acmg-variant.acmg-sentence-predictions