HKU-BAL/Clair3
Clair3 - Symphonizing pileup and full-alignment for deep learning-based long-read variant calling
Employs a dual-model architecture combining fast pileup-based calling with computationally intensive full-alignment models for complex variants, optimizing both speed and accuracy across coverage depths. Recently migrated to PyTorch with GPU acceleration (~5x speedup on Linux/Apple Silicon) and added signal-aware variant calling via Dorado move tables. Provides pre-trained models for multiple sequencing chemistries (Guppy 2-6, R10.4 Q20) and supports both standard germline VCF/GVCF output with modular training pipelines for custom datasets.
343 stars.
Stars
343
Forks
36
Language
Python
License
—
Category
Last pushed
Mar 12, 2026
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/HKU-BAL/Clair3"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
google/deepvariant
DeepVariant is an analysis pipeline that uses a deep neural network to call genetic variants...
cgoliver/rnaglib
Toolkit for structure-based deep learning on RNA.
Kuanhao-Chao/OpenSpliceAI
🤖 Open‑source deep-learning-based splice‑site predictor that decodes splicing patterns across species
google/deepsomatic
DeepSomatic is an analysis pipeline that uses a deep neural network to call somatic variants...
Merck/deepbgc
BGC Detection and Classification Using Deep Learning