Ruibang Luo

Clair3-Trio improves variant calling in family trios using Nanopore long-reads with a Trio-to-Trio neural network and a novel loss function.

SENSV detects structural variations with high sensitivity and precise breakpoints using low-depth Oxford Nanopore WGS data from a single MinION flowcell, outperforming existing tools in pathogenic SV detection.

HKG is a variant database of 205 Hong Kong Cantonese exomes that fills a gap in southern Chinese genetics, showing unique variants and improving variant imputation.

RENET2 extracts gene-disease associations from full-text articles with higher accuracy than existing tools, identifying 3.72M associations from 1.89M PMC articles.

The paper tracks cytosine depletion in SARS-CoV-2 over time using genomic data from GISAID and finds a monthly loss of cytosine correlated with mortality.

Clair is a germline small variant caller using single-molecule sequencing data that achieves high precision, recall, and speed, approaching the accuracy limit for pileup-based deep learning methods.

Skyhawk is an AI tool that quickly reviews genomic variants, reducing manual review time by identifying false positives and flagging uncertain variants for further analysis.

Clairvoyante is a deep learning model for variant calling in single molecule sequencing that achieves high accuracy across multiple sequencing technologies.

16GT is a variant caller that uses a 16-genotype probabilistic model to improve speed and sensitivity for SNP calling while maintaining comparable indel sensitivity and accuracy.

LRSim generates linked reads to evaluate genome partitioning strategies and provides recommendations for sequencing parameters.