Sentieon
Sentieon develops high-performance, accurate software solutions for secondary analysis of genomic data, supporting various sequencing platforms and providing scalable workflows that significantly reduce analysis time and costs. Their tools are designed to meet the needs of the genomics industry for fast, reliable, and precise data processing.
Industries
Nr. of Employees
small (1-50)
Sentieon
Mountain View, California, United States, North America
Patents
Products
Short-read variant-calling pipeline (WGS/WES/Panel)
A high-performance secondary-analysis pipeline for short-read data that performs alignment, active-region detection, local assembly, and ML-based variant calling to produce VCF/GVCF outputs with reported high accuracy and reduced runtime versus common open-source workflows.
Long-read variant-calling and SV pipeline (HiFi)
Pipeline optimized for high-fidelity long reads that performs multi-round variant calling, phasing, and haplotype-resolved structural variant detection with published high accuracy and competitive runtimes.
Long-read variant-calling and SV pipeline (nanopore)
Pipeline optimized for nanopore chemistry that integrates a fast aligner and variant caller to produce VCF/GVCF and haplotype-resolved SV calls with lower error rates and reduced runtime compared to selected open-source tools.
Platform-adapted model packs for sequencing chemistries
Pre-trained model files and recommended workflow configurations tailored to specific sequencing platforms and capture kits for improved WGS/WES/Panel accuracy.
Short-read variant-calling pipeline (WGS/WES/Panel)
A high-performance secondary-analysis pipeline for short-read data that performs alignment, active-region detection, local assembly, and ML-based variant calling to produce VCF/GVCF outputs with reported high accuracy and reduced runtime versus common open-source workflows.
Long-read variant-calling and SV pipeline (HiFi)
Pipeline optimized for high-fidelity long reads that performs multi-round variant calling, phasing, and haplotype-resolved structural variant detection with published high accuracy and competitive runtimes.
Long-read variant-calling and SV pipeline (nanopore)
Pipeline optimized for nanopore chemistry that integrates a fast aligner and variant caller to produce VCF/GVCF and haplotype-resolved SV calls with lower error rates and reduced runtime compared to selected open-source tools.
Platform-adapted model packs for sequencing chemistries
Pre-trained model files and recommended workflow configurations tailored to specific sequencing platforms and capture kits for improved WGS/WES/Panel accuracy.
Services
Provision of licensed secondary-analysis pipelines for short- and long-read sequencing, deployable on cloud compute instances with platform-adapted ML models.
Provision and release of machine-learning model files tailored to specific sequencing platforms, read lengths, depths, and capture kits to improve variant-calling accuracy.
Provision of licensed secondary-analysis pipelines for short- and long-read sequencing, deployable on cloud compute instances with platform-adapted ML models.
Provision and release of machine-learning model files tailored to specific sequencing platforms, read lengths, depths, and capture kits to improve variant-calling accuracy.
Expertise Areas
- Secondary analysis of next-generation sequencing data
- Machine-learning driven variant calling and genotyping
- Cloud-native, CPU-optimized genomics workflows
- Long-read and short-read small-variant and structural-variant calling
Key Technologies
- Machine learning-based variant calling
- CPU-parallelized algorithms for genomics
- Cloud deployment on EC2-class instances (x86 and ARM)
- Haplotype-resolved structural variant calling