Case studies.
Clients come to us when bioinformatics, software engineering, and biology all have to work together and the results have to be right: clinical diagnostics, new sequencing chemistries, and non-model species.
Each study covers what the client needed, what we built, and what it changed.
Clinical diagnostics.
Bringing the Summit™ liquid biopsy to clinical launch
Belay had a promising error-corrected NGS technology for brain and spinal cancers, and needed to turn it into a CAP/CLIA-grade clinical pipeline with auditable releases and cloud architecture built to scale. We embedded two scientists with their team to move the assay from prototype to clinical deployment.
“Fulcrum has been an essential partner in shaping our clinical-grade workflows and processes. This partnership has empowered our ability to deliver patients reliable, high-quality results with confidence.”
Kyle Hernandez, Ph.D., Sr. Director of Translational Informatics, Belay Diagnostics
What we built
- A production error-corrected sequencing pipeline with high variant detection accuracy
- Runtime safety checks plus unit, integration, and end-to-end tests to lock in performance
- A redesigned cloud dataflow that eliminated resource exhaustion and reduced costs
- Versioned, auditable release infrastructure with separate dev / staging / production environments
Sequencing chemistry.
Independently validating the TrueLink™ library prep
Canal’s dual direct ligation chemistry skips end-repair and A-tailing, which targets the elevated error rates seen in the first cycles of short-read sequencing. Ahead of the TrueLink™ launch, they asked us to take a close look at the chemistry, going beyond aggregate QC to measure error by cycle, by strand, and by reference base context.
“From day one, Fulcrum Genomics operated like an extension of our team — iterating quickly with us as our protocols evolved and translating cycle-resolved error signatures into clear, actionable insights.”
Yu Zheng, Ph.D. — Founder & co-CEO, Canal Biosciences
What we built
- Built a flexible, tested analysis framework that kept pace with rapidly evolving lab protocols
- Broke down sequencing error by cycle, strand, and reference base context
- Used ssDNA spike-ins and well-characterized reference materials to separate true biology from sequencing artifact
- Confirmed chimera rate, mappability, and coverage uniformity remained equivalent or improved
Population genomics.
Building the largest feline imputation reference panel
Darwin’s Ark is a citizen-science project linking feline genetics to traits and behavior, with 20,000+ cats enrolled. Cat genomics infrastructure, however, lags far behind human and agricultural species. Working with the Darwin’s Ark team, we built the reference panel and pipelines that make low-pass sequencing viable at population scale.
“Fulcrum has been a game changer for us. We now have pipelines that outperform anything else available and produce data of such high quality that we’ve expanded our vision for the entire project.”
Elinor Karlsson — Co-Founder & Chief Scientist, Darwin’s Ark
What we built
- Aggregated and standardized 1,100+ cat genomes from public repositories and sequencing efforts
- Joint variant calling and phasing with rigorous QC for coverage, duplication, relatedness, and missingness
- A containerized GLIMPSE2-based imputation pipeline built for reproducibility and scale
- Validation that reproduced known genotype–phenotype associations (coat color, fur length, tabby)