Scientific software, built with the rigour of the science it serves.
Quantivum is a software company from Montreal and Toronto. We make MRMpipe, an automated peak-review platform for targeted LC/GC-MS, and we build databases, pipelines and dashboards for research labs that have outgrown spreadsheets and one-off scripts.
Origin
Fifteen years of the same bottlenecks.
Quantivum was founded by Dr. Peter Kubiniok after fifteen years building omics pipelines in core facilities and biotech. Every lab had the same friction: scattered spreadsheets, fragile scripts and hours of manual curation.
We exist to give that work proper engineering. Everything we deliver ships with tests, documentation and a container, so it keeps running after hand-over. MRMpipe applies that idea to the most repetitive task in targeted mass spectrometry: reviewing peaks.
We work remotely from Montreal and Toronto with labs in North America, Europe and Asia.
Founder
Peter Kubiniok, PhD
Proteomics scientist and software engineer. Author of twelve peer-reviewed papers on mass-spectrometry methods and immunopeptidomics, including work in Nature Communications and Molecular Systems Biology; maintainer of MHCvalidator, MhcVizPipe and the CRAN package RHybridFinder. Through a partnership with Mila, the Montreal AI institute, he built the machine-learning expertise behind Quantivum's review models.
- Background
- Omics pipelines at academic core facilities and biotech start-ups, 15 years
- Focus
- Targeted and discovery mass spectrometry, machine-learning rescoring, reproducible pipelines
- Links
- PubMed Google Scholar LinkedIn
How we work
What every engagement includes.
- Reproducibility
- CI tests, Conda or Docker environments and versioned data snapshots. The pipeline gives the same answer a year from now.
- Transparency
- Fixed-price scopes, weekly written progress reports, a shared Slack channel and an issue tracker you can read.
- Handover
- Documentation written for the next person, a training session for your team, and no lock-in: you own the code.
- Performance
- Pipelines that finish overnight, APIs that answer in milliseconds, ML inference lean enough to run on the lab's own hardware.
- 2024Machine-learning rescoring for immunopeptidomics (MHCvalidator)Nature Communications
- 2021MhcVizPipe: a quality-control dashboard for immunopeptidomicsMolecular & Cellular Proteomics
- 2015Systems-level analysis of phosphoproteome dynamicsMolecular Systems Biology
Working on something similar?
Tell us what your team needs and we'll scope it together.