About

Between biology and computation.

We are an AI-powered proteomics company. We read proteins out of mass spectrometry data and turn what comes back into something a scientist can actually use.

Mission

We turn proteins into data, making the molecular foundations of life accessible in the era of personalized medicine.

Purpose

Unlocking the dark proteome.

A large share of the proteome is effectively unreadable by conventional methods. A database search can only tell you about proteins somebody has already written down, so anything new comes back as nothing at all. More often than not, that is the part you were looking for.

The proteins nobody has catalogued yet are usually the ones worth finding.

The problem

Mass spectrometry produces enormous volumes of tandem spectra. Interpreting them has traditionally meant matching each spectrum against a protein database, an approach that is fast and well understood, but bounded by the database itself.

The result is a workflow that is database-dependent, expert-heavy, and weakest exactly where discovery matters most.

Existing approaches are constrained by database dependence, fragile interpretation logic, and a high expert burden.

Our approach

Seqora sequences peptides directly from their spectra. A transformer-based de novo model proposes the sequence that produced each spectrum; a rescoring and calibration layer attaches a confidence estimate and controls the false discovery rate; and the results roll up to peptide and protein level evidence that stays inspectable all the way back to the raw ions.

Just as important is that this arrives as software. Projects, runs, filtering, quality control and spectrum inspection sit in one application that a proteomics group can actually operate, locally or on a cluster, rather than a model that needs a specialist to run it.

Raw spectra in. Peptides, proteins, and the evidence behind them out.

Where we are

Seqora is an early-stage company built around a working product and published, independently benchmarked science. The near-term focus is narrow on purpose: make the software operational and scalable for standard proteomics workflows and protein sequencing.

Beyond the standard workflows, these all run on the same platform today, and each keeps getting sharper as the models and the workflows around them improve.

  • Immunopeptidomics
  • Metaproteomics
  • Antibody sequencing
  • PTM-heavy workflows
  • Degradomics

Biology comes first, the software follows it.

From protein to data

If you work with mass spectrometry data and the database is the thing standing in your way, we would like to hear from you.