Publications

Benchmarks on public NanoPOTS and DISCO single-cell data: Tesorai Search identified up to 29% more unique peptides than FragPipe at 1% FDR, with match-between-runs off.

Our collaborators at UCSF have published their RNA editing study in Cell, using Tesorai Search to confirm 7,997 recoding sites at the protein level, more than an order of magnitude beyond previous reports.

Tesorai Search is now peer-reviewed and published in the Journal of Molecular Biology, detailing the model's design and benchmarking against other leading proteomics search engines.

Five public brain proteomics cohorts, reprocessed on Tesorai, support a seven-protein blood panel for Alzheimer's staging, including recovered p-tau217. Presented at HUPO 2026.

On a 94-case proteomics benchmark, Tesorai Chat reached 92.2% accuracy, ahead of Phylo, GPT-5.4 Codex, and Claude 4.6 Sonnet. Presented at HUPO 2026.

Our collaborators at MD Anderson Cancer Center have deposited a preprint on bioRxiv benchmarking narrow-window DIA with Tesorai Search, achieving the deepest proteome coverage among 12 acquisition-analysis pipelines tested while maintaining rigorous FDR control.

A new Tesorai preprint validates a proposed seven-protein blood biomarker panel for Alzheimer's staging by reanalyzing five independent brain proteomics cohorts, recovering a key biomarker, p-tau217, that was missing from the original studies.

Fine-tuning a single-cell foundation model on proteomics data matches or beats scaling to models over 40x larger — showing multimodal training can outperform parameter scaling alone.

Mass spectrometry (MS)-based proteomics enables high-throughput identification and quantification of proteins in complex biological systems.

Mass spectrometry (MS)-based proteomics is a poweful tool for high-throughput protein identification and quantification.

Accurate protein identification and quantification are central to extracting meaningful biological insights from mass spectrometry (MS)-based proteomics.

Our collaborators at NYU have deposited a preprint on bioRxiv describing the use of Tesorai for identifying cancer antigens that could be used for therapeutic targeting.