Platform
Provotics takes a single RNA-seq expression profile and returns a structured, calibrated, explainable report, the kind of read that usually takes several separate assays and a lot of manual interpretation.
Six linked predictions from the same transcriptome, each returned with its confidence and the genes behind it.
Classifies the tumor across 25 anatomical sites, ~94% balanced accuracy on held-out tumors.
Resolves intrinsic subtype where it applies, for example PAM50 on breast tumors.
Surfaces over-expressed, druggable targets as research hypotheses toward candidate therapies.
Estimates the tumor immune context, from immune-cold to inflamed.
Every call carries a calibrated probability, and low-confidence samples are flagged or abstained on.
Each prediction is traceable to the driver genes behind it, so it stays reviewable.
Start from a single tumor expression profile, gene-level counts or TPM.
The pipeline maps to a common gene space, normalizes, and screens quality before any prediction.
An ensemble scores and calibrates the profile and returns the full report, with the genes behind each call.
Explore a live sample report on the homepage, then request access to run your own profiles under our access agreement.
View the sample report