our technology
Veracyte is setting new standards in translating complex genomic data into clinically meaningful information that changes patient care. We develop our genomic classifiers by leveraging innovations in clinical science, RNA sequencing and machine learning to answer specific clinical questions. Today, these answers can often only be obtained using patient samples obtained through surgery.

Our machine learning approach recognizes patterns of genes that correspond with “clinical truth” (for example: a benign thyroid nodule). We use next-generation RNA sequencing to extract rich feature sets – gene expression, DNA variants, RNA fusions, mitochondrial DNA content and loss of heterozygosity – from the RNA transcriptome of patient samples obtained through minimally invasive procedures. These are the “pixels” that enable us to create the highest-resolution genomic picture possible.

We train our proprietary machine learning algorithms to interpret this vast genomic data using large numbers of diverse patient samples that represent the broad spectrum of disease that our genomic tests may likely encounter in a clinical setting. In some cases, we deploy ensemble algorithms – or, “algorithms of algorithms” – to differentiate between complex biologies.

As scientific understanding continues to progress, additional features that inform disease status, such as more-refined genomic features or even imaging data, can potentially help us identify “clinical truth” with ever-more precision, enabling development of tests that provide even clearer answers to important clinical questions.