“…Sequencing data were analyzed via the following advanced machine learning-based classifiers (all developed/licensed by Veracyte, Inc., South San Francisco, CA, USA) that are applied to the diagnosis of thyroid cancer, interstitial lung disease (ILD), and lung cancer, and may be applicable to other disease indications: the Afirma ® Genomic Sequencing Classifier (GSC), which uses RNA exome sequencing information derived from a fine-needle aspiration (FNA) biopsy to classify cytologically indeterminate thyroid nodules as Afirma GSC Benign or Suspicious [ 12 , 22 , 23 , 24 , 25 ]; the Afirma Xpression Atlas (XA), which provides genomic alteration information from FNA samples with an Afirma GSC Suspicious result, or cytologically suspicious for malignancy or malignancy nodule [ 26 ]; the Envisia ® Genomic Classifier (GC), a diagnostic test that uses a 190-gene expression signature to differentiate between usual interstitial pneumonia (UIP) and non-UIP subtypes in lung transbronchial biopsy (TBB) samples from patients with ILD [ 13 ]; the Percepta GSC, which uses RNA exome sequencing information from bronchial brushing samples for classification of lung cancer risk among bronchoscopy nondiagnostic samples [ 16 ]; the Percepta Nasal Swab (NS) classifier that uses RNA exome sequencing information from nasal (inferior turbinate) samples to classify the risk of lung cancer in patients with previously detected lung nodules [ 27 ]; and the Percepta Genomic Atlas (GA) test, which was developed to detect lung cancer-associated fusions and MET exon-skipping events in diagnostic TBB specimens and transbronchial needle aspirates (TBNAs) [ 28 ]. Beyond expression signatures, interrogation of RNA-Seq data in these samples revealed specimens with rare and novel genomic events, including gene fusions, mitochondrial variants, and loss of heterozygosity (LOH).…”