Algorithms for disease diagnostics
Inventors
Kennedy, Giulia C. • Chudova, Darya I. • Wang, Eric T. • Wilde, Jonathan I. • Anderson, Bonnie H. • Wang, Hui • Pagan, Moraima • Rabbee, Nusrat
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-10672504-B2
Publication Date
2020-06-02
Expiration Date
Abstract
The present invention relates to compositions and methods for molecular profiling and diagnostics for genetic disorders and cancer, including but not limited to gene expression product markers associated with cancer or genetic disorders. In particular, the present invention provides algorithms and methods of classifying cancer, for example, thyroid cancer, methods of determining molecular profiles, and methods of analyzing results to provide a diagnosis.
Core Innovation
A method processes a sample of thyroid tissue of a subject by sequencing nucleic acid molecules from the sample to yield data comprising one or more levels of gene expression products. The gene expression product levels correspond to a plurality of genes selected from ALK, CALCA, DICER1, IGF2BP3, MET, NTRK1, NTRK3, PAX8, PTEN, PTH and THADA.
A trained algorithm in a computer processes the gene expression product data to generate a classification of the thyroid tissue sample as positive or negative for thyroid cancer. The trained algorithm is trained with a plurality of training samples comprising known benign samples and known non-benign samples different from the sample of tissue, and the classification is generated at an accuracy of at least 85%.
The method electronically outputs a report that identifies the classification of the thyroid tissue sample as positive or negative for thyroid cancer. The disclosed analysis framework also describes a large gene-marker set associated with thyroid tissue subtype labels, including benign vs malignant categories such as NHP, FA, LCT, FVPTC, PTC, and rest, and a "Trident Analysis" output providing 210 unique TCIDs mapping to 237 genes from a union of three statistical sub-analyses.
Claims Coverage
The disclosed independent claim is directed to a thyroid-tissue sample processing and analysis method that sequences nucleic acids, derives gene-expression levels for specified genes, applies a trained algorithm to generate a positive or negative thyroid-cancer classification at a stated accuracy threshold, and electronically outputs a report identifying the classification. Three inventive features are identified across the independent claim.
Sequencing gene-expression data from thyroid tissue for specified genes
Sequencing nucleic acid molecules from a sample of thyroid tissue to yield data comprising one or more levels of gene expression products corresponding to a plurality of genes selected from ALK, CALCA, DICER1, IGF2BP3, MET, NTRK1, NTRK3, PAX8, PTEN, PTH and THADA.
Trained algorithm classification as thyroid cancer positive or negative at an accuracy threshold
Using a trained algorithm in a computer to process the gene-expression data to generate a classification of the thyroid tissue sample as positive or negative for thyroid cancer at an accuracy of at least 85%, where the trained algorithm is trained with training samples comprising known benign samples and known non-benign samples different from the sample of tissue.
Electronically outputting a report identifying the classification
Electronically outputting a report that identifies the classification of the thyroid tissue sample as positive or negative for thyroid cancer.
The core inventive coverage centers on producing gene-expression data from thyroid tissue for specified genes, applying a trained algorithm trained on known benign and non-benign samples to generate a positive or negative thyroid-cancer classification meeting a stated accuracy threshold, and electronically outputting a report identifying the classification.
Stated Advantages
Generates a positive/negative classification for thyroid cancer at an accuracy of at least 85%.
Documented Applications
Diagnosing thyroid cancer status of a thyroid tissue sample as positive or negative using gene-expression data and a trained algorithm, with an electronically output report identifying the classification.
Interested in licensing this patent?