Machine learning implementation for multi-analyte assay development and testing
Inventors
Drake, Adam • Delubac, Daniel • Niehaus, Katherine • Ariazi, Eric • Haque, Imran • Liu, Tzu-Yu • Wan, Nathan • KANNAN, Ajay • White, Brandon
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-11847532-B2
Publication Date
2023-12-19
Expiration Date
Abstract
Systems and methods that analyze blood-based cancer diagnostic tests using multiple classes of molecules are described. The system uses machine learning (ML) to analyze multiple analytes, for example cell-free DNA, cell-free microRNA, and circulating proteins, from a biological sample. The system can use multiple assays, e.g., whole-genome sequencing, whole-genome bisulfite sequencing or EM-seq, small-RNA sequencing, and quantitative immunoassay. This can increase the sensitivity and specificity of diagnostics by exploiting independent information between signals. During operation, the system receives a biological sample, and separates a plurality of molecule classes from the sample. For a plurality of assays, the system identifies feature sets to input to a machine learning model. The system performs an assay on each molecule class and forms a feature vector from the measured values. The system inputs the feature vector into the machine learning model and obtains an output classification of whether the sample has a specified property.
Core Innovation
The invention is a method of screening an individual for a colorectal advanced adenoma by assaying a plurality of classes of molecules in a biological sample from the individual using a plurality of assays. The biological sample is whole blood, plasma, or serum, and the plurality of classes includes cell-free DNA (cfDNA) as a first class of nucleic acids and endogenous polyamino acids as a second class, with a first assay applied to the cfDNA that includes methylation sequencing.
Measured values from each class are converted into features corresponding to properties of each class and used to form a feature vector of feature values. The feature vector includes at least one feature value obtained using each set of measured values representative of the plurality of classes of molecules, with each feature value corresponding to a feature of the set of features and including one or more measured values.
The method loads, into memory of a computer system, a machine learning model trained using training vectors from training biological samples labeled as having the colorectal advanced adenoma or not having the colorectal advanced adenoma. The prepared feature vector is input into the machine learning model to obtain an output classification of whether the individual has the colorectal advanced adenoma.
Claims Coverage
One independent claim is provided, covering colorectal advanced adenoma screening based on multi-class, multi-assay molecular measurements from a whole blood, plasma, or serum sample and machine-learning classification using a combined feature vector. The independent claim includes 5 main inventive features.
Multi-class assaying with cfDNA methylation sequencing
Assaying a plurality of classes of molecules in a biological sample using a plurality of assays to obtain sets of measured values representative of the plurality of classes of molecules, wherein the biological sample is whole blood, plasma, or serum; the plurality of classes includes cfDNA and endogenous polyamino acids; and a first assay applied to the cfDNA includes methylation sequencing.
Feature extraction for properties of each molecular class
Identifying a set of features corresponding to properties of each of the plurality of classes of molecules to be input to a machine learning model.
Combined feature vector from multiple assay-derived measured values
Preparing a feature vector of feature values from the plurality of sets of measured values representative of the plurality of classes of molecules, where each feature value corresponds to a feature of the set of features and includes one or more measured values.
Loading a trained machine learning model with labeled training subsets
Loading, into a memory of a computer system, the machine learning model trained using training vectors obtained from training biological samples, with a first subset identified as having the colorectal advanced adenoma and a second subset identified as not having the colorectal advanced adenoma.
Machine-learning classification of colorectal advanced adenoma
Inputting the feature vector into the machine learning model to obtain an output classification of whether the individual has the colorectal advanced adenoma.
Overall, the independent claim covers integrating assay-derived measured values from multiple molecular classes into a combined feature vector, using a machine-learning model trained on labeled adenoma and non-adenoma samples to output an adenoma classification.
Stated Advantages
Not explicitly described in patent.
Documented Applications
Screening an individual for a colorectal advanced adenoma using multi-class, blood-based molecular measurements and machine-learning classification.
CRC/colorectal cancer and advanced adenoma example cohorts are referenced in the partial content as used to demonstrate the framework.
Interested in licensing this patent?