Method for the Diagnosis And/or Classification of a Disease in a Subject

Simple SummaryContent extracted from patent full text and abstract with AI.

This invention provides a rapid, flexible method for diagnosing and classifying diseases (such as cancers, autoimmune disorders, and infections) in a subject by analyzing genetic and/or epigenetic information from a biological sample. The method collects genetic or epigenetic data (such as DNA methylation, mutations, or structural genomic variations) from a random subset of the genome—using technologies like nanopore sequencing—and inputs this data in real-time to a computational model (e.g., a Naive Bayes classifier). The model has been pre-trained on reference samples and can assign a probable disease class for the patient even when only partial, sparse, or shallow coverage genetic data is available. This enables potentially intraoperative or same-day diagnosis for clinical decision-making.

Use CasesContent extracted from patent full text and abstract with AI.

• Intraoperative molecular diagnosis of brain tumors during neurosurgery, enabling immediate surgical decisions.
• Real-time classification and diagnosis of various cancers (e.g., brain, lung, breast) from biopsy or liquid biopsy samples.
• Noninvasive disease detection from circulating cell-free DNA (cfDNA) in blood or cerebrospinal fluid for cancer or infectious diseases.
• Rapidly distinguishing between different types or subtypes of diseases (such as cancer subtyping or autoimmune disorders) for tailored treatment strategies.
• Point-of-care diagnostics in emergency or intensive care situations, e.g., identifying infectious or septicemia causes.
• Monitoring for disease relapse or progression through minimally invasive sampling (e.g., liquid biopsy follow-up).

BenefitsContent extracted from patent full text and abstract with AI.

• Significantly faster diagnosis (can be performed within minutes to an hour), supporting intraoperative and point-of-care decisions.
• Works with real-time, sparse, or shallow sequencing data; no need for complete genome coverage or preselected biomarkers.
• Does not require retraining or updating the computational model for each new sample, enhancing regulatory approval and product stability.
• Can integrate diverse types of genomic and epigenetic data (methylation, mutations, copy number variations, etc.), improving diagnostic accuracy.
• Minimally invasive sampling possible (e.g., from blood, CSF), widening clinical applicability including for patients where tissue biopsies are risky.
• Enables immediate stratification and subclassification of disease, supporting precision medicine and individualized treatments.
• Reduces laboratory time, technical complexity, and costs compared to standard diagnostic workflows.

Technical Classifications (CPCs)

Inventors & Applicants

Patent Abstract

The present invention relates to a method for the diagnosis and/or classification of a disease in a subject based on the genetic and/or epigenetic information of a sample obtained from the subject, the method comprising the steps of: a) providing data from said sample, wherein said data comprises genetic and/or epigenetic information of a random subset of genomic positions; b) assigning said sample to a sample class based on genetic and/or epigenetic information of said random subset of genomic positions by employing a computational model, which discriminates a plurality of sample classes based on genetic and/or epigenetic information of a set of genomic positions comprising said random subset, wherein the computational model has been trained with pre-determined genetic and/or epigenetic information obtained from a plurality of pre-classified samples of known diseases and wherein said computational model processes the genetic and/or epigenetic information of a genomic position of said random subset independently of the genetic and/or epigenetic information of another genomic position of said random subset, wherein said computational model is preferably in the form of a linear classifier with independent feature sampling.