Fluctuating Methylation Clock for Hematological Condition Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current diagnostic methods for hematological diseases, particularly clonal hematopoiesis (CH), are inadequate as they fail to detect CH in patients with unknown oncogenic drivers or structural genomic variants, and they cannot determine the aggressiveness of CH clones or their life histories.
Innovation Solution
A computer-implemented method using fluctuating methylation clock (FMC) data and a trained machine learning model to predict hematological conditions, including CH, by analyzing DNA methylation fluctuation data from blood specimens.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If DNA sequencing or SNP arrays are used to diagnose hematological diseases, then point mutations with variant frequencies ≥2% can be detected, but structural variants (copy number and translocations) and patients with unknown oncogenic drivers cannot be detected
Solution Approach 1:
The patent changes the detection parameter from sequence-based (DNA sequencing/SNP arrays) to methylation-based (Fluctuating Methylation Clock). This parameter change enables detection of all CH types including structural variants and unknown drivers, as methylation fluctuations occur regardless of the underlying genetic alteration mechanism.
Solution Approach 2:
The patent uses DNA methylation levels as an intermediary marker to indirectly detect clonal hematopoiesis. Instead of directly sequencing DNA to find mutations, the methylation clock serves as a mediator that reflects clonal expansion through epigenetic changes, capturing CH cases that escape traditional mutation-based detection.
2Measurement precision
If traditional diagnostic methods are used, then presence/absence of CH or MDS can be determined, but aggressiveness of CH clones and life histories cannot be determined
Solution Approach 1:
The patent segments the CH diagnosis into multiple dimensions: detection of clonal expansion, characterization of clone aggressiveness, and estimation of life history. By analyzing specific methylation fluctuation patterns across different CpG sites, the system provides granular information about clone behavior and prognosis, not just presence/absence.
Solution Approach 2:
The patent adds new dimensions to CH characterization by introducing methylation-based metrics that capture temporal dynamics and clonal heterogeneity. This dimensional expansion enables differentiation between indolent and aggressive clones, and estimation of clone age, providing prognostic information that was previously inaccessible.
3Productivity
If molecular diagnostics are used to screen for point mutations, then known driver mutations can be identified, but structural genomic variants that are strong prognostic indicators are missed
Solution Approach 1:
The patent changes from sequence-based detection to methylation-based detection, capturing all types of genomic alterations through their epigenetic consequences. This parameter change ensures that both known drivers and structural variants are detected, improving prognostic accuracy without sacrificing screening efficiency.
Data Source
AI summary
Systems and methods for predicting hematological conditions using methylation data are described herein. An example computer-implemented method includes: receiving patient data associated with a blood specimen from a subject, the patient data including fluctuating methylation clock (FMC) data; inputting the FMC data into a trained machine learning model; and predicting, using the trained machine learning model, a hematological condition in the subject.