cfDNA Domain Adaptation for Tissue-of-Origin Deconvolution

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current methods for detecting diseased tissue-of-origin, such as ATAC-Seq, require invasive biopsies and are not scalable for routine diagnostics, while cfDNA-based approaches suffer from sensitivity limitations and incomplete genomic coverage, especially for early-stage diseases.

Innovation Solution

A domain adaptation engine that integrates chromatin-informed models with cfDNA fragmentation data, leveraging ATAC-Seq data to deconvolute cfDNA and predict tissue-of-origin using machine learning, enabling detection of diseased tissue at lower fractions without invasive sampling.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If ATAC-Seq is used to detect diseased tissue-of-origin, then measurement precision is improved, but ease of operation deteriorates due to invasive biopsies

Engineering Contradiction:

Improvetissue-of-origin detection accuracyVSAvoidinvasive biopsy requirement

Core Design Contradiction:

Measurement precisionVSEase of operation

Solution Approach 1:

The patent uses cfDNA as an intermediary substance that carries chromatin accessibility information from diseased tissue to the bloodstream, where it can be non-invasively sampled. The domain adaptation engine then acts as a mediator to translate cfDNA fragmentation patterns into tissue-of-origin predictions by mapping them to reference ATAC-Seq data, thereby eliminating the need for direct tissue biopsy while preserving diagnostic accuracy

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent creates a computational copy of ATAC-Seq chromatin accessibility signatures from reference tissues and stores them in a database. The domain adaptation engine compares cfDNA fragmentation patterns against these copied reference signatures to infer tissue-of-origin, effectively replicating the information obtained from direct ATAC-Seq without requiring actual tissue sampling

Inventive Principle:

Principle #26Copying

2Ease of operation

If cfDNA-based approaches are used for disease detection, then ease of operation is improved by avoiding biopsies, but measurement precision deteriorates due to sensitivity limitations

Engineering Contradiction:

Improvenon-invasive samplingVSAvoiddetection sensitivity

Core Design Contradiction:

Ease of operationVSMeasurement precision

Solution Approach 1:

The patent transforms the analysis parameter from direct cfDNA concentration or mutation detection to cfDNA fragmentation pattern analysis. By changing the measured parameter to nucleosome positioning patterns and fragment size distributions that reflect chromatin accessibility, the system achieves higher sensitivity for detecting low-abundance diseased tissue signals while maintaining non-invasive sampling

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The patent combines multiple data sources into a composite analytical framework: cfDNA fragmentation data from the patient sample is integrated with reference ATAC-Seq data from multiple tissue types, and both are processed through the domain adaptation engine that merges chromatin accessibility information with fragmentation patterns to produce enhanced tissue-of-origin predictions with improved sensitivity

Inventive Principle:

Principle #40Composite materials

3Measurement precision

If domain adaptation engine deconvolutes cfDNA using ATAC-Seq data, then measurement precision is improved for tissue-of-origin identification, but device complexity increases

Engineering Contradiction:

Improvetissue-of-origin prediction accuracyVSAvoidmachine learning system complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent performs preliminary action by pre-computing and storing chromatin accessibility signatures from ATAC-Seq data of multiple reference tissue types in a database before clinical use. The domain adaptation engine is pre-trained on these reference data to learn the mapping between chromatin patterns and tissue identities. This preliminary preparation eliminates the need for complex real-time computations during patient testing, reducing operational complexity while maintaining high prediction accuracy

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS20260018291A1Domain Adaptation Engine(s) For Cell-Free DNA Fragmentomics

Publication Date: 2026.01.15 THE REGENTS OF THE UNIVERSITY OF COLORADO

AI summary

Systems and methods for predicting tissue-of-origin for diseased tissues using cell-free DNA (cfDNA) are provided herein. In an aspect, a domain adaptation engine receives a cfDNA sample from a subject and generates cfDNA fragmentation data from the cfDNA sample. The domain adaptation engine deconvolutes the cfDNA fragmentation data using Assay for Transposase-Accessible Chromatin using sequencing (ATAC-Seq) data from multiple tissue and cell types to generate deconvoluted cfDNA fragmentation data. In an example, the domain adaptation engine deconvolutes the cfDNA fragmentation data using a machine learning (ML) system trained to translate between the ATAC-Seq data and cfDNA data. Subsequently, the domain adaptation engine detects a diseased tissue signature within the deconvoluted cfDNA fragmentation data and generates a tissue or cell-type-of-origin prediction for the diseased tissue signature based on the deconvoluted cfDNA fragmentation data.