Granted Patent B2

US 12,462,935 · App. 17/043,629 · Granted Nov 4, 2025

Inventors: Mathias Ehrich (San Diego, CA); Lawrence Du (San Diego, CA); Dirk Van Den Boom (Encinitas, CA)

Assignee: Nucleix Ltd.

G16H50/20G16H10/60G16H50/30G16H50/70

Abstract

Claims (22)

1 . A method comprising:

a) obtaining or providing a biological sample obtained from a subject, wherein the biological sample comprises circulating cell-free nucleic acid molecules;

b) amplifying at least a portion of the circulating cell-free nucleic acid molecules to generate amplified cell-free nucleic acid molecules;

c) sequencing at least a portion of the amplified cell-free nucleic acid molecules to produce a set of sequencing reads;

d) processing each sequencing read in the set of sequencing reads to generate one or more probability values using a machine learning algorithm, wherein the machine learning algorithm is trained using an input data set comprising one or more sets of sequencing reads from subjects of different classes, and wherein the one or more probability values represents a probability that the set of sequencing reads corresponds to a particular class of the different classes;

e) classifying the set of sequencing reads as belonging to a particular class when the one or more probability values comprise at least one probability of the particular class; and

f) detecting a presence or an absence of at least one genetic abnormality in the subject, or fetus if the subject is pregnant, based on the classifying of e).

2 . The method of claim 1 , wherein the processing of d) does not include alignment of the set of sequencing reads to a reference genome or reference sequence.

3 . The method of claim 1 , wherein the processing of d) comprises a calculation of the length of each sequencing read in the set of sequencing reads, the GC content of each sequencing read in the set of sequencing reads, a value corresponding to the number and ordering of nucleotide bases in each sequencing read in the set of sequencing reads, a feature weighting factor, or any combination thereof.

4 . The method of claim 1 , wherein the machine learning algorithm is a deep learning algorithm.

5 . The method of claim 4 , wherein the deep learning algorithm comprises a feedforward neural network, a convolutional neural network, or a recurrent neural network.

6 . The method of claim 4 , wherein the input data set comprises one or more sets of sequencing reads from one or more control subjects comprising known euploid subjects, known aneuploid subjects, or any combination thereof.

7 . The method of claim 4 , wherein the input data set comprises in silico sequence data obtained from a publicly-available database, a private institutional database, a commercial database, or any combination thereof.

8 . The method of claim 4 , wherein the input data set comprises simulated sequence data for normal subjects, abnormal subjects, or any combination thereof.

9 . The method of claim 4 , wherein the input data set comprises personal health data for one or more control subjects, wherein the personal health data is selected from the group consisting of subject age, gestational age, sex, weight, blood pressure, number of previous offspring (if female), ultrasound markers, biochemical screening results, smoking history, history of alcohol use, family history of disease, or any combination thereof.

10 . The method of claim 1 , wherein the input data set further comprises values corresponding to personal health data for the subject that is selected from the group consisting of subject age, gestational age, sex, weight, blood pressure, number of previous offspring (if female), ultrasound markers, biochemical screening results, smoking history, history of alcohol use, family history of disease, or any combination thereof.

11 . The method of claim 4 , wherein the input data set resides in a cloud-based database that is periodically or continuously updated with sets of sequencing reads, input data sets, and previously-performed deep learning analysis results that are generated locally or remotely.

12 . The method of claim 1 , wherein the at least one genomic abnormality comprises a copy number variation, a full or partial duplication of at least one chromosomal arm, a full or partial deletion of at least one chromosomal arm, or any combination thereof.

13 . The method of claim 1 , wherein the biological sample is blood, plasma, serum, urine, interstitial fluid, vaginal cells, vaginal fluid, buccal cells, or saliva.

14 . The method of claim 1 , wherein the particular class of the set of sequencing reads in d) is not determined with respect to a specific target chromosome.

15 . The method of claim 1 , wherein the one or more probability values represents a probability that a sequencing read of the set of sequencing reads corresponds to a particular genomic region.

16 . The method of claim 1 , wherein the machine learning algorithm generates a probability vector for each sequencing read of the set of sequencing reads.

Assignments (5)

Continuity (3)

References Cited (342)