Methods and systems for diagnosing from whole genome sequencing data

A processor-based method using Gaussian mixture models for analyzing whole genome sequencing data addresses the challenge of genotyping SMN1/SMN2 and CYP2D6/CYP2D7 genes, enabling precise copy number determination and improved diagnostic and treatment strategies.

AU2020341336B2Pending Publication Date: 2026-07-16ILLUMINA INC
View PDF 2 Cites 0 Cited by

Patent Information

Authority / Receiving Office
AU · AU
Patent Type
Applications
Current Assignee / Owner
ILLUMINA INC
Filing Date
2020-08-26
Publication Date
2026-07-16

AI Technical Summary

Technical Problem

Genotyping of SMN1 and CYP2D6 genes is challenging due to their high sequence similarity with their paralogs SMN2 and CYP2D7, respectively, leading to difficulties in accurately determining copy numbers and genotyping.

Method used

A processor-implemented method using Gaussian mixture models to analyze whole genome sequencing data, aligning sequence reads to specific gene regions, and determining normalized numbers to accurately calculate copy numbers of SMN1/SMN2 and CYP2D6/CYP2D7 genes, followed by identifying the most likely combination of copy numbers based on alignment support and posterior probabilities.

Benefits of technology

Enables precise determination of SMN1 and CYP2D6 gene copy numbers and genotyping, facilitating accurate diagnosis and treatment recommendations for spinal muscular atrophy and drug metabolism, respectively.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 00000001_0000
    Figure 00000001_0000
  • Figure 00000119_0000
    Figure 00000119_0000
  • Figure 00000119_0001
    Figure 00000119_0001
Patent Text Reader

Abstract

Disclosed herein include systems, devices, computer readable media, and methods for paralog geno typing, such as determining a copy number of survival of motor neuron 1 gene and genotyping cytochrome P450 family 2 subfamily D member 6 gene using a Gaussian mixture model comprising a plurality of Gaussians each representing a different integer copy number.
Need to check novelty before this filing date? Find Prior Art