Genome sequencing and AI analysis reveal hidden genetic markers of breast cancer microbiologystudy

Hidden genetic markers of breast cancer discovered in new study
Variant analysis pipeline. (A) Preprocessing flow. (B) ML scoring EVE (EVEmodel), ESM (ESM-1v), FIRM (functional impact rating at the molecular level) and structural algorithm DIST. (C) Di and Firepresent the dataset and family ranking of variant i⁠, m the number of ML models 𝟙1ij indicator of whether variant i was scored by model j⁠, Si(j) the variant normalized sore of model j (min max normalization over entire dataset), Mi is given by 𝟙∑j=1m1ij⁠, μ and σ are the mean and SD of DSRank over the entire dataset. (D) Analysis flow. Credit: Briefings in Bioinformatics (2024). DOI: 10.1093/bib/bbae346

A recent study has developed a novel method to analyze genetic variations in families with a high incidence of breast cancer. By examining 1,218 genetic variants in 12 families, the research identified 80 genes linked to an increased risk of the disease. It also highlighted the significant—yet previously overlooked—roles of peroxisomal and mitochondrial pathways in breast cancer predisposition and patient survival.

Breast cancer is the most common malignancy among Western women, with up to 10% of cases attributed to genetic variants. Despite this, the roots of many familial cases remain unexplored, largely due to the complex nature of the genetic factors involved.

Addressing this critical gap, a recent study led by Prof. Dina Schneidman-Duhovny from the Rachel and Selim Benin School of Computer Science and Engineering at the Hebrew University of Jerusalem has provided new insights into the genetic underpinnings of familial breast cancer, especially prevalent in families of Middle Eastern descent.

The paper is published in the journal Briefings in Bioinformatics.

The study utilizes an innovative analysis method tailored for examining genetic variations in families with a history of breast cancer. This method combines cutting-edge machine learning with detailed analysis of protein structures to investigate rare genetic variants.

Through the examination of 1,218 variants found among members of 12 families, researchers identified 80 genes that could significantly influence breast cancer risk. This discovery includes 70 genes previously unknown to be linked to breast cancer, significantly expanding our understanding of the genetic landscape of the disease.

Hereditary or familial breast cancer accounts for about 15% of all breast cancer cases. Historically, mutations in well-known genes like BRCA1 and BRCA2 have been linked to increased risks of familial breast and ovarian cancer. Yet, they only account for about 30%-40% of familial breast cancer cases. This leaves a substantial number of cases with unknown genetic origins, particularly in families where the illness is evident across generations.

The study revealed key roles for certain cellular pathways related to peroxisomes and mitochondria in predisposing individuals to breast cancer and affecting patient survival. These pathways were found to be particularly significant across a diverse range of ethnic groups in seven of the families studied, highlighting the broader applicability and importance of the findings.

The researchers used full genome sequencing and AI analysis to study genetic variations in women from Middle Eastern families. This approach identified significant genetic changes, linking subgroups of genes to critical cellular pathways involving peroxisomes, which play a key role in fat metabolism.

Prof. Schneidman-Duhovny remarked, “Our research not only sheds light on the elusive genetic factors behind familial breast cancer but also heralds the possibility of new, targeted treatment strategies that could eventually benefit a wider array of patients, particularly those from underrepresented groups.”

These discoveries open up potential avenues for genetic testing and the development of targeted therapies, promising a significant impact on the management and treatment of breast cancer across diverse populations. Additionally, the findings may eventually support the creation of a specialized genetic testing panel for these patient groups, enhancing early detection and personalized treatment plans as research progresses.

More information:
Gal Passi et al, Discovering predisposing genes for hereditary breast cancer using deep learning, Briefings in Bioinformatics (2024). DOI: 10.1093/bib/bbae346

Provided by
Hebrew University of Jerusalem


Citation:
Genome sequencing and AI analysis reveal hidden genetic markers of breast cancer (2024, October 31)
retrieved 31 October 2024
from https://medicalxpress.com/news/2024-10-genome-sequencing-ai-analysis-reveal.html

This document is subject to copyright. Apart from any fair dealing for the purpose of private study or research, no
part may be reproduced without the written permission. The content is provided for information purposes only.



Source link

Leave a Comment

Your email address will not be published. Required fields are marked *

Scroll to Top