Related Experiment Video
Updated: Jul 30, 2025
Performing Data Mining And Integrative Analysis Of Biomarker in Breast Cancer Using Multiple Publicly Accessible Databases
Published on: May 17, 2019
Mutational signatures for breast cancer diagnosis using artificial intelligence
Patrick Odhiambo1, Harrison Okello2, Annette Wakaanya3
1Department of Biological Sciences, School of Natural and Applied Sciences, Masinde Muliro University of Science and Technology, P.O. Box 190, Kakamega, 50100, Kenya. sbfg01-540862019@student.mmust.ac.ke.
This study used artificial intelligence (AI) tools to analyze breast cancer genetic mutations, identifying key signatures like BRCA1 and TP53 for improved diagnosis and prognosis. These AI platforms offer a novel approach to understanding breast cancer development and potential therapeutic targets.
Area of Science:
- Oncology
- Bioinformatics
- Computational Biology
Background:
- Breast cancer is a leading global cancer in women, with current diagnostic and prognostic methods lacking precision.
- Genetic mutational signatures are implicated in breast cancer progression, yet their role in diagnosis and prognosis is under-researched.
- Existing breast cancer research often overlooks the potential of artificial intelligence (AI) tools for prognostic prediction.
Purpose of the Study:
- To investigate the relationship between breast cancer genetic mutational profiles using AI models.
- To develop accurate prognostic predictions based on identified breast cancer genetic signatures.
- To explore the utility of AI platforms like Cytoscape, Phenolyzer, and Geneshot for breast cancer diagnosis and prognosis.
Main Methods:
- Utilized AI algorithms to simulate human cognitive abilities for analyzing complex biological abnormalities in breast cancer.
- Employed Geneshotsav 2021, Cytoscape 3.9.1, and Phenolyzer to mine breast cancer-associated mutational signatures.
- Correlated imaging phenotypes with genetic mutations, tumor profiles, and hormone receptor status to develop imaging biomarkers.
Main Results:
- Identified specific DNA-maintenance defects, exposures, and cancer genomic signatures linked to breast cancer.
- A PubMed (Geneshot) search yielded 21,921 breast cancer-associated genes, screened by Phenolyzer for mutation propensity.
- AI platforms revealed significant mutational signatures including BRCA1, BRCA2, TP53, CHEK2, PTEN, CDH1, BRIP1, RAD51C, CASP3, CREBBP, and SMAD3.
Conclusions:
- AI tools like Cytoscape, Phenolyzer, and Geneshot demonstrate potential for accurate breast cancer diagnosis and prognostic prediction.
- The identified mutational signatures provide insights into breast cancer development pathways.
- Massive datasets and AI tools can aid in developing novel pharmaceuticals and treatment alternatives.
