Heterogeneidad intratumoral en cáncer de mama

LUIS DIAMBRA; DANIELA SENRA; NARA GUISONI

Buenos Aires

Taller; XXII Giambiagi Winter School: Artificial intelligence and deep learning in physics; 2020

DF, UBA

Breast cancer is the most prevalent cancer form and the leading cause of cancer deathsamong women worldwide according to the World Health Organization (WHO). Breastcancer is a heterogeneous disease and displays variability between patients (intertumoralheterogeneity) and within individuals (intratumoral heterogeneity).Historically, attention has been focused on studying heterogeneity between patients, interms of histopathology, DNA tests (microarrays), clinical outcomes, among others.The gene expression profile of breast cancer is generally performed in the total biopsy(bulk RNA-seq data), which can oversimplify the complexity of the disease. With theadvent of new technologies, the assessment of intratumoral heterogeneity is increasing. Inparticular, with single-cell RNA sequencing (scRNA-seq) it is possible to observe geneexpression at the cell level. The identification and characterization of cell subtypes withina tumor is crucial for the diagnosis and development of targeted therapies.Single-cell RNA sequencing has great potential, however it also presents new challenges.From the computational and statistical point of view the challenge lies in the treatment ofvast amounts of sparse data.Here, we use single-cell RNA-seq raw data available from international publicrepositories and preprocess them, the downstream analysis consists of reducing thedimensionality of the dataset and performing graph based clustering. We identify geneticmarkers for different types of breast cancer cells within individual tumors and analyze thepopulation composition in each patient.