Project

Project

monet project

Omics, Machine Learning,
and Precision Medicine:
A New Era for
Glioma Treatment

Gliomas, the most common primary brain tumors, originate from glial cells and are typically associated with poor prognoses. Among them, glioblastoma (grade IV) stands out as the most aggressive form in adults, with a median survival time of only 12 to 15 months. Recent advances in next-generation sequencing and imaging technologies have generated vast datasets, collectively known as omics, encompassing DNA variants, DNA methylation, RNA sequencing, and radiological images. These omics datasets hold great potential for identifying novel gene targets to combat gliomas. Such technological breakthroughs have driven the development of advanced statistical and machine learning methods to translate this complex data into clinically actionable insights. By modeling omic data in the framework of precision medicine, it will be possible to stratify and treat patients based on their unique biological profiles, uncover new biomarkers, and ultimately help redefine novel therapeutic directions.

.cls-1 { fill: url(#Nova_amostra_de_gradiente_2); }
MONET - Multi-omic Networks in Gliomas

The MONET Goals

The MONET project will advance research on the identification and validation of biomarkers and multi-omic networks in gliomas, with the ultimate goal of improving patient diagnosis, prognosis, and therapeutic decisions. This will be achieved through the development of machine learning and bioinformatics tools, focusing on network discovery and analysis, model regularization, and causal inference, key topics at the forefront of machine learning research. 

The MONET Consortium

The MONET consortium brings together four internationally recognized research institutions with expertise spanning information systems, data science, machine learning, and translational and clinical research on brain tumors.

The MONET Data

The MONET project accounts for publicly available databases and datasets from glioma studies for evaluating the proposed methodologies. The predictive performance of the generated models and the resulting glioma molecular signatures will be validated experimentally using relevant in vitro and in vivo models, with a focus on patient-derived cultures and tumors. The scientific advancements achieved through the MONET project are anticipated to enhance our understanding of glioma biology, uncover valuable diagnostic and prognostic markers, and contribute to clinical decision-making and therapeutic research.

The MONET Tasks

01

Data acquisition
and Integration

Collection and integration of omics glioma data from publicly available databases and research studies.

03

Machine
Learning

Identification of biomarkers and causal relations predictive of glioma outcomes using machine learning tasks.

02


Imaging

Extraction of glioma radiomic features from images and association with histologic and genomic outcomes.

04

Biomarker
Validation

Validation of selected candidate biomarkers using complementary in vitro and in vivo models, genetically and pharmacologically modeled to evaluate functional relevance, tumor aggressiveness, and molecular correlates.

01

Data acquisition
and Integration

Collection and integration of multi-omics glioma data from publicly available databases and research studies.

02


Imaging

Identification of biomarkers and causal flows predictive of glioma outcomes using machine learning tasks.

03

Machine
Learning

Extraction of glioma radiomic features from images and association with histologic and genomic outcomes.

04

Biomarker Validation

Validation of selected candidate biomarkers using complementary in vitro and in vivo models, genetically and pharmacologically modeled to evaluate functional relevance, tumor aggressiveness, and molecular correlates.