Starting Year: 2021
Title: MONET: Multi-omic networks in gliomas
Abstract: Gliomas, the most common primary brain tumors, arise from the glial cells and typically exhibit bad prognosis. In particular, glioblastoma (GBM), the most common and aggressive
(grade IV) form of brain tumors in adults, presents median survival times of 12 to 15 months. In addition, the infiltrative nature of GBM hampers surgical cure and eludes current targeted therapies.
The advances in next-generation gene sequencing and imaging technologies are currently delivering large amounts of data, collectively designated as omics, including DNA variants,
DNA methylation, RNA sequencing, and radiological images, holding the promise of disclosing new gene targets to fight gliomas. Such remarkable advances have fostered the
development of statistical and machine learning methods able to translate the complex omic data into meaningful clinical-oriented formats. Modelling omic data from cancer patients
in the context of precision medicine enables categorizing and treating patients according to their biology, while disclosing new potential biomarkers that ultimately will help redefining
therapeutic directions.
The MONET project will advance current research on the identification and validation of glioma networks across multiple omic layers and molecular biomarkers, aiming at improving
patient diagnosis, prognosis, and therapeutic decisions, in a two-fold goal: i) identifying intertumoral multi-omics network-based differences and similarities in different glioma
subtypes; and ii) assessing intratumoral heterogeneity in GBM, the most heterogeneous and malignant type of brain cancer, through the identification of distinct and shared single cell network-based profiles across GBM clones, towards the development of new therapies targeting multiple clones. This will be accomplished through the development of machine
learning and bioinformatic tools encompassing network analysis, model regularization, and causal inference, topics at the forefront of machine learning research.
The MONET consortium is composed of four partners: a) NOVA.id.FCT, represented by the Centre for Mathematics and Applications (CMA) and NOVA Laboratory for Computer
Science and Informatics (NOVA LINCS), b) Universidade do Minho, by the Life and Health Sciences Research Institute (ICVS), a R&D Unit of the Associate Laboratory ICVS/3B’s,
incorporated in the School of Medicine, c) Instituto de Telecomunicações (IT), and d) Instituto de Engenharia de Sistemas e Computadores – Investigação e Desenvolvimento
(INESC-ID). These research institutions have top-ranking international recognition on information systems, data science, machine learning, and translational and clinical research on
brain tumors. CMA is devoted to cutting-edge research in both pure and applied mathematics, encompassing thematic lines in Mathematical Biology, Mathematics for Health, and
Data Science. NOVA LINCS works on key areas of Computer Science and Informatics, doing fundamental and applied research and problem-driven innovation, fostering interactions
with external technology and knowledge users. IT focuses its activity on advancing the state of the art in conceptual, algorithmic, and application directions, with worldwide
recognition on efficient optimization algorithms for inverse problems, particularly in medical images. INESC-ID has a long experience in computational methods for biological data
management, analysis and service-oriented online infrastructures for decision support system implementation and governance. ICVS is strategically located in the Northern region of
Portugal within a fast growing Cluster of Biomedical Science, Technology, and Healthcare institutions, with extensive expertise in brain cancer research, which fosters translational
research with Hospital de Braga and the “2CA” Clinical Academic Centre.
Public databases, such as The Cancer Genome Atlas (TCGA), The Cancer Imaging Archive (TCIA), and the Clinical Proteomic Tumor Analysis Consortium (CPTAC), as well as datasets
available from single-cell glioma studies, will be used for evaluating the methods proposed. Moreover, data from glioma patients and relevant in vitro and in vivo models routinely
used at ICVS, including gene expression, DNA methylation, and therapy response, will complement the multi-omics datasets. The predictive ability of the models generated and the
resulting glioma molecular signatures will be validated experimentally in relevant in vitro and in vivo models, particularly prioritizing the use of patient-derived cultures/tumors. In
addition, advanced visualization tools will be used to complement and strengthen the communication channels between bioinformaticians, computer scientists, molecular biologists
and clinicians. The scientific contributions by the MONET project are expected to improve understanding of the biology of gliomas and disclose diagnostic and prognostic markers,
invaluable contributions to clinical decision and therapy research.
Funding Source: FCT – Fundação para a Ciência e a Tecnologia, I.P.
Typology: R&D Project
Reference: PTDC/CCI-BIO/4180/2020
Principal Investigator (PI): Marta B. Lopes
PI’s institution: NOVA Math
NOVA Math members involved: Marta Lopes, Roberta Coletti, Marcos Raydan, Jorge Orestes