I am a molecular biologist and histologist with years of experience in cancer research. My main focus areas have been the tumour microenvironment and drug discovery to target non-cancer cells, especially immune components, involved in the disease. Examples of that is my research on VEGF isoforms and VEGF inhibitors to target angiogenesis, discovery of pathways to target cancer using known TKIs, such as Sunitinib, to block metastasis through targeting bone-marrow stromal cells, or neutrophil protease inhibition to block metastasis spread in the lung.
During the last years, I have also switched my focus to the fields of computational biology, computational analysis of microscopy, and artificial intelligence to analyse multiparametric molecular and pathology data from patient tissue samples. Having experienced by myself multiple times that target discovery (and in some cases, rediscovery) can open real opportunities for the patients, and in today ́s new scenario of multiple data sources and –omics being integrated in a holistic manner, partly with the help of the upcoming AI-based technologies, I envision a new era of target discovery and rediscovery led by computational analysis with machine learning and other AI- based technologies. This is an exciting area of research and development that I feel ready to join and contribute with my dual experience as an oncology experimentalist and computational biologist.
Computation is deeply impacting biomedical research, and one of my aims is to develop tools and graphical user interfaces that are approachable to scientists, that present and analyze data easily while keeping the biological context and meaning (Catena et al, Journal of Pathology 2018, Nature Methods 2017, Genome Biology 2016, Development 2016, BMC Bioinformatics 2015). These tools Include analysis, 3D visualization, biological atlases, and image analysis tools.
At IBM, am working in developing a comprehensive systems approach to better understand pathology images, with the use of advance computer vision and artificial inteligence methods. The ultimate aim is to provide medical practitioners with quantitative, automatically curated data extracted from digital pathology images, to augment and ease their practice.