Currently, I am a Ph.D. candidate in Probabilistic Machine Learning applied to Personalized Medicine and Genetics at Universidad Carlos III de Madrid (UC3M) under the supervision of Dr. Pablo M. Olmos, and a predoctoral researcher at the Instituto de Investigación Sanitaria Gregorio Marañón (IiSGM) under the supervision of Dr. Carolina Martínez-Laperche. My research is mainly focused on the development and application of Probabilistic Machine Learning models for Personalized Medicine problems such as biomarker discovery for post-transplant complications after stem-cell transplantations, biomarker discovery for Chimeric Antigen Receptor (CAR) T-cell therapy complications, risk stratification in Acute Myeloid Leukemia (AML) patients or sleep characterization based on smartphone passively sensed data. Throughout 2023, I collaborated with Professor Anders Krogh’s team at the Center for Health Data Science (HEADS) on the development of representation learning and generative models for studying genetic data related to cancer.

Current Projects

• Improving Inference in Discrete Variational Autoencoders using Error Correcting Codes. In this project, we conceptualize the generative model as a communication channel and demonstrate that introducing ECCs to safeguard the latent space can lead to a tighter posterior approximation. Preprint available under request. Supervisor: Dr. Pablo M. Olmos.

• Generative models applied to Cardiac Electrophysiology. In this project, we employ generative models to analyze cardiac electrophysiological maps associated with atrial fibrillation patients. Our goals include enhancing resolution from low-resolution maps, filling in missing or sparsely populated areas, generating synthetic data, and exploring the latent structure of the data, potentially linked to the physiological traits of the patients. Supervisors: Dr. Gonzalo R. Ríos Muñoz and Dr. Pablo M. Olmos.

• Identification of Polymorphisms in Immune System Genes Associated with Early Progression after CAR T-Cell Therapy. In this study, we aim to identify genetic markers that could serve as potential predictors of early progression after CAR T-cell therapy. Identifying these markers could provide valuable insights for further biological investigations into the underlying mechanisms of the disease. Supervisors: Dr. Pablo M. Olmos and Dr. Carolina Martínez-Laperche.

• Unveiling the Latent Structure of Cancer using Generative Models. In this work, we aim to apply the Deep Generative Decoder to obtain an informative and interpretable latent space that capture the underlying structure of cancer genomics. Supervisors: Prof. Anders Krogh and Dr. Pablo M. Olmos.

News!

• September 2023: Happy to announce I’m starting a research stay at the Center for Health Data Science (HEADS) at the University of Copenhagen, under the supervision Professor Anders Krogh.
• May 2023: NEW PUBLICATION ACCEPTED! Our paper ‘Handling ill-conditioned omics data with deep probabilistic models’ has been accepted for publication in the ‘IEEE Journal of Biomedical and Health Informatics’.
• April 2023: NEW WORK ACCEPTED! Our work ‘Identification of Biomarkers and Risk Factors for Inmune Effector Cell-Associated Neurotoxicity Syndrome (ICANS) in CD19-directed CAR T-Cell Therapy: a Retrospective Machine Learning-based Analysis’ has been accepted at the EHA2023!
• April 2023: Happy to announce I’ve been awarded a full scholarship to attend to the Nordic Probabilistic AI School 2023 in Trondheim (Norway)!
• March 2023: I’ll be volunteering at AISTATS 2023!
• January 2023: NEW PREPRINT AVAILABLE! Our work ‘Sleep Activity Recognition and Characterization from Multi-Source Passively Sensed Data’ is already available at arXiv and is currently under review.
• December 2022: Reviewing for AISTATS 2023!
• December 2022: NEW PREPRINT AVAILABLE! Our work ‘Handling ill-conditioned omics data with deep probabilistic models’ is already available at bioRvix and is currently under review.
• November 2022: NEW PUBLICATION! Our work ‘Identification of Predictive Models Including Polymorphisms in Cytokines Genes Associated with Post-Transplant Complications after Identical HLA-Allogeneic Stem Cell Transplantation’ has been published in Blood as part of the 64th Annual Meeting and Exposition of the American Society of Hematology.
• October 2022: NEW WORK ACCEPTED! Our work ‘Sleep Activity Recognition from Multi-Source Passively Sensed Data’ has been accepted at the Machine Learning for Health (ML4H) symposium!
• September 2022: I’m attending to the PyConES at Universidad de Granada!
• August 2022: NEW WORK ACCEPTED! Our work ‘Caracterización de pacientes con leucemia mieloide aguda mediante el uso de un modelo de aprendizaje automático. Caso práctico en un centro de tercer nivel.’ has been accepted at the LXIV Congreso Nacional de la SEHH, XXXVIII Congreso Nacional de la SETH Y 38º Congreso Mundial de la International Society of Hematology
• July 2022: I’m attending to the ELLIS Machine Learning Summer School at Cambridge University!!
• August 2021: NEW WORK ACCEPTED! Our work ‘Identificación de modelos de riesgo que incluyen nuevos polimorfismos en genes del sistema inmune y su relación con complicaciones post-trasplante en pacientes sometidos a un trasplante HLA-idéntico’ at the LXIII Congreso Nacional de la SEHH y XXXVI Congreso Nacional de la SETH

Contact

If you want to contact me, please send me an email to mmargarc@pa.uc3m.es .