Daniel Guerrero Romero

Biography

My expertise lies at the intersection of mathematical oncology, computational biology, and precision medicine. I develop mathematical models and statistical frameworks that integrate multi-omics datasets—including genomics, transcriptomics, and drug-response profiles—to dissect tumour evolution and unravel mechanisms of treatment resistance. A central component of my work is the analysis of patient-derived tumour xenografts (PDTXs) and ex vivo systems to characterize persister cell states and identify vulnerabilities that can be exploited therapeutically. I began my academic journey in pure mathematics at the National Polytechnic Institute in Mexico City, where I cultivated a passion for algebraic topology. During my MSc, I transitioned into applied mathematics with a focus on stochastic processes in game theory. I completed my PhD at the CRUK Cambridge Institute under Professor Carlos Caldas and Dr. Oscar Rueda, designing high-throughput computational pipelines and formulating models of clonal evolution in breast cancer. Currently, I am a Postdoctoral Fellow in the EAZPOD programme (EMBL-EBI and AstraZeneca DISC), where I extend my research into multi-modal characterizations of persister cell populations to inform personalized therapeutic strategies.

Research Interests

I’m driven to understand why certain cancers evade therapy and to predict, in advance, which treatments will succeed for individual patients. By fusing mechanistic mathematical modelling with rich multi-omics data—spanning genomic mutations, transcriptomic states, and quantitative drug-response metrics—I aim to chart the evolutionary trajectories through which tumour cells acquire resistance, with particular attention to the role of persister cells in driving relapse. My goal is to bridge the translational divide between promising laboratory discoveries and clinical application by building predictive models that leverage patient-derived tumour models (PDTXs), ex vivo cultures, and in vitro assays. These models illuminate how tumour heterogeneity and clonal dynamics shape differential drug sensitivity across breast cancer subtypes. To handle the complexity of longitudinal and high-dimensional datasets, I develop computational pipelines that automate data preprocessing, feature extraction, and biomarker selection. I am especially interested in integrating machine learning techniques—such as Bayesian inference and reinforcement learning—with mechanistic frameworks to refine treatment recommendations. Ultimately, I seek to create robust mathematical tools that not only explain resistance mechanisms but also guide the optimal sequencing and combination of therapies to improve patient outcomes.

Professional Career

Education

Contact Information