I started my journey in science with a Master’s in Medical Biotechnology, where I first discovered how data could reveal patterns hidden in biology. That curiosity soon became a commitment, I wanted to understand not just the biology itself, but how algorithms could help us ask better questions about it. Through my academic journey, I found myself drawn to lab techniques, data and genomics. I spent years working with RNA-seq and NGS data, learning how to handle their complexity and, more importantly, how to translate numbers into stories about cells, tissues, and diseases. Those experiences shaped my perspective: data isn’t just statistics, it’s another lens through which we can study life. I like to bridge machine learning and biology to study them in ways that weren’t possible before. It feels like the natural next step in my journey bringing together my background in medical biotechnology, my training in health data science, and my passion for using AI to illuminate biology’s more hidden corners. I’m excited by the unknowns, by the chance to build tools that might change how we look at something as small and essential as cell. For me, research is less about chasing buzzwords and more about following that spark of curiosity wherever it leads.

Predicting variant effects within ciliary proteins

DC6 will be dedicated to predicting variant effects within ciliary proteins. (1) As the group has done for other diseases and protein classes, we will define features that are most predictive of ciliary pathogenesis including many features used by the group previously and details on gene expression and the interactome. These features will then be used to predict variant details using ML as we have done previously. Beyond just whether variants are “pathogenic” or not the focus will be on specific protein (e.g. stability, interactions, molecular mechanism etc.) or phenotypic outcomes, particularly focusing on tissue relevance and overall disease severity as is of particular interest in the ciliopathies. (2) DC6 will also interrogate and attempt to predict sets of candidate modifiers and other potential variant types of interest. We envision that the fellow will also work very closely with experimental partners in the consortium, particularly those generating omics data (e.g. P6-UT and P1-RUMC) and those with a focus on patient variant data (e.g. P5-Imagine and AP4-UEDIN).  It is with these partners that we plan secondments. We also expect numerous collaborations with other partners on particular variants, genes or datasets of interest.