Biography
Amith is a London Interdisciplinary Doctoral Programme (LIDo) student at the UCL UK Dementia Research Institute, supervised by Dr Mathieu Bourdenx and Professor Kenneth Harris. His research applies spatial and single-cell transcriptomics to questions of ageing and neurodegeneration, with a particular focus on cell-type-specific vulnerability to DNA damage. Sitting at the interface of biology, maths and computation, he is drawn to graph-based and machine learning approaches as tools for extracting deeper biological insight from complex spatial data.
Amith completed his undergraduate and integrated master's degrees at University College London, studying Natural Sciences with a major in Neuroscience and a minor in Science and Technology Studies. During this time he gained research experience in Caswell Barry's lab, where he worked on analysis of hippocampal calcium imaging data and contributed to a literature project on the subiculum.
He then undertook a two-month research internship at the RIKEN Centre for Brain Science in Tokyo, working under Professor Louis Kang. There he used computational modelling to simulate lesions to grid cell continuous attractor networks, aiming to reproduce the navigational deficits characteristic of early Alzheimer's disease.
Amith joined the LIDo programme, completing two research rotations before starting his current PhD project. His first rotation, with Professor Abhishek Banerjee and Professor Caswell Barry, involved developing reinforcement learning models to characterise human decision-making strategies in cognitive flexibility tasks. His second rotation, with Dr Mathieu Bourdenx and Professor Kenneth Harris, focused on machine learning-based gene imputation methods to computationally enrich spatial transcriptomic datasets. He has since continued with the Bourdenx lab, where his doctoral research investigates the spatial and molecular determinants of selective neuronal vulnerability to DNA damage in ageing and neurodegeneration.
Research interest
Amith's research sits at the intersection of molecular neuroscience and computational biology, using spatial and single-cell transcriptomics to investigate how the ageing brain loses its resilience to cellular stress. His primary focus is understanding why certain cell types are selectively vulnerable to DNA damage, a hallmark of both normal ageing and neurodegenerative disease, and how the spatial organisation of the brain, including the composition and structure of cellular neighbourhoods, shapes these vulnerabilities. To address these questions, he draws on graph-based representations of tissue architecture and machine learning models to extract biologically meaningful patterns from high-dimensional spatial datasets, alongside modelling the trajectories of gene module activity across ageing, with the broader aim of identifying the molecular and spatial determinants that drive selective neuronal vulnerability in ageing and neurodegeneration.