We apply cutting-edge data science approaches to two core scientific goals: 1) to develop a better understanding of the brain across the lifespan, in both health and disease; and 2) to predict development, cognition and mental health, at the individual level.
The methods we use include network science, machine learning and Natural Language Processing. Importantly however, our work is scientifically driven- focusing first on the two scientific goals above and then developing methods tailored to our questions of interest. We work with a range of data modalities, including brain MRI, speech, genetics and genomics and behavioural and cognitive data.
Key recent contributions include work relating brain connectivity markers of schizophrenia to transcriptomic data, to shed fresh light on the biological mechanisms underlying the condition; and a new tool for mapping semantic speech graphs which showed speech graphs from first episode psychosis patients were more fragmented than healthy control subjects. The latter was developed as part of a project at The Alan Turing Institute, exploring the potential of speech data to predict outcome for patients with psychotic disorders.
Publications
Below are a few key publications. For a full list of publications, please see here.
- “Semantic speech networks linked to formal thought disorder in early psychosis”, Nettekoven, Diederen, Giles, Duncan, Stenson, Olah, Gibbs-Dean, Collier, Vertes, Spencer, Morgan*, McGuire*, Schizophrenia Bulletin, vol. 49, S142-S152, 2023
- “Cortical patterning of abnormal morphometric similarity in psychosis is associated with brain expression of schizophrenia related genes”, Morgan, Seidlitz, Whitaker, Romero-Garcia, Clifton, Scarpazza, Amelsvoort, Marcelis, van Os, Donohoe, Mothersill, Corvin, Pocklington, Raznahan, McGuire, Vértes*, Bullmore*, PNAS, 116, 9604-9609, 2019
- “Natural Language Processing markers in first episode psychosis and people at clinical high-risk”, Morgan, Diederen, Vertes, Ip, Wang, Thompson, Demjaha, De Micheli, Oliver, Liakata, Fusar-Poli, Spencer*, McGuire*, Translational Psychiatry, 11, 1-9, 2021
- “Functional MRI connectivity accurately distinguishes cases with psychotic disorders from healthy controls, based on cortical features associated with neurodevelopment”, Morgan*, Young*, Patel, Whitaker, Scarpazza, van Amelsvoort, Marcelis, van Os, Donohoe, Mothersill, Corvin, Arango, van den Heuvel, Kahn, McGuire, Brammer*, Bullmore*, Biological Psychiatry CNNI, 6, 1125-1134, 2020
- “Multimodal Graph Coarsening for Interpretable, MRI-Based Brain Graph Neural Network”, Sebenius, Campbell, Morgan, Bullmore, Lio, IEEE 31st International Workshop on Machine Learning for Signal Processing (MLSP), 1-6, 2021
- “Transcriptomic and cellular decoding of regional brain vulnerability to neurodevelopmental disorders”, Seidlitz, Nadig, Liu, Bethlehem, Vértes, Morgan, Vasa, Romero-Garcia, Lalonde, Clasen, Blumenthal, Paquola, Bernhardt, Wagstyl, Polioudakis, de la Torre-Ubieta, Geschwind, Han, Lee, Murphy, Bullmore, Raznahan, Nature Communications, 11, 1-14, 2020
- “A network neuroscience approach to typical and atypical brain development”, Morgan, White, Bullmore and Vértes, Biological Psychiatry CNNI, 3, 754-766, 2018