Meet our 2020 CDT cohort. Jan DabrowskiPhD project: Biogerontology: understanding the human aging processSupervisors: Tamir Chandra and Catalina Vallejos Based on the data of 235,435 survivors of cancer diagnosed when young, recent studies have shown a substantial 5-fold excess of death during adult life when compared to read more the expected survival from the general population (30% versus 6% death rate respectively). The second most common cause of the excess of number of deaths is caused by circulatory disease. Among studies on cancer diagnosed when young, none has looked at mental health long-term effects of survivors. Two UK-wide research priority setting projects overseen by the James Lind Alliance, defined in the top 10 UK priorities for cancer survivors, their families and their healthcare professionals. Almost half of them concerned the long-term effects of a cancer diagnosis and treatment at young age, the importance of finding a way to compare treatments on the late-effects and developing targeted follow-up care to prevent serious health consequences. The project responds to the increasing demand from government bodies, patients and carers on quantifying the impact of cancer survivorship, exploiting the NHS investment in electronic health data and the UK drive in developing machine learning (ML)-tools for the healthcare. We will analyse big data of cancer registries linked with routine NHS data and public sector data through the National Safe Haven analytic platform for anonymised extracts of data (electronic Data Research and Innovation Service - eDRIS). We will use standard statistical approach and ML tools for better prediction of cancer outcomes. Marcin KedzieraPhD project: Direct and Rapid Antimicrobial Susceptibility Testing of Whole Blood Using Deep LearningSupervisors: Till Bachmann and Kartic Subr For the majority of their lives, human beings are enjoying a period of relative fitness and resistance to illness. However, in the process we call aging, people are increasingly suffering from diseases and their risk of death becomes greater. read more Scientists are considering many explanations for this progressive decline of the human body, a field of study named biogerontology. In this project, we will be considering the yet unexplored avenues in so-called hallmarks of aging. The first subject we will focus on is cellular senescence, a gradual accumulation of damaged and shutted down cells in the human body. Senescence in the context of human aging is a young research area. While first described in, only recently in (Baker et al. 2011), a therapeutic potential of senescent cell removal was proved on mice. However, the underlying biological process is not clear. We will be improving this understanding by determining the content of senescent in different tissues of the human body at different points of time. We will then link that understanding with person's lifestyle factors, genetic variance, diseases, and such, to provide actionable measures to intervene in the process. On the first stage of the project, we will be focusing on cellular senescence. The project will answer questions across a broad range of fields, taking advantage of the skills of both supervisors. First, we will work with scRNAseq data from two sources: the published in house data and third-party scRNAseq data. This will provide a set of phenotypic marker genes - a focus of the Chandra lab. We will then develop a novel tool that can deconvolute bulk RNA-seq, inferring cell-type and phenotypic contributions using aforementioned markers in which Dr Vallejos has significant experience having developed the widely used BASiCS which utilized Bayesian inference to assess cell-to-cell heterogeneity. To bridge the gap between understanding senescence on the individual cell and understanding in the context of the whole organism, we will be taking advantage of big datasets already generated for healthy individuals. GTEx database provides RNAseq data of hundreds of individuals, across 30 primary human tissues.This will ultimately lead to our long term goal of using these proportions to understand the genetic and co-transcriptional effectors in a population study, namely in the GTEx dataset in which the supervisors have significant experience. In addition, our data will, for the first time, provide a comprehensive ageing and senescence signature which will lay the basis for future, more specific senolytic development. In addition, biomarker discovery of senescent cells will lead to early detection avenues for uncleared, pre-malignant, detrimental senescent cells. We believe this project has two components that are eminently publishable - a deconvolution tool and a systems/population study - that appeal to audiences in genetics, computation and systems biology. All findings - both biological and computational - will be provided on free open source services as tools or datasets. We also believe that aside from the senescence focus, this methodology has broad application (ie. deconvolution of premalignancy markers) which could form the basis for future funding applications to sequence scRNA-seq datasets where marker genes could be derived ultimately engaging further collaboration across fields. In addition, information gained under this proposal can be followed up for potential biomarker discovery/validation for early detection of uncleared senescent cells and specific senolytics, targeting specific organs or subpopulation of senescent cells. Bryan LiPersonal webpagePhD project: Foundation model as a digital twin of the mouse visual cortexSupervisors: Arno Onken and Nathalie Rochefort Understanding how the visual system processes information is a fundamental challenge in neuroscience. Recently, predictive models of neural responses to naturally occurring stimuli have shown to be a successful approach toward this goal, read more serving the dual purpose of generating new hypotheses about biological vision and bridging the gap between biological and computer vision. With the advent of large-scale neural recordings and the emergence of visual foundation models, a foundation model of the mouse visual cortex holds tremendous potential. This project aims to design large-scale multi-modal methods that can accurately predict visual responses to natural stimuli across animals. This approach relies on the idea that high-performing predictive models can account for the nonlinear response properties of neural activities thus explaining a large part of the stimulus-driven variability. Moreover, we are interested in interpretable approaches that can illuminate the modulation of neural responses by visual input and behaviour variables, thus providing a platform to investigate the computation in the visual system in silico. Craig NicolsonPhD project: Optimising Organ Donation through the use of Machine Learning to Predict Time to Asystole in Intensive CareSupervisors: Thanasis Tsanas, Nazir Lone, Kathryn Puxty and Martin Shaw Understanding how the visual system processes information is a fundamental challenge in neuroscience. Recently, predictive models of neural responses to naturally occurring stimuli have shown to be a successful approach toward this goal, read more serving the dual purpose of generating new hypotheses about biological vision and bridging the gap between biological and computer vision. With the advent of large-scale neural recordings and the emergence of visual foundation models, a foundation model of the mouse visual cortex holds tremendous potential. This project aims to design large-scale multi-modal methods that can accurately predict visual responses to natural stimuli across animals. This approach relies on the idea that high-performing predictive models can account for the nonlinear response properties of neural activities thus explaining a large part of the stimulus-driven variability. Moreover, we are interested in interpretable approaches that can illuminate the modulation of neural responses by visual input and behaviour variables, thus providing a platform to investigate the computation in the visual system in silico. Matthew WhelanPhD project: Interpretable AI modelling to predict cognitive/mental health outcomes from rest-activity patterns in UK BiobankSupervisors: Daniel Smith, Jacques Fleuriot, Stephen Lawrie and Amy Ferguson Rest-activity patterns, which measure patterns related to sleep and activity levels throughout the day, have shown to associate with a wide range of health outcomes, including overall mortality risk. However, the association between read more rest-activity patterns and certain mental and cognitive health outcomes, such as brain volume and dementia risk, is unclear. This project applies explainable AI modelling methods that aim to predict mental/cognitive health risks from rest-activity patterns using the UK Biobank dataset. Interpretable AI methods provide clearer understanding on the features within a dataset most important for making the predictions, in contrast to many of the black-box AI methods currently in popular use, even if their predictive power is often inferior. Whilst balancing predictive power with interpretability is challenging, interpretability is a critical component if AI modelling approaches are to be adopted within clinical practice. This article was published on 2024-11-22
