There are 20,000 closed (historic) landfill sites in the UK that potentially could pose a threat to human health as they are largely unmonitored. Previous work within our Centre has shown inconclusive evidence of adverse birth outcomes with residence close to landfill sites, but understanding population level exposures is hampered by a lack of detailed characterisation of the chemical signatures associated with wind-blown dust, volatile organic carbon emissions and leachates from these sites. To address these questions, we will capitalise on the ‘exposome’ approach using state-of-the-art omics and other analytical technologies to gain a better understanding of exposure, biological dose and adverse biological responses.

This project will utilize biomarker and questionnaire data from the BEED cohort of pregnant women living around municipal waste incinerators across the UK, in order to investigate the potential impacts of the environment on breast milk and subsequent early life development. The project includes epidemiological and metabolomic data analyses but also a wave of data collection from the children of the BEED women.

This project aims at developing statistical approaches to analyse exposures to mixtures of chemicals and their biological responses, ultimately to identify biomarkers of these complex exposures and develop hypotheses on the molecular mechanisms underlying these responses. The project comprises a strong methodological component focusing on the development and the application of novel statistical models and machine learning algorithms accommodating multivariate and interacting compounds. The project will also feature applications to existing and de novo data to identify validated biomarkers of exposure mixtures.

There are approximately 20,000 closed (historic) landfill sites in the UK that potentially pose a threat to human health. This PhD will identify historic landfill sites in the UK, develop an exposure assessment specific to exposure from landfill sites and conduct epidemiological analysis on potential adverse health effects with a focus on reproductive outcomes.

In recent years, concern over the presence of neurotoxins in the environment, whether accidentally or deliberately spread, has grown substantially. These compounds can persist in the environment for long times and endanger the public. Even sampling for these can be extremely risky. Recently, ionic liquids have been shown to be able to reduce the risk of exposure to highly toxic chemicals when analysing, or destroying these. This project will combine theoretical investigations of compounds of concern with experiments on safer stimulants to understand the solubility and reactivity of these neurotoxins with the aim of creating safer analytical and disposal technologies.

Identification and monitoring of potentially hundreds of new highly toxic agents (HTAs) in drinking water remains a significant analytical challenge. This project will characterise and classify new chemical HTAs in UK municipal, well and bottled drinking waters and assess their potential risks for human exposure. Objectives: i) to develop AI-assisted high resolution analytical methods for suspect screening of large numbers of neurotoxins and HTA markers including pharmaceuticals, pesticides, organometallics, plastic-related compounds, disinfectant by products; ii) to conduct a temporal survey of drinking water sources across the UK; iii) analyse urine samples to assess individual-level exposure and analyse wastewater from the corresponding city to extrapolate to both human population and environmental exposure; iv) evaluate public health and environmental risk of HTAs from drinking water sources.

Exposure to metals via inhalation is a relatively new and important area of research given the potential for neurological effects within an ageing population. This project aims to combine recent advances in measurement of airborne metals in different urban settings (indoor, outdoor, transport) with highly detailed population exposure modelling, to provide new insight into individual exposure, representative of the entire London population. We will estimate the important sources of toxic metals, assess policies for exposure reduction and provide data for ongoing health research.

Particulate matter air pollution (PM) detrimentally affects lung development and function. This project will investigate real-world PM samples, including different biodiesel exhaust materials, that drive inappropriate inflammatory responses (e.g. in bronchial epithelial cells, neutrophils, dendritic cells, monocytes/macrophages), to link this with PM compositional data in order to identify relevant signalling pathways and explore mediators that may mitigate detrimental PM effects. Studies in patient cohorts (e.g. asthma, COPD) will test the role of the pathways identified in comparison to matched healthy subjects

Separating short-term health effects of nitrogen dioxide (NO2) and fine particulate matter (PM2.5) is important for policy. The student will study whether (i) the relative risk for NO2 in time-series analysis reduces as the NO2/PM2.5 ratio increases, with particle trap use, (suggesting an apparent effect due to PM2.5 rather than NO2); (ii) controlling the NO2 effect for PM2.5 actually controls more for secondary than primary PM and (iii) causal inference methods provide further insight.

The degradation and abrasion of plastic products and textiles can emit microplastic particles and fibres to the air, potentially resulting in human exposure via inhalation. There is a possibility these cause lung injury following inhalation, but little evidence. This project will investigate the cellular and molecular effects of microplastics on the human airway to determine whether there are physicochemical properties which render them a (low or high) risk.

Co-use of fentanyls with heroin substantially increases risk of overdose and death. Their potential use in a CBRN attack also requires emergency preparedness capacity including detection methods for a suite of fentanyls and urinary metabolites. The establish a mechanism for the monitoring the use of fentanyls in the UK and establish preparedness for monitoring of potential CBRN incident involving fentanyls. Objectives: i) Literature review of the known fentanyl usage trends; ii) charactering fentanyl metabolism using liver microsomes and AI-assisted liquid chromatography-high resolution mass spectrometry (LC-HRMS); iii) Quantitative analytical method development and validation using rapid LC-tandem mass spectrometry; iv) Analysis of human urine samples and a UK-wide wastewater pilot study to estimate national fentanyl usage.

Aim: to quantify and monitor the scale of illicit and abused drug consumption nationally using wastewater-based epidemiology (WBE). We currently provide London WBE data for four drugs to the EU: cocaine, amphetamine, methamphetamine and MDMA. Objectives: (1) to monitor >100 compounds including abused medications, alcohol, nicotine, opioids and novel psychoactive substances; (2) spatiotemporal analysis of UK wastewaters to estimate per capita drug consumption regionally and nationally; (3) to investigate associations between drug consumption and health using hospital admission and mortality data from the NHS Digital Health Episode Statistics and ONS death registrations.

Aim: to better understand exposure and sources of chemical contamination bound to airborne particulate matter (PM). Objectives: (1) to develop comprehensive analytical PM characterisation methods using gas and liquid chromatography coupled to high resolution mass spectrometry; (2) to perform ‘omics-type analysis of current and historical PM filter materials and cohort samples to understand the principal components of chemical exposure, occurrence and potential effects; (3) source apportionment modelling to better understand the health risks of identified chemical PM contaminants.

Aim: to better characterise disinfectant by-product (DBP) exposure from chlorinated, chloraminated, ozonated and chlorine dioxide-treated water sources. The majority of DBPs and their transformation products have yet to be characterised and research has mainly focussed on trihalomethanes and small chain halo-organic acids, both of which are genotoxic and carcinogenic. An assessment of public health exposure and impacts is also still required. Objectives: (1) to characterise more DBP classes using several sources of ozonated and chlorinated drinking water using AI-assisted high resolution mass spectrometry-based chemical analysis; (2) to determine transformation and excretion of DBPs in cohort urine samples and (3) to link identified DBPs and internal molecular signatures of exposure using multi-platform ‘omics technologies.

Aim: To develop methods to measure plastic-derived particulate and chemical pollutants in air and biofluids. Plastic particles (‘microplastics’) contaminate food, drinking water, dust and air. Their presence in human stool, as discovered in a study albeit pilot in nature, highlights the plausibility of human exposure. However, there is an overall lack of data on exposure, an absence of biomonitoring, and no indication of internal measures of effect. This project aims to characterise the plastic and plastic-associated (additives, flame retardants, monomers) chemical exposome of children living in urban areas, based on monitoring within homes and schools and assessment of metabolites in body fluids: blood, saliva and urine.

There is significant and increasing concern over the safety of travel on the London Underground. This studentship offers a programme of environmental and occupational epidemiology, based around the health records of the vary large number of workers on the Underground, to assess the potential adverse health effects of PM2.5 exposure. The programme will consist of two main parts: a retrospective cohort study of sickness absence to understand the short-term, adverse health impacts, and a cohort analysis of mortality to understand the long-term, adverse health impacts. Each will be linked to detailed estimates of workplace particulate exposures. This will be an excellent opportunity for a student who wishes to work on a project of immediate importance. You would work closely with the Occupational Health team at TfL and the Environmental Research and Occupational Medicine Groups at Imperial College.

Particle air pollution is a major health concern in the UK and has been linked to around 29,000 attributable deaths annually. Home burning of solid fuel (wood and coal) is estimated to produce around 39% of the particle pollution (PM2.5) emitted in the UK, greater than that from road transport. Local air pollution hotspots have been identified by the local air quality management process in the UK but this has focused mainly on major roads, enclosed streets and busy road junctions. Localised emissions from solid fuel burning are not well captured in existing emission inventories and current measurement networks. A better understanding of solid fuel exposure hotspots will be necessary to enable a focus on interventions. The PhD project will: • Investigate measurement methods of wood burning PM. • Devise and pilot street-by-street mapping of solid fuel PM concentrations and compare these with datasets from national fixed measurement locations. • Undertake longer term measurements of wood burning PM outside and inside UK homes • Determine the range of PM exposures that might be expected from wood burning in UK urban areas.

There is increasing evidence of life course health effects from air pollution, i.e. that air pollution exposure at particularly points in our life might have a long lasting impact. It is difficult to study such effects prospectively. Looking retrospectively, a study by Hansel et al. (2016) followed a 1% sample of the 1971 English and Welsh census for forty years. They found that mortality rates could be associated not only with contemporary exposure but also with that over the previous three decades. Uniquely the UK has a national measurements of air pollution from around 1920. A recent donation of paper copies of annual data sets to the King’s College London library allows these data to be studied by researchers for the first time. The PhD investigation will: • Investigate the different measurement methods, sample locations and calibrations systems used over the last century to determine time spans when consistent time series can be created and to create conversion factors to smooth inconsistencies. • Create time series of air pollution metrics to span the century for cities of the UK and tie these in with policy interventions and social changes including coal strikes, wartime and Clean Air Act implementation. • Gather long-term health metrics for the UK and investigate associations between changes in air pollution and in health over time. Further investigations will look at between city contrasts in exposure and health outcomes.

Aim: To determine whether long-term exposures to urban air pollution is associated with evidence of accelerated aging and chronic disease risk. Ageing is one of the principal risk factors for non-communicable disease, however the rate of age-related declines in physiologic function and immunity varies considerably between individuals. It has been proposed that DNA methylation age biomarkers may be good predictors of age-related diseases and mortality risk and therefore in this study we wish to examine whether long term exposure to urban air pollution, particularly that derived from traffic emissions promote these age-related changes, after adjustment for other lifestyle factors. In addition, we wish to examine how these epigenetic markers relate to other measures of biological age, including telomere shortening and changes in the metabolome.

General aims of the project include: 1) Investigate socioeconomic determinants of epigenetic aging in children form the HELIX, ALSPAC and Generation 21 cohorts, 2) Develop novel multi-omic biological age (development ) assessment in HELIX and ALSPAC children using epigenetics, transcriptomics, metabolomics and proteomics and developmental endpoints (cognition, growth, adiposity) 3) Investigate role of physiological stress in biological aging, using glucocorticoid steroid profiling in HELIX children.

The environment contains a large number of chemicals whose potential toxicity for humans is unknown. We propose an agnostic approach based on metabolomics. The aim of the fellowship is to apply statistical and laboratory techniques of data mining for the detection of low levels of metabolites and xenobiotics in blood, to fill the knowledge gap in the understanding of the exposome. New analytical methods will consist in adding concentration steps prior to analysis, improving sensitivity and using state-of-the-art equipment. The candidate will then use in depth statistical analysis, multi-OMICs and network analysis to establish links between measured environmental exposures and selected disease outcomes. This will be done in several thousands of individuals over five cohorts (Exposomics project). We will evaluate the specificities of xenobiotic signatures, and the pathways they may affect to understand how exposures influence critical biochemical processes. These results will provide unprecedented information on xenobiotic exposures in humans.

The newly funded PRISM project is a joined UK-Korea initiative funded by the MRC investigating molecular phenotypes that are predictive of the asthma severity and of the responsiveness to targeted biological treatment. Using serial measurements of high resolution OMICs data (RNA-seq, proteomics,…) in a natural experiment design, this PhD project will focus on the development novel statistical models and machine learning algorithms to identify molecular phenotypes, as defined by sets of molecular markers that are jointly predictive of disease severity or of its main risk drivers. Resulting molecular targets will subsequently be investigated, in a longitudinal set-up, in relation the different therapeutic options included in the natural experiment. The research proposed in this PhD project (i) involves the development and implementation (in R, Python, and/or C/C++) of advanced OMICs profiling and integration models as well as machine learning approaches accommodating repeated measurements, (ii) combines both predictive and aetiological research with a particular attention on producing interpretable and reproducible results. As such this research will be conducted in a multi-disciplinary environment involving experts in data analytics, bioinformatics, molecular biology, and clinicians.

LongITools is a newly funded H2020 project investigating the health effects of the exposome in the life course. The project compiles data including (i) multiple high-resolution OMICs profiles (metabolomics, DNA-methylation, gene expression….), (ii) well-characterised environmental exposures, and (iii) multiple cardio-metabolic health outcomes, in several European cohorts totalling thousands of participants. Using a longitudinal design, all measurements are available at different time points in the life course. The present PhD project will aim at developing and applying novel statistical approaches to (i) explore exposome composite scores capturing the joined evolution of the exposures in the lifecourse (via clustering and/or dimensionality reduction techniques), (ii) characterise the internal exposome as defined by the (multi-)OMIC signature of the exposome scores in the life course (using penalised regression, penalised PLS and network models for OMICs profiling and integration), and (iii) assess the marginal and joined effect of the external and internal exposome on the risk of cardio metabolic outcomes (using risk prediction models), (iv) assess the causal nature of the identified association (using Bayesian Path Analysis). This work will rely on the development of bespoke scripts (in R) which will be applied to the wealth of already existing data compiled in the consortium.

Type 1 diabetes is an autoimmune disease affecting ~400,000 people across the UK and is being compounded by an increase in youth and early-onset type 2 diabetes as well. Both type 1 and type 2 diabetes result from gene–environment interactions, with environmental factors likely triggering the disease process in genetically susceptible individuals. This project aims to identify and quantify associations between type 1 diabetes and early onset of type 2 diabetes in children in relation to environmental risk factors. The project will make use of routinely collected health data and broad range of environmental risk factors held within the Small Area Health Statistics Unit.

Cardiometabolic risk factors and their subsequent conditions, including obesity, diabetes and cardiovascular disease, are major causes of multimorbidity in the UK population. The ongoing COVID-19 pandemic is likely to have strong indirect effects on the management, control, and outcomes of these conditions, and conversely, these conditions also represent high risk conditions for COVID19 and other infections. As chronic conditions are consistently associated with social deprivation and their outcomes are affected by access and continuity of care, healthy behaviours and supportive environments, the COVID-19 pandemic in the UK could have enormous variation in the levels of disruption and consequent health outcomes. The overall aim of this project is to quantify the indirect impact of COVID19 on chronic disease management, care, and outcomes and determine the area- and environmental-level risk variation and risk factors for adverse impact of COVID10 on chronic disease care and management. Specific aims in this work will be to: 1) Describe variation and map the levels of care, management, and disease related to cardiometabolic disease; 2) Examine the impact of COVID19 on the short-term and long-term impact of cardiometabolic risk factors and outcomes; 3) Identify and quantify the major environmental and socio-demographic risk factors for adverse impact of COVID on chronic disease management, care, and outcomes.

Determining whether an exposure causes an outcome is a fundamental problem in epidemiology. Mendelian randomization (MR) is a causal inference approach that uses genetic variation as “instrumental variables” to infer if an exposure has a causal effect on an outcome of interest. Existing MR frameworks can only accommodate a few carefully selected variables at a time, which severely limits their application to high-dimensional biomarker sets from molecular high-throughput data, e.g. metabolomics. This project will develop novel statistical approaches combining high-dimensional network inference with an instrumental variable design to infer directions of effect between many exposures and improve our understanding how the exposome causes disease.

Ultrafine particles (UFP) are composite pollution measures (similar to larger size Particulate Matter); in order to understand their health effects, it is necessary to move from total UFPs back to their sources, to design and implement effective policies targeting the reduction of those sources showing evidence of negative health effects. The project looks at developing and applying machine learning tools to answer the question: which sources of UFPs around Gatwick airport are detrimental for the health of the local population? The project will entail statistical methodological developments in two phases: i) the student will work on non-parametric methods such as Dirichlet processes, or other machine learning approaches to apportion the UFPs to their sources; ii) then the student will develop a flexible regression model to link the apportioned sources and the health outcome(s) – both separately from the source apportionment or in a joint fashion, to allow uncertainty to fully propagate across the model components.

With the term big data analytics we refer to the collection, processing, and analysis of large sets of data with the aim of discover patterns and meaningful information. Processing and computing for such large volumes of (multidimensional) data raise, however, new challenging problems, which require scalable algorithms and creative statistical and probabilistic solutions. This project will focus on Earth observations and, in particular, on satellite-derived observations for mapping time-evolving and spatially-varying climate and environmental factors associated with mosquito-borne disease dynamics. Taking advantage of a longstanding relationship with the University of Sao Paulo (Brazil), the student (i) will extend and develop spatio-temporal geostatistics methods stemming from Bayesian inference to retrieve information from remote sensing measurements, then (ii) will investigates the synergistic effects of the local environmental and climate variations on mosquito-borne diseases in Brazil to support outbreak risk management, providing inputs for strengthen early warning systems.

Health surveillance is well established for infectious diseases, but less so for non-communicable diseases. A wide range of spatio-temporal methods are currently available, ranging from traditional tools such as SatScan and Cusum to more recent Bayesian approaches. Choices of methods are often based on easy of use, rather than on best performance and accuracy. The aim of this project would be i) to conduct a review of available methods, ii) to define the appropriate simulation framework to compare the different methods identified; and iii) to create guidance on the strengths and weaknesses of each method considered and on when to use each method.

When assessing the effect of air pollution concentration on health outcomes, it is crucial to understand how the different pollutants acts and interact, however, this is challenging due to their high correlation. This project will compare and contrast different multi-pollutant methods to detect pollutant-specific effect on health. In particular, the student will consider (i) propensity score based methods, as a way of disentangling the effect of each pollutant, adjusting for the other ones and (ii) machine learning approaches, to estimate pollutant profiles. The modelling will be based on a Bayesian framework, to account for uncertainties throughout.

This project will use routinely collected de-identified primary care electronic healthcare record data linked to hospital, mortality and environmental data to explore the relationship between exposures (e.g. temperature, pollution, pollen) on incidence and progression of asthma in a population aged 6 and upwards.