Jasmina Panovska-Griffiths

Background

I am a Senior Research Fellow in Modelling Infectious Diseases within Christophe Fraser’s Pathogen Dynamics Group at The Big Data Institute at University of Oxford and a Lecturer in Probability and Statistics at The Queen’s College, Oxford University. I am also affiliated with The Wolfson Centre of Mathematical Biology at Oxford University.

I am also an honorary Associate Professor at the Institute for Global Health at University College London, where I lead the Modelling for Decision Science Module.

I am a Fellow of The Institute of Mathematics and its Applications,The Royal Statistical Society and The Royal Society for Public Health. I am actively involved in promoting mathematics and statistics as well as other STEM subjects across schools in the UK.

Research

I am an experienced mathematical modeller with extensive training in applied mathematics and statistics who delivers ground-breaking, innovative research in infectious disease modelling that is policy relevant, impactful and has methodological rigour. My undergraduate and masters degrees are in mathematics and my DPhil is in computational mathematics, all from University of Oxford, and my research combines mathematical and statistical methods with data analysis and numerical simulations to answer existing and emerging questions in infectious disease and public health.

I have over nine years experience of modelling infectious diseases. Since the onset of the COVID-19 crisis, I have been actively modelling the transmission of SARS-CoV-2 and its variant with my research advising policy decision bodies in the UK. My modelling findings have been shared with and used by the Scientific Pandemic Influenza Group of Modelling (SPI-M) and the Scientific Advisory Group for Emergencies (SAGE). I have been working with NHS Test and Trace and Public Health England (PHE) to determine aspects to improve service performance. I have also been working closely with researchers at the Joint Biosecurity Centre (JBC) and Public Health England (PHE) within the UK Health Security Agency in answering imminent and emerging questions around controlling COVID-19 resurgence. I actively advise the UK Department of Health and Social Care, the World Bank, the Bill and Melinda Gates Foundation and FIND working across different projects.

COVID-19 research work

Since the onset of the pandemic, I have been leading a portfolio of COVID-19 modelling work, collaborating accross instititions and countries.

My current COVID-19 projects include:

• Covasim model: In early 2020, together with colleagues from The Institute for Disease Modelling, I was involved in developing the Covasim model as an individual-based model for COVID-19 transmission. Methodological description of the model was published at PLoS Comp Biol and an application of Covasim to USA was published at Nature Communications. I have been leading the application of the model in the UK, developing a UK-specific technical framework and applying it to explore different strategies to reopen schools in September 2020 published at the Lancet Child & Adolescent Health and widely quoted in the media. An extension of this work that evaluates the impact of combining face coverings with different test-trace-isolate (TTI) strategies was published at Scientific Reports and work exploring optimal strategies of reopening schools after the third national lockdown.
• SEIR-TTI model: In 2020, I lead work on extended the classic SEIR model to incorporate probabilistic framework to in details model contact tracing. The new SEIR-TTI model allows different TTI strategies to be explored. The methodological paper was published PLoS Comp Biol. Follow-on work that includes vaccination against COVID-19 is in the process of preparation for publication.
• Epi-Economic modelling: The SEIR-TTI model was combined with economic analysis and a user-friendly interface to explore patterns of infections and economic costs of different NPIs to reduce coronavirus transmission. Preprint of this work is available here.
• Rule-Based-Model (RBM): Lead by collaborators at Univerity of Edinburgh, we also developed a rule-based modelling framework that generalises reaction-based models with reagents with internal state that may be bound together to form complexes, as in chemistry. The methodological paper was published at the Journal of Theoretical Biology. The model is currently being applied across a number of Low- and Middle-Income Countries (LMICs) in collaboration with researchers from Imperial College London and the London School of Hygiene and Troppical Medicine. On-going work is underway in applying the model framework to develop within-host immunity model.
• Social distancing and COVID-19 model: Working with NHS North East London Commissioning Alliance, we developed an SEIR model to explore the level of social distancing necessary to safely reopen socialty following the nationwide lockdown. This work was published in Scientific Reports and available here.
• Determining the cohorts at risk of COVID-19: Working with NHS North East London Commissioning Alliance, we use statistical analysis to untangle the key cohorts at risk of hospitalisation, intensive care admission and deaths associated with COVID-19. This work was published in Influenza and Other Respiratory Viruses and available here.
• The impact of RT-PCR vs anti-body tests for COVID-19: Working with collaborators at Queen Mary University London and across different institutions in Austria, I am leading statistical analysis to evaluate the effect of primary care testing for COVID-19 across different epidemic waves. Two papers were published at BMJ Open and at EClinMed and the work is ongoing.
• Exploring the susceptibility to and transmissibility of COVID-19 accross different age groups: I am involved in work led by Prof Russell Viner that aims to untangle the differences in susceptibility to and transmissibility of COVID-19 in children compared to adults. Initial work has recently been published at JAMA Paediatrics and the work is ongoing.
• I have also been involved in modelling and genomic surveillance studies of different variants, collaborating with the COVID-19 Genomics UK Consortium. Collaborative work published in Nature in October 2021, reconstructed the dynamics of different lineages in England between September 2020 and June 2021, suggesting that a number of different variants were circulating in England in early autumn of 2020. But they became dominated by the more transmissible B.1.1.7/Alpha lineage during the second national lockdown between November and December 2020, whuch was then suppressed by the third national lockdown.

COVID-19 other publications

• I have written a number of commissioned pieces for The Conversation and an editorial for BMC Public Health.
• I have had two commissined opinion pieces at the BMJ discussing asymptmatic COVID-19 infections and how we balance protecting lives and livelihoods as lockdown is relaxed.
• I have also been commisioned to write a chapter for the Data Science: Theory and Applications, Volume 44 in the Handbook of Statistics series, that highlights new advances in the field, with this new volume presenting chapters on a variety of interesting topics. My chapter will look into the current state of modelling the COVID-19 pandemic and give examples of new models that my collaborators and I have developed during 2020.

Workshops on modelling and COVID-19

• Together with collaborators from University of Edinburgh and University of Warwick, and supported by the International Centre for Mathematical Sciences (ICMS), in 2020, I organised a Workshop on mathematical modelling and COVID-19 to discuss how modelling informs policy decision making around the UK epidemic.
• During 2020 I was actively participating in Infectious Dynamics of Pandemics workship organised by the Issac Newton Institute. I was part of a working group comprising colleagues across different Universities in the UK and across the World, that have written a number or articles of lessons learned from the pandemic, one published at the Stat Methods Med Res and a number in submission for a special issue of Epidemics.

Other infectious diseases research work

• Modelling influenza transmission. In the period 2015-2018, I lead a portfolio of work funded by the Department of Health and Social Care in the UK, on comparing the impact of dfferent vaccine policies against pandemic influenza. The methodology was published in Methods X while the application to the UK context was published in Vaccine. Additional work on the net-benefit of different vaccine policies on the last four pandemics was published in the Journal of Theoretical Biology. Furthermore, a user-friendly interface was also developed for the purpose of this work using the Shiny app and can be found here.
• Modelling childhood vaccine-preventable diseases. In the period 2018-2019, I lead work funded by the Department of Health and Social Care, and working with Prof Martin Utley at UCL, to develop a novel formal method to evaluate the impact of vaccination changes to one aspect of the programme across an entire vaccine programme. The study was publlished in Vaccine.
• Modelling HIV in UK. During 2018-2020 I co-lead work on applying statistical analysis to evaluate the effect of post-trial nurse-led enhanced primary care testing for HIV. The longitudinal analysis was published in EClinical Medicine. The cross-sectional analysis is currently being prepared for publication.
• Optima Model. Since 2015 I have been part of the Optima Consortium for Decision Science working accross different models to determine how to optimally allocate resources to minimise burden accross a number of diseases and settings. Publications from the group are listed on the Optima website and I have been involved in papers published in AIDS, J Int AIDS Soc and more recently in PloS Comp Biol.
• Modelling HIV across West Africa. I lead a group across UCL, LSHTM and Coventry University that aims to determine the key determinants of risk accross different West Africa settings. We recently developed a novel mathematical model that explores the role of different populations cohorts in gearing up the HIV epidemic which is currently under review.

Data analytics and machine learning research

• Statistical analysis to evaluate primary care interventions. I am part of the PRECODE consortium that uses qualitative and quantitative analysis to understand the impact of COVID-19 on woman experiencing Domestic Violence and Abuse (DVA). Previously, I lead work on evaluating the impact of disruption of primary care service for DVA published at BMC Health Service Research. More recently, the protocol for PRECODE was published at BMC Fam Pract and the first quantitative analysis is under review at BMC Public Health.
• Application of different machine learning and AI tool to neurology and neuro-oncology. I am part of a consortium of statisticians, physicists and neurologists that work on application of different statistical techniques to answer neurology and neuro-oncology questions. In 2020, a paper co-lead by myself applied Random Forest algorithm to classify different gliomas by grade and mutation status and was published at BMC Medical Informatics and Decision making. Recently, work led by my PhD student extended this to apply a Support Vector Machine methodology and was published at Neuroradiology. Ongoing work is looking to extend this analysis on a larger cohort and apply Neural Networks methodology. Additionally, work is ongoing in applying high-order statistics to MRIs to identify superagers as people with superior cognative ability compared to people of their age.

Key collaborating groups

• Christophe Fraser’s Pathogen Dynamics Group, The Big Data Institute, University of Oxford
• IDM COVID-19 response team
• Isaac Newton Institute for Mathematical Sciences
• PTTI modelling group
• Optima Consortium
• NHS North East London Commissioning Alliance

PhD Students (current)

• Secondary supervisor to Loizos Siakallis, at the UCL Institute for Neurology from September 2021. The work will include application of different statistical methods to classifying brain gliomas.

PhD Students (completed)

• Primary supervisor to David Hodgson on his thesis Modelling RSV transmission and vaccination, UCL Department of Mathematics. David successfully obtained his PhD in August 2020. This work has been published at Influenza Other Respir Viruses and BMC Medicine. Follow up work is looking to a build a user-friendly interface for this model and to better understand the burden of RSV during the COVID-19 pandemic.
• Secondary supervisor to Amirah Alsaedi on her thesis The role of Arterial Spin Labelling (ASL) in classification of primary adult gliomas, UCL Institute of Neurology. Amirah submitted her thesis in August 2021. Iinitial work from this thesis has been publsihed at Oncotarget with follow-on owrk on three papers currently in the process of submission.
• Tertiary supervisor to Freya Rhodes. Ongoing work has been published at BMJ open, J Gastroenterol hepatol and BMC Gastroenterology.

MSc Students

• To date I have supervised 47 MSc projects across LSHTM and UCL. Over the period 2015-2021, I have supervised over 30 MSc students across the Department of Mathematics, Institute of Epidemiology & Healthcare and the Institute for Global Health at UCL.