Reconstructed influenza A/H3N2 infection histories using multipanel serology, paper finally out! Original thread here:

@sakitakahashi1 @IsobelRoutledge

One thing I'm particularly proud of is showing that virtually all serodynamics models and data, from the basic serocatalytic model through to complex time-since-infection models, are described by a common data-generating process. This is in the Appendix so please check it out!

Are you interested in analysing serological data for infectious disease epidemiology? Check out our new review article on serodynamics! With Saki Takahashi at JHU and Isobel Routledge at UCSF. (See next comment if you're into serology modeling)

How do repeated #flu infections generate such a complex immune landscape? @jameshay218 @SRileyIDD &co estimate #influenza A/H3N2 lifetime infection histories for 1,130 people, yielding incidence & antibody patterns across age, location & time #PLOSBiology https://t.co/MFqqvkWmDS

Come work with us at Oxford PSI on vaccine trial modelling for pandemic preparedness! Multiple positions available. @PSIOxford @ChristoPhraser

If you are at @options2024 and interested in modelling methods to estimate antibody dynamics and infection histories from multi-antigen serological data, come find me at the poster session this evening or tomorrow! Poster P657.

Postdoc openings in our new PRESTO project: using modelling to explore in advance which types of vaccine trial will work best for future pandemics, with partners @CEPIvaccines. Join us at @PSIOxford Modeller https://t.co/Rtdwkn8bu4 Statistical modeller

It was great to have @adamjkucharski visiting @PSIOxford over the past few days. Brilliant seminar on how we can learn more about pandemics as they unfold using incidental data from NPIs. More seminars to come!

Join us on the 20th of September at 12pm when @WoltersRachael from @OHSUNews will be presenting a talk titled 'Transforming viral defense: The role of Monoclonal antibodies in modern medicine' @HumanGeneticsOx as part of the @PSIOxford satellite seminars https://t.co/oRzvhVJTWX

New in @ScienceMagazine: app-based contact tracing for covid 🤒🤳📲 as well as reducing transmission, generated data for epidemic monitoring & analysis in real-time with unprecedented resolution. Example below: sharp spikes in transmission in England during last Euros, 2021 🇪🇺⚽

Interested in promoting open science, exploring immunity (from SARS-CoV-2 to influenza) and building analysis dashboards? Then you might be a good fit for this new research software engineer post with @cmmid_lshtm. Deadline 15 April 2024:

New preprint: Reconstructed influenza A/H3N2 infection histories reveal variation in incidence and antibody dynamics over the life course https://t.co/w5THpfjLYO 1/12

This has been a massive saga over many years. Although the data are cool enough on their own, the modelling work also tackles many challenges on serodynamics modeling in general. Code/data here: https://t.co/u5Xn88cvYo. Serosolver package: https://t.co/AKDF8MvVP6 12/12

Given the rise in multiplex antibody assays and technologies like PepSeq and PhIP-Seq, modeling methods like these will help us to understand the mechanisms and consequences of how immunity builds over the life course to pathogens like influenza and SARS-CoV-2. 11/12

An exciting output of our inference is the well-known relationship between antibody titer and probability of infection. Using our method, we can understand not just serological patterns, but also immunity patterns using these multi-antigen serology panels 10/12

We find: 1. Serology-based attack rates are high, at around 18% infected per year. 2. Influenza A/H3N2 infection rates are highest in children, decrease with age and plateau in adulthood. 3. Incidence rates are highly correlated at this small spatial scale . 9/12

Fitting serosolver gave us estimates for: 1) each individual’s sequence of lifetime influenza infections; 2) incidence at a fine spatial scale; and 3) parameters of an antibody kinetics model describing boosting, waning, cross-reactivity and measurement error. 8/12

This new paper brings these pieces together: we fit serosolver to our massive dataset of over 70,000 HI titers, summarizing antibody profiles against 20 A/H3N2 strains for 1,130 individuals from Guangzhou, China. 7/12

We then developed this method into an R package, serosolver: https://t.co/ksGejNaWyZ. Around the same time, @BingyiY analysed and summarised the serodynamics of our massive serological study from Guangzhou, China ( https://t.co/YIbmXgvSl6). 6/12