SISTM

The relevant information is extracted from large omics data, and this signal is then incorporated into mechanistic models, thanks to prior biological knowledge, to estimate their parameters. Those models can then inform the optimal vaccine strategies to be evaluated in the next clinical trials. Such in silico trials, allowing for optimized clinical trial designs and personalized strategies, in turn generate new data that require and feed anew the above methodological developments.

The team is structured around three research axes focused on these shared objectives.

The axis on “High-dimensional statistical learning” aims to

• Unlock the analysis of high-dimensional longitudinal data by developing suitable statistical approaches, in particular for applications to longitudinal high-throughput data (e.g. microbiome, transcriptome, cytomics) generated in vaccine trials.
• Leverage prior biological knowledge and formally incorporate it into statistical models to tackle the small n large p setting, one of the characteristics of early phase vaccine trials.
• Advance adaptive clustering methods of high-dimensional data in both supervised and unsupervised settings, especially to infer the proportions of cellular population from gene expression measurements and also to identify gene whose expression is key in segmenting transcriptomic measurements across vaccine arms or disease severity for instance.
• Perform feature selection and dimension reduction of high-dimensional molecular and cellular data based on surrogacy potential and prior biological knowledge, as a first step to feed such information into mechanistic models.

The axis on “Mechanistic modelling” aims to

• Develop methods for statistical inference of differential equations model parameters in population framework.
• Use Artificial intelligence for hybrid modelling and infer unknown mechanisms
• Model within-host immunological and virological dynamics in samples of individuals.
• Model between-host dynamics of epidemics in populations.
• Use mechanistic model as in silico platform for exploration of counterfactual scenarios with application in implementing control strategies toward personalized medicine.

The axis on “Translational vaccinology and design” aims to

• Facilitate secure data sharing, integration, and analysis in collaborative clinical vaccine research projects.
• Accelerate the vaccine development by in depth analysis of data generated in early clinical trials
• Design the next trials with development of new adaptative designs in collaboration with immunologists and clinicians

All this work is done in collaboration with our partners from the Vaccine Research Institute, EUCLID/ANRS-MIE CMG platfor, the UMS MART and the Bordeaux Hospital.

The research conducted by the SISTM team is funded through a combination of national, European, and international programs, as well as public-private partnerships. SISTM is involved in two axis of the PEPR Santé Numérique programme (SMATCH ; AI4scMed). The team also secures major European funding as partners through programs such as Horizon Europe (SOLVE ; IP-CURE-B), IMI2 (CARE), and EDCTP2 (PREVAC-UP, ASCENT), which support statistical modelling and data analysis for vaccine research. International collaborations, including with CEPI (MUSICC), UT Austin (RISE), Rand corporation (DESTRIER), and Latin American partners (MATH AmSud), further enhance methodological innovation. The team also engages with industry partners like Johnson & Johnson, Gilead, and Ipsen for clinical trial design and analysis.