The CMI-Flu prediction challenge series follows on from the CMI-PB prediction challenges. Computational models capable of forecasting an individual’s immune response to the influenza vaccine can provide mechanistic insights and pave the way for personalized vaccination strategies. In this challenge we provide influenza vaccination data collected pre- and post-vaccination from a cohort of donors. The data is multi-modal, consisting of cytokine measures, transcriptomics, genetic information, and cell frequencies and phenotypes. Participants are asked to train a model on the provided training data, and to predict a post-vaccination response on an unseen challenge dataset.

The prediction challenge series is hosted on Kaggle. Contestants are asked to predict a number of different tasks (each task has its own dedicated Kaggle page).

• Tasks 4.1 - 4.10 : here you can submit ALL tasks simultaneously. The metric is calculated as a mean across all (shortlisted tasks). If you would like to submit predictions for non-shortlisted tasks (to be scored at a later date), this is the place!
• Task 4.4 : predict magnitude of antibody response
• Task 4.5 : predict breadth of antibody response