The research project aims to build a model that can predict immune phenotypes. The data used is gene expression and genetics, and measured immune phenotypes in the 500FG cohort. This data will be splitted randomly in 10 folds and entered the elastic net function ten times with different alpha levels. The model will then be used to predict immune phenotypes in LLDeep with gene expression, genetics, or genetics combined with gene expression data. This will produce immune phenotypes that were not available in this cohort before, and can be used to forecast links between the immune system and microbiome. This will help understand diseases, disease manifestation, or help develop disease treatment which can restore microbiome composition or function.
