Figure 1.

Flowchart of predictive modeling using neuroimmune guidance cues (NGCs). Subjects in the CIRCULATING CELLS cohort were recruited based on inclusion criteria. The clinical characteristics were collected, peripheral blooded CD14-positive monocytes were isolated and their transcriptomes were profiled. The expression of NGCs was subset. The individuals were randomly assigned to the Training Set or Test Set for external assessment of the model performance. Classification models that were built on the Training Set data and model performance were internally assessed with cross-validation. Finally, we made comparisons between the models and gained insights on the choice of model and features.