Figure 2 | Scientific Reports

Figure 2

From: Predicting type 2 diabetes via machine learning integration of multiple omics from human pancreatic islets

Figure 2

A schematic presentation of our integrative multiOmics analysis. The full dataset is randomly split into train (n = 88, 80%) and test (n = 22, 20%) datasets. Feature pre-selection and multiOmics integration are performed on the train set, and the model is evaluated on the test set. This procedure was repeated 100 times, and confidence intervals of the predictive model were built via the hold-out cross-validation strategy, i.e. multiple splitting of the data into train and test datasets.

Back to article page