Maple

1. Maple improves pre-disease status detection accuracy in silico (mean AUC 0.85).

   “0.85 for pre-disease status detection”

   In silicopre-disease status detection

   A robust computational framework for methylation age and disease-risk prediction based on pairwise learning

   Nature computational science·Jan 13, 2026

More on Pre-disease status detection accuracy

1. Maple improves disease identification accuracy in silico (mean AUC 0.97).

   “mean areas under the curve of 0.97 for disease identification”

   In silicodisease identification

   A robust computational framework for methylation age and disease-risk prediction based on pairwise learning

   Nature computational science·Jan 13, 2026

More on Disease identification accuracy

1. Maple decreases methylation age prediction error in silico (median absolute error of 1.6 years).

   “achieving a median absolute error of 1.6 years across 31 benchmark tests”

   In silicoacross 31 benchmark tests

   A robust computational framework for methylation age and disease-risk prediction based on pairwise learning

   Nature computational science·Jan 13, 2026

More on Methylation age prediction error