RINAMI: Residue-attributed interpretable neural network for predicting absolute folding free energy by merging structure and sequence information.

Furthermore, Wilcoxon signed‐rank tests comparing the absolute prediction errors of RINAMI and the baseline model showed highly significant performance differences across the three Mega‐scale test subdatasets, with p ‐values below 0.01 in all splits ( p = 5.95 × 10 −145 , p < 1.00 × 10 −300 , and p = 1.39 × 10 −25 ).

also in 60,174 other papers