Machine Learning in Protein Science

Aimed at researchers in the molecular life sciences, this unique reference summarizes current approaches for harnessing the power of machine learning for more efficient full quantum mechanical (FQM) calculations in protein systems. Application examples range from property calculations (energy, force field, stability, protein-protein interaction, thermostability, molecular dynamics) to protein structure prediction to protein design and the optimization of enzymatic activity. From a methodological point of view, the practical reference covers the most important machine learning models and algorithms, from deep neural network (DNN) and transfer learning (TL) to hybrid unsupervised and supervised learning.