“The machine learning method partial order optimum likelihood (POOL) is used to obtain predictions of the effects of missense mutations on biochemical function, based on the effects of the mutation on computed electrostatic and chemical properties of the catalytic residues. Two human enzymes, DNA polymerase kappa (PolK) and ornithine transcarbamoylase (OTC) are presented as test cases. Seventeen cancer-associated variants of PolK were studied. … The ability to predict the effects of missense mutation on catalytic function has important implications for protein engineering and for a variety of biomedical questions.”

Find the paper and full list of authors in Biophysical Journal.