Abstract
Protein Engineering, 5, 647–657 An error was found (Leberman, 1993) in the selection of non-homologous proteins used to train the machine learning program Golem. The proteins 155C (cytochrome C550) and 2C2C (cytochrome C2 oxidized) have 43.5% sequence homology. Removal of the result for 2C2C (the most accurately predicted of the two proteins) produces a Q3 accuracy of 78% (no change) and a Matthew's correlation of 0.53 (a reduction of 0.04). Removal of 2C2C makes no difference to results for the independent test set of non-homologous proteins. Reference. Leberman, R. (1993) Protein Engng, 6, 547.