Enzyme modeling as a course of study is in dire need of systematization. Many disparate analyses across the discipline have been proposed, but they need to be integrated into a coherent whole in order to provide predictive capacity.
In order to facilitate this, enzymodeling is being developed in three capacities:
- A repository of features corresponding to the [contributions to catalysis].
- Evaluation of these features in enzyme and non-enzyme protein structures using deep learning.
- An API that provides sensible access to (1) and (2).