Post-translational modifications (PTMs) on protein contribute to variouse protein isoforms with little evolutionary cost, regulating protein functions in cell signaling events and being involved in many diseases. The increasingly wealth of information on PTMs presents the challenge of understanding the dynamical properties of PTM sites, by which mechanism the allosteric regulation underlying PTMs would extremely enlarge the target space in drug design. Here, we integrate the sequence information, structural topology and in particular dynamics features to characterize the PTMs in the well known targets—kinase dataset. We demonstrate that machine learning can successfully classify the PTM sites and active sites compared with other residues, especially with the dynamics features.
For more details, please see each subfolder.
All the calculations were done with Ubuntu 18.04.4 LST and python 3.7.7.
More details to run deep learning and random forests models can be found at the corresponding folders.
Mode details can be found at DL and RF.
git@github.com:ComputeSuda/PTMKinase.git
For usage of the package and associated manuscript, please cite according to the enclosed.
Sijie Yang
Fei Zhu, et al, Dynamics of Post-Translational Modification Inspires Drug Design in the Kinase Family， J Med Chem，2021, 64, 15111−15125
This repository is distributed under GNU General Public License v3.0.