Illustration of how researchers used their new “SimplifiedBondfinder” algorithm to reveal hidden chemical bonds in proteins. (IMAGE)
Caption
Illustration of how researchers used their new “SimplifiedBondfinder” algorithm to reveal hidden chemical bonds in proteins. It combines machine learning (UMAP*), quantum mechanical modelling, and structural refinement methods. “*Uniform Manifold Approximation and Projection” is a machine learning algorithm for dimensionality reduction. It was used to project high-dimensional chemical descriptor data into a low-dimensional space, enabling the identification of clusters corresponding to known and new bonds.
Credit
Sophia Bazzi (protein structural data from the Protein Data Bank, visualization using Pymol and Coot software)
Usage Restrictions
None
License
Original content