Publication date: Jan 21, 2022
We present the NaviCatGA package, a versatile genetic algorithm capable of optimizing molecular catalyst structures using well-suited fitness functions to achieve a set of targeted properties. The flexibility and generality of this tool are demonstrated with two examples: i) Ligand optimization and exploration for Ni-catalyzed aryl-ether cleavage manipulating SMILES and using a fitness function derived from molecular volcano plots, ii) multiobjective (i.e., activity/selectivity) optimization of bipyridine N.N'-dioxide Lewis basic organocatalysts for the asymmetric propargylation of benzaldehyde from 3D molecular fragments. We show that evolutionary optimization, enabled by NaviCatGA, is an efficient way of accelerating catalyst discovery that bypasses combinatorial scaling issues and incorporates compelling chemical constraints.
No Explore or Discover sections associated with this archive record.
|2.5 KiB||Describes the content of example_1.zip and example_2.zip in more detail.|
|182.1 MiB||Contains structures for the GA runs in Example 1, training data for the ML model and code snippets of the GA setup.|
|1.6 MiB||Contains structures for the GA runs in Example 2, training data for the MLR model and code snippets of the GA setup.|