Paper published on Kartograf, a novel atom-mapping framework
Alchemical free energy methods are, in my opinion, one of the coolest ideas in computational chemistry. Basically, you transmute one molecule into another, and that you can do this in a way that gives you scientifically valid information about the differences between them.
However, how you transmute one molecule into another is a non-trivial problem. That’s part of what we tried to address with Kartograf. Our view was that how the atoms were connected to each other (known as the 2D information) needed to be combined with the actual coordinates of the atoms (known as the 3D information). So we developed a 3D-first atom mapping algorithm.
I should be clear that my own contributions to this were not huge. I was helpful in setting up the software framework that was used, and I had many questions and suggestions for how we could compare our mappings to methods that were considered the current state of the art. But the majority of the work, both science and software development, was performed by Ben Ries.
David W.H. Swenson
paper