A team of researchers from the University of Muenster in Germany has demonstrated a combination of Monte Carlo Tree Search and deep neural networks based on machine learning and artificial intelligence to make chemical syntheses called retrosyntheses.
According to the researchers, there was a time in 1996 when a computer won a match against the ‘then’ reigning world chess champion Garry Kasparov. After that breakthrough match, the board game Go was long considered to be a bastion reserved for human players due to its complexity. Nowadays, the world’s best player has no longer any chance to win against the software. So, the scientist used the combination to make retrosyntheses. The study has been published in the current issue of the Nature journal.
Retrosynthesis is the method to design the production of chemical compounds. The principal says that going backward mentally, the compound is broken down into ever smaller components until the basic components have been obtained. This analysis provides the cooking recipe, which is then used for working forwards in the laboratory to produce the target molecule, proceeding from the starting materials.
In this, the Monte Carlo Tree Search is a method for calculating moves in a game. At every move, the computer provides a number of variations. For example, how a game of chess might end. The best move is then selected.
In a similar way, the computer now looks for the best possible “moves” for the chemical synthesis. It is also able to learn by using deep neural networks. To this end, the computer brings all the chemical literature ever published, which includes 12 million chemical reactions. The new method is about 30 times faster than conventional programmes for planning syntheses.