The Chemist’s Gambit | Opinion

Elegant retrosynthesis is an art form, it is strategic: you have to plan ahead, but also be able to adapt if reactions fail. When the lockdown kicked in in 2020, I encountered those same patterns in a completely different area – chess. With everyone sitting at home, online chess suddenly became hugely popular, especially after the series’ release The Queen’s Gambit. My friends and I were among those who gave it a shot. Out of curiosity, I also read something in the history of the game: interesting anecdotes, big names like Paul Morphy or Bobby Fischer and – about 25 years ago – the rise of the machine, when Deep Blue became the first computer to become a reigning world champion. , Garry Kasparov.

While delving deeper into the rabbit hole of chess, I spoke to a colleague about the use of machine learning in chemistry. With Deep Blue in mind, I was a little concerned: Will all my college years be useless if artificial intelligence (AI) takes over research once I finish my studies?

I think this fear of the ‘rise of the machines’ lies with many people. Hollywood has added more than its fair share with villains like Skynet in the terminator. Frankly, with only evil examples like this and my future job at stake, I was also infected by that fear. Therefore, to follow The art of warI decided to get to know my rival better and delve into the subject.

Unlike AI in the classic sense, where the program follows predefined laws, machine learning predicts the results of patterns and parameters it has found in the data it has been given. The more data you give it, the better the result. For example, Deep Blue had a predefined database of chess openings, but also analyzed thousands of chess games itself to perfect the decision-making process.

The machine learning approach is useful in any field that involves a lot of trial and error, such as research and development in the pharmaceutical industry. AI is already being used there to predict the properties of molecules or to design compounds with similar effects to known drugs. The method becomes even more powerful when it is combined with other prediction systems or empirical data in a process called stacking. Even problems that are simply too complex for the human mind, such as protein folding, can be simulated with AI. Machine learning programs have so far been unparalleled in handling the plethora of parameters required for these tasks – and even they are not always 100% accurate.

So, could AI steal my job? Not really. A scientist must be involved to validate AI-generated predictions and synthesize the molecules it suggests. Personally, I’d like to see AI help me find a synthetic strategy for a new compound; it basically works like an automated, super-fast literature search – without having to spend hours clicking the web. It feels like those futuristic spaceship computer assistants from science fiction movies. They tend to get evil, but again, Hollywood is a bad influence.

A worthy comparison, in my opinion, is the implementation of computers in science. Without programs like ChemDraw or Word, I could hardly survive today. I can hardly imagine how time consuming and tedious it must have been to solve multiple pages of math equations by hand. Evaluating crystallographic or NMR data without computers must have been nearly impossible, or at least unimaginably slow. However, we’re not done yet. Computer automation is increasing in all areas: flash chromatography and automated catalytic screenings, just to name a few that use robotic approaches (I’ve never seen a self-aware flash chromatograph trying to take over the world in a movie, by the way). In 20 years, scientists will view AI the same way: a tool that allows us to do our job faster, better and easier than ever before. This is important not only from an economic point of view, but also from an ethical point of view, as it is our duty to society to make progress as quickly as possible. To do this, we need to make use of the best technological innovations available.

After Deep Blue was victorious, chess initially lost some of its magic. The computer’s success provided tangible mathematical proof of which moves were the best, allowing players to abandon analysis and remember only the best strategies. To counter this, more and more players tried unusual positions to push their opponent into unfamiliar territory and then overwhelm them. As a result, matches became much more attractive to watch.

I see a similar trend for chemistry: regular, tedious tasks like column chromatography or solvent screening are automatically performed by computers, and AI can help our decision-making. In return, we, the scientists, can focus on checkingmate new, unknown challenges.

Leave a Comment

Your email address will not be published.