Computational Chemistry

TL;DR -- Machine learning and its use of physics was the focus for the Nobel award. This was followed up with the chemistry award being given to protein research. This is a step forward for various reasons. Ultimate consequences are not really known at this time. Research deals with the future.  

--

Yesterday, we noted that the Nobel Prize for Physics in 2024 was given for work in machine learning. ML which is usually known as AI (artificial intelligence) deals with the technology of a computer systems taking as input huge streams of data and reducing these through various techniques into information that people can understand and use. But, this information can feed back into the computer model itself. 

The emphasis is on modeling which is accomplished through computational techniques. "computational" goes along with "theoretic" which deals with analytic and quantitative means applied to a field of study in order to effect desirable effects such as design, prediction and continuing analysis, generally in support of scientific endeavors. If one surveys the STEM discplines, one sees a huge influence of computational modes. With respect to some of the harder problems such as life and intelligence, one could very well suggest that theoretical chemistry will play a huge role. So, this is a necessary step; but, we have huge problems yet to resolve. 

Along with that of Physics, the Prize for Chemistry in 2024 was given to researchers who used a variant of ML from Google (AlpaFold). This work portends to future benefits that have been elusive using other means. Needless to say, the success is heavily dependent upon the researchers who are working now as well as the continually developing frameworks of their disciplines. 

A technical paper explains some of the details: COMPUTATIONAL PROTEIN DESIGN AND PROTEIN STRUCTURE PREDICTION. The following is a quote from the paper:

In summary, the achievements of David Baker, Demis Hassabis and John Jumper in the

fields of computational protein design and protein structure prediction are truly profound. Their

work has opened up a new era of biochemical and biological research, where we can now predict

and design protein structures in ways that had not been possible before. Hence, a long-standing

goal has finally been met, and the impact of this will have far-reaching consequences. 

As it says, this is research and applies to the future. We will look at some recent applications of the techniques and discuss the matters that relate. 

The use of AI is somewhat unfortunate. The work is an example of applied mathematics, computational modeling of such, people who are in the position to exploit the facilities in their work, and long years of work captured various ways as science has been doing for the past 200 years or so. 

AI deals with intelligence which we know is mostly associated with living forms. Now, we can look more closely at modeling these by using improved protein analysis plus a whole lot of other information. Like with physics which is strongly using normals in order to reduce problems to a proper scope, we will see this with other scientific domains. 

At we see with GenAI, what are the costs of this research in comparison with other work that is necessary, in terms of the resources used (say power), or the potential for misuse given our proclivity to not know how to manage complexity? 

Congratulations to the Laureates. Looking forward to watching things unfold. 

Remarks: Modified: 10/09/2024

10/09/2024 --