ERROR 1
ERROR 1
ERROR 2
ERROR 2
ERROR 2
ERROR 2
ERROR 2
Password and Confirm password must match.
If you have an ACS member number, please enter it here so we can link this account to your membership. (optional)
ERROR 2
ACS values your privacy. By submitting your information, you are gaining access to C&EN and subscribing to our weekly newsletter. We use the information you provide to make your reading experience better, and we will never sell your data to third party members.
COVID-19 and AI
The messenger RNA–based vaccine development efforts to counter COVID-19, while clearly welcome, remain a distant consolation prize for those who had already lost their near and dear, for those still hospitalized, and for those who have developed COVID-19-related longitudinal complications.
The recent advances in artificial intelligence to predict proteins’ structure perhaps are more than a silver lining. AI and existing force fields can be used to predict, design, and generate in silico mutations in proteins and large topologies to create folds that are more robust than their wild-type counterparts.
We can therefore posit that AI will effectively predict the future robust variants of viruses. It can potentially be used to determine which virus is the most effective in jumping host barriers. It is likely to inform us about which cellular architectures are the most receptive to viral invasion.
The computational horsepower and AI–machine learning (AI-ML) techniques at our disposal today to execute such predictive studies easily and by far outstrip the rate at which a virion has historically jumped from its original host to a foreign dwelling and mutated therein.
AI today can potentially predict what the constitution of the Omega variant in 2023 may be (as long as Darwin’s laws of natural selection and survival of the fittest hold). Against these determinations, mRNAs can be rapidly (and economically) created. By preintroduction of these mRNAs into humans, antibodies can be generated against future strains or kept on standby for rapid deployment once the predicted strain was actually realized.
A rate-limiting step in mRNA-based vaccines is the development of effective liposomal coats that facilitate delivery. AI-ML can intervene here as well.
In theory, have we finally won the war? Can AI-ML rescue humankind? Or are there unforeseen problems that have to occur for a new training set of problems and side effects to be generated for the AI-ML machine to chew on and learn? It is difficult to predict!
What are the outcomes associated with having designer antibodies against future viruses circulating in our vasculature today? May AI answer this?
I am grateful to professor Luis Echegoyen (immediate past president of the American Chemical Society) for scintillating and thought-provoking discussions on this topic.
Mahesh Narayan
El Paso, Texas
Join the conversation
Contact the reporter
Submit a Letter to the Editor for publication
Engage with us on Twitter