We often imagine futuristic AI computers, 3D-printed organs, and robot surgeons when we think of new medical advances. However, the more ambitious and less-explored approaches to drug discovery and development that are currently being used could prove to be just as interesting.

According to a recent global data survey, over 70% of pharma sector respondents believe that smart technology deployment will have the greatest influence on drug development. Here's a look at some of the technology advancements and approaches that could alter drug research in 2022 as the year draws to a close.

Using supercomputing to harness AI

In terms of speed and performance, supercomputers are far superior to general-purpose computers, and they are especially useful for scientific and data-intensive jobs. It's no surprise that researchers are attempting to use supercomputing to speed up the time-consuming process of medication discovery and development

Gene Writing

Gene editing, which involves inserting, deleting, modifying, or replacing DNA in a genome, is a promising and relatively new method of treating genetic problems. The concept of gene writing to combat inherited disease is even more recent.

Quantum Computing

While using computational approaches to identify drugs isn't new, using ultra-efficient quantum computers to uncover previously unknown molecules has only lately surfaced as a promising topic.

Top three predictions for pharma supercomputing:

  • Speedy drug discovery: Molecular simulations assist to model target and drug interactions completely in silico, which speeds up drug discovery by a million times. AlphaFold and RoseTTAFold's innovations, which resulted in a thousand-fold increase in known protein structures, as well with AI's ability to synthesise a thousand more new chemical compounds, have expanded the opportunity to identify medications by a million times.
  • Multimodal AI: There are almost ten thousand diseases for which there is no treatment. Whether it's to find new treatments or treat patients, multiple sources of health data must be utilised. Multimodal AI will take us to a new level in finding disease pathways and personalising patient treatment and prognosis by leveraging the world's greatest data sources.
  • Federated learning: To help application developers industrialise their AI technology and broaden the application's economic impact, AI must be trained and validated on data that sits beyond their group, institution, and location. Federated learning is essential for enabling such collaboration in order to construct and verify robust AI models without having to share sensitive data. Federated learning will be a critical capacity for AI learning and evaluation in the future.