How Artificial Intelligence Is Changing Drug Discovery
The massive advances humans have made in the medical field are prominent. We have developed and created medicine, treatments, and vaccinations that keep human life-expectancy high. But, these advances come at a high price. A massive amount of money is required to conduct researches, trials, and experiments for new medications. However, with the help of technology, scientists have found a solution to this economic problem. In the article “How artificial intelligence is changing drug discovery”, Nic Fleming discusses the discovery and use of artificial intelligence (AI) systems in finding new biopharmaceutical drugs that are possible cures to incurable diseases. Although the article’s main focus is on AI, it is not the innovation of the year.
However, scientists discovered new ways of utilizing AI in the world of biotechnology which can ultimately lead to new valuable drugs. The idea of AI first began in the 1950s and was said to “likely remain in the realms of science fiction” (Fleming 2018). Currently, scientists use ‘narrow AI’—technology that concentrates on a single task—for drug-discovery. In his article, Fleming delves into a particular biotechnology company near Boston, Massachusetts called Berg. Through their specific scientific methodology, Berg was able to create a cancer drug that showed potential healing characteristics. They observed both cancerous and healthy human cell samples. Their AI was built to “identify previously unknown cancer mechanisms using tests on more than 1,000 [cells]” (Fleming 2018). The controlled variable in their experiment was the variation in the levels of sugar and oxygen the diseased human cells were exposed to. They then tracked its lipid, metabolite, enzyme, and protein profiles. The AI totaled massive amounts of data from patients to contrast cancerous and healthy cells. The result: BPM31510, a drug for individuals with advanced pancreatic cancer and is presently undergoing the second phase of a clinical trial.
Apart from Berg, there are also other companies that Flemington shares knowledge about. Benevolentbio, a company based in London, uses AI like a search engine. It gathers data from various sources like research papers, patents, clinical trials, and patient records that make up a large database of “inferred relationships between biological entities”. Flemington brings up how the company uses their AI system to find suggestions that could treat ALS (amyotrophic lateral sclerosis). To make the long story short, in December 2017, four out of five of the 100 existing compounds the AI was able to flag had promising results. Furthermore, in early April, Nature Biotechnology reported that the FDA approved an AI-powered software that sends an instant message to specialists about the nature and treatment of stroke patients. Viz. ai LVO Stroke System speeds up the process of detecting LVO (large vessel occlusion) stroke by 6 minutes.
The importance of artificial intelligence in drug discovery is evidently growing. The fact that AI provides a cheaper, more effective and defined method of researching encourages scientists to develop it further. The more we learn and innovate AI systems like Berg, Benevolentbio, and Viz. ai, the further we unlock answers to the unknown and find cures for the incurable. Artificial intelligence provides a fighting chance for patients with terminal diseases. It is improving the drugs we use for cancer patients and our understanding of certain cancerous diseases. In the article “AI-powered drug discovery captures pharma interest”, Eric Smalley states that the “two-decade-long downward trend in clinical success rates has only recently improved” due to AI. This shows that AI shows an abundant success rate. Another point worth noting is that some scientists believe that learning how to navigate technology is vital. They urge students that aim to work in drug disovery to “become more informed and flexible” (Flemington 2018). The article shows a statement made by Anthony Bradley, a medical chemist at the University of Oxford, UK, which points out that “remaining versatile [makes] best use of the power of the available tools”.
To sum up, this article dissertates the role AI plays in the pharmacology department and specific areas in the medical field. Nic Fleming eloquently explains in simple, understandable terms how AI provides a more effective solution and a more defined method of researching. He expands on a few companies that use AI in their work and proves the potential impacts they have on the subjects they are working on. Yet, with all these great examples, Fleming ends with the note on the matter that there are currently no approved drugs developed by AI. There are also people concerned with the ethical controversies surrounding artificial intelligence. But, personally, I am hopeful that it is the key to finding the solutions to the existing medical uncertainties we have in our world. Technology is only continuing to prove that with the right innovations and proper applications, it can save lives.
