More than 40% of active start-ups in drug research use artificial intelligence to test through chemical reserves.
In ancient Greece, Hippocrates tried very early to introduce scientific rigor into the “art” of medicine. Over the next thousands of years many intellectual reflections have made medicine the scientific field we know today.
The drug-making process has become a huge field of investigation, characterized by highly complex, time-consuming and costly multi-disciplinary procedures conducted by countless local, national and international organizations.
Drug discovery and design have come a long way since the days of repeated application of known natural toxins against specific diseases (such as those found in fungi or plants) until a therapeutic effect was observed. Currently, due to the recent drop in the cost of gene sequencing, biology is digitizing at a dangerous pace.
Using AI, a team of BenevolentAI researchers identified Barisitinib as a possible treatment for Covid-19 in four days.
However, the large amount of machine-readable data represents both an opportunity to acquire new knowledge and a daunting challenge, as with increasing volumes of data it becomes increasingly difficult to utilize them. It is more complicated to maintain an overview of major developments in adjacent research areas, which can often be useful for drug design. Recent studies have highlighted this point, estimating that about 80% of medical data remains unorganized and unused after it has been created (Kong, 2019).
After enabling significant advances in other markets such as cloud computing and cybersecurity, can AI play a leading role in drug discovery?
Progress thanks to Multidisciplinary Open Archive, HAL
Theoretically, drug discovery and design is exactly the kind of challenge that must lend itself to intelligent automation. For example, the number of possible drug molecule transfers is approximately 1,060, which presents an interesting challenge for AI, which can be trained to identify potential lead compounds and verify target design and drug composition. This work can be both potential and previous.
The power of AI is illustrated by the fact that in just four days, a team of BenevolentAI researchers identified Baricitinib as a possible treatment for COVID-19. Eli Lilly’s drug, commonly used to treat rheumatoid arthritis, can attack both the Covid-19 virus and the body’s inflammatory response. This is the first time AI has discovered an existing drug to detect a new problem.
Analyzing its practical benefits, many companies are taking advantage of intelligent automation. In 2020, for example, Pfizer will not be able to automatically move data through any of its libraries containing data on 4.5 billion commercially available compounds. Today, he can scan the entire database in 48 hours, greatly accelerating his ability to detect potential new drugs.
Currently, AI is not expected to replace human skills. Instead, AI is seen as a way to improve it.
According to Deep Knowledge Analytics 2019, more than 170 AI-driven research and development (R&D) companies and 35 major R&D centers worldwide use AI. A 2019 Deloitte study found that more than 40% of drug discovery start-ups use AI to search chemical stores for potential drugs, 28% use AI to find new drug targets, and 17% use it for computer-assisted molecular design. . Miraj Rahman, a professor of pharmaceutical chemistry at King’s College London, estimates that within ten years, all major pharmaceutical companies will add AI to drug design.
Belief is verified anyway
It is important to note that at present, AI is not expected to replace human skills. Instead, AI is seen as a way to improve it. Subject matter experts are essential for determining data for AI analysis and for equivalent review and verification of results. Furthermore, like any powerful tool, AI can be used for malicious purposes if not checked. In a recent demonstration, an AI model was trained with a set of elementary molecules and tasked with calculating how to adapt to them so that they became increasingly toxic. The result was disturbing: within hours, the model offered more than 40,000 potentially dangerous molecules.
Drug research is just one step in the process of approving larger drugs. Like the discovery, there are inefficiencies in other parts of the process that can be improved. Genetic sequencing continues to improve in speed, accuracy and cost. The Illuminati are influential players in the industry, but newcomers like Oxford Nanopur continue to innovate. Cell manipulation is seeing dramatic progress thanks to companies like Berkeley Lite. In addition, Genmab has made excellent progress in antibody testing. Finally, clinical research firms like Icon allow large pharmaceutical companies to outsource some of the most demanding work and focus on the most complex research. Each part of the value chain improves and contributes to a larger outcome.
Drug discovery remains an important part of the overall treatment development process, and it lends itself to an extended role for potentially trained automation. AI is expected to play an increasingly important role in how we discover new therapies, with the right balance between humans and machines, with the right controls to ensure the work is performed rigorously. The future of new drug research is promising. Hippocrates will be proud of how far he has come.