Are you aware of why the process of creating a new pharmaceutical agent (PA) takes so long? How can this process be accelerated? Alexander Zholobko, a first-year Master’s student in biotechnology, discussed all of this in his presentation at practical classes for the course “Modern Pharmaceutical Technologies”.
Drug development is a sufficiently labor-intensive and financially expensive process. Chemists, biologists, pharmacists, pharmacologists, toxicologists, physicians, and specialists in many other disciplines are involved in it. According to authoritative sources, less than 1% of substances that are potentially endowed with pharmacological properties ultimately become new pharmaceuticals.
Currently, methods for obtaining new active substances, or so-called “formulas,” are known. Among them are: modification of the structure of already known substances; discovery of new properties in already known drugs; synthesis of analogs/antagonists of natural hormones or neurotransmitter substances; randomized screening of new chemical substances. In this regard, it is important to consider a number of factors that significantly affect the development of new drugs. These include both medical-biological factors (changes in the human immune system, emergence of new diseases, resistance of disease-causing pathogens, environmental conditions) and socio-economic factors (losses by pharmaceutical companies due to patent expiration on highly profitable drugs, the need to increase the number of scientific developments and, accordingly, expenses for them).
How do leading pharmaceutical companies solve this problem today? They combine their efforts with specialists in IT, software development, and computer technologies.
Currently, demand for the use of artificial intelligence is growing rapidly worldwide in many areas of human life, and the pharmaceutical industry is not lagging behind this trend.
One of the trends is the developments of companies such as Atomwise, Turbine, and Deep Genomics, which use artificial intelligence to identify the most suitable drugs for certain conditions, doing so in record time while reducing costs. A striking example is that Atomwise provided its algorithm to conduct research on pharmaceutical substances for the purpose of developing an effective treatment method against Ebola virus infections, in partnership with the University of Toronto and IBM.
Through this algorithm, two drugs were identified that can significantly reduce morbidity caused by this infection. This analysis, which normally takes months or even years, was completed in less than one day.
Another well-known trend is the launch by American computer company NVIDIA of the most powerful supercomputer Cambridge-1 in the United Kingdom to help British healthcare researchers solve some of the most pressing problems in healthcare. Along with the launch of Cambridge-1, NVIDIA also announced a series of collaborations with pharmaceutical giants AstraZeneca and GlaxoSmithKline, as well as institutions such as Guy’s and St Thomas’ NHS Foundation Trust, King’s College London, and Oxford Nanopore Technologies.
The supercomputer Cambridge-1 has the potential to significantly accelerate and optimize every stage of drug research.
One area of close collaboration between NVIDIA and AstraZeneca is the creation of a generative artificial intelligence model based on transformers for chemical structures, which will allow researchers to utilize vast datasets (BigData), using self-supervised learning methods, enabling faster discovery of new drug “formulas”.
GSK’s own research focuses on genetically validated targets, which are twice as likely to become approved treatment methods and currently account for over 70% of the company’s drugs. NVIDIA also collaborates with GSK and its artificial intelligence team to obtain vast amounts of genetic and clinical data and help the company develop effective drugs and vaccines faster.
This is how the use of artificial intelligence will enable humanity to cope more effectively with new medical and biological challenges of our time. This is only a fraction of the interesting insights shared by our researcher; the rest will appear in subsequent publications. Follow us—it will be interesting!

