Statistics show that it usually takes a biopharmaceutical company $500 million to $1 billion and 10 to 15 years to successfully develop a new drug. The high risk, long cycle and high cost of new drug research and development are all big challenges for pharmaceutical companies. AI technology can help analyze the structure-activity analysis of compounds, which can, to some extent, ease the burden of researchers.
Artificial intelligence for structure-activity analysis of compounds
The structure-activity relationship of a drug refers to the relationship between the chemical activity of the drug and the drug effect. The earliest structure-activity relationship studies qualitatively inferred the relationship between the structure and activity of physiologically active substances in an intuitive way, and then inferred the structure of the active site of the target enzyme and designed new active substance structures. With the development of information technology, computer-assisted quantitative structure-activity relationship has become the main direction of structure-activity relationship research, and has become one of the important methods for rational drug design. According to the degree of influence of the chemical structure of the drug on the biological activity, macroscopically, the drugs are divided into non-specific structure drugs and specific structure drugs. The relationship between the biological activity and structure of the former is mainly determined by the specific properties of these drugs. For most drugs, their chemical structure and activity are related to each other. Drugs generally bind to receptors on the body’s cells and then exert their effects.
Many software can simulate the process of structure-activity relationship analysis (SAR analysis) of compounds on a computer, and make predictions about the possible activities of the compounds, and then carry out targeted screening of the compounds most likely to become drugs, which can greatly reduce time for drug mining.
With the help of artificial intelligence, the speed of structure-activity analysis of drugs can be further improved. When there are thousands of compounds that may show a certain effect on a disease, but it is difficult to judge their safety, artificial intelligence can be used to quickly select the safest compound, namely, the best candidate for new drugs. Secondly, for new drugs that have not yet entered the stage of animal experiments and human trials, AI can also be used to test their safety. Because the targeted proteins and receptors of each drug are not specific, if it acts on non-targeted receptors and proteins, it will cause side effects. AI can screen and search the side effects of nearly a thousand known drugs to determine whether they will have side effects, or whether the side effects are large or small, so as to select those with the least harmful side effects. Drugs can then enter animal experiments and human trials, which greatly increases the probability of success and saves time and costs. In addition, the use of AI can also simulate and detect the absorption, distribution, metabolism and excretion of drugs after entering the body, and the relationship between dose-concentration-effects, etc., allowing drug development to be accelerated.
AI & Medicine is a representative start-up company in the field of drug mining and artificial intelligence. It uses supercomputers to analyze existing databases, and uses exclusive algorithms to simulate the process of drug development, analyze the structure-activity relationship of compounds, and evaluate the risks of new drugs in the early stages of development.