by Simon Gao February 9, 2022
There is a significant figure in drug development, average cost for a new drug is about $2.6-billion dollars. Another figure is nine out of ten new drugs entering phase I trials fail subsequent trials and get regulatory approval. The whole industry has been exploring new approaches in drug discovery, AI is widely considered a potential paradigm changer.
Leading biopharmaceutical companies are investing in AI solutions. For example, Pfizer is using IBM Watson, a system that uses machine learning, to power its search for immuno-oncology drugs. Most sizeable biopharma players have similar collaborations or internal programs, and company executives think AI and machine learning will usher in an era of quicker, cheaper and more-effective drug discovery. Undoubtedly skeptics still exist, but most experts do expect these AI tools to become increasingly important.
There are two major approaches in AI based drug discovery:
AI In Designing Drug Molecules
The vast chemical space, comprising >1060 (Decillion) molecules, fosters the development of a large number of drug molecules. However, the lack of advanced technologies limits the drug development process, making it a time-consuming and expensive task, the traditional drug discovery is a trial-and-error process – by analyzing clusters of molecules and looking for a potential drug candidates.
AI can speed up this process, recognize hit and lead compounds, and provide a quicker validation of the drug target and optimization of the drug structure design:
- Design of new drug activities
- Prediction of physicochemical property
- Prediction of new therapeutic use
By leveraging what we learnt from these chemical properties and filtering out a large amount of compound data, AI can research more molecule compounds than human can ever possibly do, and narrow target selections.
AI in identifying biological causes of disease
Researchers have developed a model to identify previously unknown cancer mechanisms using tests on cancerous and healthy human cell samples. They modelled diseased human cells by varying the levels of sugar and oxygen the cells were exposed to, and then tracked their lipid, metabolite, enzyme, and protein profiles. AI is used to generate and analyze immense amounts of biological and outcomes data from patients to highlight key differences between diseased and healthy cells.
The approach is to identify potential treatments on the basis of the precise biological causes of disease. After biological causes are identified, AI can filter out huge amount of data on sources such as research papers, patents, clinical trials, and patient records. This forms a representation, based on real world evidence (RWE) data and research, of more than billions known and inferred relationships between biological entities such as genes, symptoms, diseases, proteins, tissues, species, and candidate drugs. This can be queried rather like a search engine, to produce ‘knowledge graphs’ of, for example, a medical condition and the genes that are associated with it, or the compounds that have been shown to affect it. Most of the data that the AI platform processes are unstructured in nature and are not annotated, so it uses natural-language processing to recognize entities and understand their links to other information points.
This approach of using patient-driven biology and data to derive disease-causing hypotheses, then finding out possible molecular solution candidates through known research and publications, rather than the traditional approach of starting research with chemical compounds, is attracting more attention.
