Because of its length and significant expense, drug discovery is referred to as “bench-to-bedside”. Bringing a medicine to market takes 11 to 16 years and costs between $1 billion and $2 billion. However, artificial intelligence is currently transforming medication research, allowing for speedier and more profitable results.
AI in drug development has altered the method and strategy for biomedical research and innovation. This has helped researchers simplify disease processes and discover biological targets. Let’s take a deeper look at generative AI in drug discovery.
How is it used in drug discovery?
AI has enhanced many phases of the drug discovery process by analyzing massive volumes of data and making sophisticated predictions. Here’s how.
- Target identification
Target identification is the initial step in drug development, and it entails finding potential molecular entities (proteins, enzymes, and receptors found in the body) that might be paired with medications to create therapeutic effects against illnesses.
AI may use vast clinical databases that contain crucial target identification information such as biomedical research, biomolecular information, clinical trial data, protein structures, and so on.
Trained AI models, when combined with biomedical approaches like gene expression, may grasp complicated biological disorders and discover biological targets for therapeutic possibilities. For example, researchers have created a variety of artificial intelligence approaches to uncover novel anticancer targets.
- Target selection
AI aids researchers in identifying prospective targets based on illness correlations and projected treatment value. AI may make this decision not just based on the stated medical literature, but also on wholly novel targets that have no prior references in published patents.
- Drug prioritization
During this phase, AI assesses significant medicinal molecules and prioritizes them for future study and research to help them progress. AI-based approaches outperform earlier ranking algorithms in finding the most potential applicants. For example, researchers created a deep learning-based computer system to prioritize potential Alzheimer’s disease treatments.
- Difficult screening
AI can forecast a compound’s chemical characteristics and biological action, as well as offer information about adverse effects. They can examine data from a range of sources, including prior studies and databases, to determine any possible dangers or adverse effects linked with a certain substance.
- Drug Design De Novo
Manual screening of huge collections of compounds has long been a standard procedure in drug development. Researchers may use AI to screen novel compounds with or without previous knowledge, as well as anticipate the ultimate 3D structure of identified medications.
4 successful examples of AI-powered drug discovery
Abaucin
Antibiotics destroy and kill bacteria. However, due to a shortage of new drugs and the rapid development of bacterial resistance to old drugs, it is becoming increasingly difficult to treat the bacteria. Abaucin, a potent experimental antibiotic created using artificial intelligence, is intended to destroy Acinetobacter baumannii, one of the most deadly superbugs.
VRG50635 from Verge Genomic
Verge Genomic utilized its CONVERGE artificial intelligence technology to identify a novel drug, VRG-50635, for the treatment of ALS by analyzing human data. Data points comprised brain and spinal cord tissue from people suffering from neurodegenerative disorders like Parkinson’s, ALS, and Alzheimer’s.
The platform initially identified the PIKfyve enzyme as a potential ALS target and then recommended VRG50635 as a promising PIKfyve inhibitor that may be used to treat ALS. The procedure took around four years, and the medicine is currently in phase 1 human testing.
Target X by Insilico Medicine
Insilico Medicine uses Generative AI to develop Target X, a medication that is now in phase 1 clinical trials.
Target X is designed to treat idiopathic pulmonary fibrosis, a disorder that can cause lung stiffness in older persons if left untreated. Phase 1 will have 80 people, half of whom will gradually get greater dosages. This will assist in assessing how the medication molecule interacts with the human body.
Exscientia-A2a receptor
AI MedTech produced the first chemical to treat immuno-oncology, a type of cancer treatment in which the body’s immune system fights cancer cells. Their AI medication has entered the human clinical testing phase. Its promise stems from its ability to target the A2a receptor, which enhances anticancer activity while having fewer negative effects on the body and brain.
What does the future hold for artificial intelligence and medication discovery?
Among many other medical uses, AI accelerates and improves the drug development process by evaluating large data sets and identifying prospective targets and drug candidates. Using generative AI, biotech companies may quickly detect patient response signals and generate individualized treatment programs.
