|In recent years, the healthcare industry has been adopting new technologies such as 3D-printed medication, smart pill technology, and mobile apps to improve its services / Photo by: Max Pixel|
In recent years, the healthcare industry has been adopting new technologies such as 3D-printed medication, smart pill technology, and mobile apps to improve its services. One of the main aspects that the industry is focusing on is drug discovery, which takes long years to implement. Reports showed that 95% of the experimental medicines that are studied in humans fail to be both effective and safe despite undergoing years of development and assessments.
A 2013 study by Forbes, a global media company focusing on business, investing, technology, entrepreneurship, leadership, and lifestyle, revealed that a company hoping to get a single drug to market can potentially spend $350 million before the medicine becomes available for sale. Other large pharmaceutical companies working on several drug projects at once usually spend more than $5 billion per new medicine. Thus, the number of drugs invented per billion dollars of R&D investment has been cut in half every nine years for half a century now, reported Nature Reviews Drug Discovery, a peer-reviewed review journal covering drug discovery and development.
Francis Collins, the director of the National Institutes of Health, decided to launch a new National Center for Advancing Translational Sciences, which aims to remove the roadblocks that keep new drugs from reaching patients. “We would love to contribute to making that failure rate lower, to identifying those bottlenecks and to trying to reengineer the pipeline so if failures happen, they happen very early and not in later stages where the costs are higher,” Collins said.
This is where artificial intelligence comes in. For years, AI has brought a new era of automation to the healthcare industry. Reports showed that investments in AI healthcare tech are increasing, which is expected to reach a value of $36.1 billion by 2025. This is higher compared to only $2.1 billion in 2018. Thus, it’s not surprising to see that AI has great potential in drug research and discovery.
The Potential of AI in Drug Discovery
According to Stat, an American health-oriented news website, AI can make information from the scientific literature and databases useful. This can help in identifying potential approaches to treat diseases by proposing a drug target, designing a molecule, and defining patients. It can also help in determining relevant information faster and making links between biomedical entities.
As of now, many leading biopharmaceutical companies are using AI in drug research and discovery. For instance, Roche subsidiary Genentech is utilizing an AI system to search for cancer treatments; UK start-up Exscientia’s AI platform is looking for metabolic-disease therapies, and Pfizer IBM Watson, a system that uses machine learning, is searching for immuno-oncology drugs.
While many companies are skeptical, AI and machine learning are considered great tools for quicker, cheaper, and more effective drug discovery. In fact, some companies have already succeeded. According to Nature, the world's leading multidisciplinary science journal, researchers at biotechnology company Berg have developed a model to identify previously unknown cancer mechanisms. The researchers used its AI platform in generating and analyzing huge amounts of biological and outcomes data from patients.
Niven Narain, Berg’s co-founder and chief executive, stated that the company aims to identify potential treatments based on the precise biological causes of disease. “We are turning the drug-discovery paradigm upside down by using patient-driven biology and data to derive more-predictive hypotheses rather than the traditional trial-and-error approach,” he said.
Aside from that, a team of pharma companies and tech providers collaborated to start a research project called MELLODDY (Machine Learning Ledger Orchestration for Drug Discovery). MELLODDY aims to apply machine learning and blockchain to data management in the drug development process. This can be done by collecting data from pharmaceutical companies and extracting those from multiple databases. Blockchain, on the other hand, will ensure data security.
“This project allows the pharma partners for the first time to collaborate in their core competitive space, invigorating discovery efforts through efficiency gains,” Hugo Ceulemans, MELLODDY’s project leader, said.
Creating Better Drugs
Recently, drug discovery researchers from Purdue University have created a new framework called Lemon to address the problem in creating a process for the computer to extract valuable information from a pool of data points. This framework helps drug researchers better mine the Protein Data Base (PDB), an essential tool for the drug discovery community. It contains more than 140,000 biomolecular structures and with new ones being released every week.
While machine learning can help in sorting through all the accumulated data in PDB, the researchers recognized the fact that a strong framework is needed to quickly analyze data in creating safe and effective drugs. According to Tech Xplore, an online site that covers the latest engineering, electronics, and technology advances, the framework was named Lemon because it was originally designed to make benchmarking sets for drug design software and identify the “lemons,” which are biomolecular interactions that cannot be modeled well in PDB.
With Lemon, data in PDB can be mined within a few minutes that originally took about 290 minutes. This framework allows the user to write custom functions as well as develop custom functions in a standard manner. These are aimed at generating benchmarking datasets for the entire scientific community.
Jonathan Fine, a Ph.D. student in chemistry, who worked with the researchers to develop the platform, stated that they created Lemon as a one-stop-shop to quickly mine the entire data bank. Also, to get useful biological information that can greatly help in developing drugs. "Experimental structures deposited in PDB have resulted in several advances for structural and computational biology scientific and educational communities that help advance drug development and other areas," he said.
There’s so much potential in AI when it comes to drug discovery and research. This can greatly help in reducing the costs and making it faster to be released in the market.
|While machine learning can help in sorting through all the accumulated data in PDB, the researchers recognized the fact that a strong framework is needed to quickly analyze data in creating safe and effective drugs / Photo by: Pöllö via Wikimedia Commons|