Despite the steady increase in R&D spending, the five-year average of new drugs approved by the US Food and Drug Administration (FDA) remains stubborn at 50 drugs per year1. As the cost of bringing a new drug to market increases, estimates suggest that it has almost doubled over the past decade, it is important to improve R&D production to ensure the future functioning of the pharmaceutical industry and advances in health care.Many drug addicts fail before they can undergo clinical trials. Pre-clinical studies are designed to identify and validate a leader. It is in this important phase of the process that 50% of the money is wasted, reaching $ 28.2 billion in the United States alone each year3.
There are three main reasons for this development:
• Unnecessary. Scientists are often absent or oblivious to the research that other scientists have done in the past, even in their own organizations.
• Failure to test. Many experiments fail to produce useful data due to faulty reagents and protocols.
• Inability to reproduce. Even if testing is successful, it is often unable to reproduce, perhaps due to problems with reagents, scripts and data collection.
As a PhD researcher, Tom Leung, the CSO of BenchSci witnessed first-hand the waste of resources in pre-clinical studies. When he worked as a cancer researcher, he lost rare patient samples due to the wrong antibody. Problems with the selection of antibodies and their specifications are widely recognized as one of the main causes of negative, consistent and unproductive results in the health sciences.
Eager to address the problem, Leung met with David Q. Chen, Elvis Wianda and LiranBelenzon at the University of Toronto, Canada, to use AI in selecting antibodies. In 2015, they founded BenchSci and began building AI-enabled applications that analyzed the full text of expert articles to extract the right information to select the most suitable reagent for specific tests.In 2017, BenchSci launched its first program: AI-Assisted Antibody Selection. “Scientists had to spend hours or days reviewing published material to discover an antibody that might be useful in their experiments. With BenchSci they can get the details of the antibody confirmation within minutes, ”said Leung.
Since then, the BenchSci platform has expanded to include a wide range of reagents and model systems, including regenerative proteins, RNAi, cell lines, CRISPR, animal models and PCR. Today, more than 49,000 scientists in 16 of the top 20 pharmaceutical companies and 4,500 educational institutions are using BenchSci’s AI-Assisted Antibody Selection or Reagent Selection applications to organize successful trials, with guaranteed savings of millions of dollars a year.
It uses fewer but better tests
By helping researchers select the most suitable reagents in seconds, BenchSci aims to halve the time it takes for new drugs to reach the market by 2025.
“Approximately 80% of the initial phase of drug testing is not required for the development of a new drug in clinical trials,” said Leung. and very cheap. ”
The BenchSci method includes:
• Collection of terabytes of data from sources including scientific publications, catalogs of resources, and customer websites. This includes easily accessible publications, thanks to partnerships with publishers.
• The use of more than 100 machine learning models that cuts millions of informal scientific texts and extracts images, as well as text.
• An easy-to-use interface that organizes resulting information using proprietary ontology. This allows researchers to understand the biomedical value of the extracted data and to establish relationships between biological organizations. Users can easily search by target and reagent- or system type to find specific reagents and compare their specifications in minutes.
Meet customer needs
Building on the success of current applications, BenchSci is upgrading its technology to provide even more comprehensive solutions to help leading pharmaceutical companies solve their major R&D challenges.
For more read: BenchSci, Software powered by artificial intelligence accelerates drug discovery, Nature News (2022), https://www.nature.com/articles/d43747-022-00125-2
READ ALSO : Politics Focus: ED calls Sonia Gandhi, Rahul Gandhi for questioning in the National Herald Case
