Toronto-based artificial intelligence startup BenchSci has raised $95 million Canadian dollars, or $70 million, in funding to help scientists discover new drugs faster.

BenchSci, officially Scinapsis Analytics Inc., announced the Series D round on Thursday. It detailed that Generation Investment Management was the lead investor. Returning backers iNovia Capital, TCV, Golden Ventures and F-Prime Capital participated as well, bringing the startup’s total outside funding to $170 million.

BenchSci provides a software platform called Ascend that uses AI to speed up the drug discovery process. According to the company, Ascend is used by 16 of the world’s 20 top pharmaceutical companies.

“Thousands of scientists use our platform every day to move their most promising drug discovery projects forward faster,” said BenchSci co-founder and Chief Executive Officer Liran Belenzon. “This investment is validation of the incredible work of our team and the traction we have in the market with the largest pharmaceutical and biotech companies in the world.”

Ascend uses AI models optimized for text and image processing to analyze research teams’ scientific datasets. The models also ingests external information, such as data from publicly available research papers. Ascend then scans the collected data for insights that can help scientists work more efficiently.

One of the most important tasks involved in the drug discovery process is finding what’s known as a target. That’s a protein or some other biological building block with which a drug can interact to cure a disease. Currently, finding targets is a highly complicated and time-consuming process.

According to BenchSci, its Ascend platform can ease the task. The platform identifies potential targets and displays them in a centralized dashboard along with key scientific data about each item. From there, scientists can compare the potential targets to identify which should be prioritized as part of their research project. 

BenchSci also promises to speed up subsequent phases of the workflow. After scientists narrow down the list of potential targets, they review existing research papers about each remaining target to evaluate its suitability. The company provides features that enable scientists to more quickly find relevant research. 

Once a target has been selected, researchers carry out a series of experiments to determine if it could become the basis of a new drug. The goal of the experiments is to quickly identify any factors that may rule out the target. The sooner scientists can complete that task, the more efficient the research workflow becomes.

According to BenchSci, its platform allows research teams to create an “AI-augmented experiment path.” The path is designed to help scientists rule out unsuitable targets as quickly as possible. It claims its platform can reduce unnecessary experiments by as much as 40% in some cases.

The platform also helps scientists optimize the experiments they do carry out. Its AI algorithms can identify research techniques that have been applied successfully in earlier, similar projects. The company likewise promises to ease the task of finding reagents, materials that scientists use to cause chemical reactions.

BenchSci says its platform can significantly speed up the drug discovery process. The platform is capable of saving an average of 12 to 18 months per research program, according to the company. That value proposition has helped the startup amass an installed base of more than 50,000 scientists at more than 4,500 research centers worldwide.

Image: BenchSci