AI & Machine Learning Helps Scientists Deal with Data Overload

Biopharmaceutical companies must collect and analyze an astonishing amount of data to take a product all the way to market. This data is collected by many different groups along the product development pathway. Furthermore, large companies have international teams who need to harmonize experimental conditions so they can combine datasets.

These complexities of product development in the life sciences industry mean that scientists, managers and biopharma executives spend a lot of time trying to understand what their data means. Many are turning to artificial intelligence (AI) and machine learning tools to help shift through reams of data.

AI Can Help Cross-Talk Between Cross-Functional Teams

During the 2019 BIO International Conference, LabTwin and Informa Pharma Intelligence invited a panel of experts to talk about how AI and machine learning can help life science companies deal with data overload.

The experts - Ron Alfa, SVP Translational Discovery, Recursion Pharmaceuticals, Krishnan Nandabalan, President and CEO, InveniAI Corporation, John Piccone Principal ZS Associates, Seth Lederman, CEO and Chairman, TONIX Pharmaceuticals, and own very own Magdalena Paluch, CEO and co-founder, LabTwin – discussed the problems biopharmaceutical companies face when trying to collate data collected by many different scientists at multiple sites, over many years.

“In big pharma, they’ve got massive amounts of data across many decades of experimentation and studies, and what they’re looking for is insights from cross-study analysis,” said John Piccone. “Looking at the enterprise level, looking across programs – companies need to be able generate insights and knowledge across all of this data. The ability to do this is a differentiator.”

The panel experts discussed how AI can help collect, sort, store and share data. AI-powered data integration facilitates collaborations, helps cross-functional teams to work together and makes it easier to protect intellectual property.

Magdalena explained how tools like LabTwin let scientists put a specific time stamp on an idea or dataset. “So later on, for IP purposes, such annotations can be tracked to the very first moment that they were discovered and this makes traceability much more manageable,” she said.

Scientists May Not Know About the AI Tools Available

One of the problems highlighted by the panellists is that many scientists still don’t know about the AI tools available to them and don’t know how to use new tools. The expert panel agreed that education is the best way to show scientists that AI and machine learning tools are not a threat. AI cannot replace humans - instead AI tools can eliminate mundane, repetitive tasks and free up human scientists for higher thinking and experimental design.

“We want to create tools for scientists to enable them to teach the machine learning algorithms and control them – this is what is needed at the moment to offer insights and innovation later on,” said Magdalena. “Our goal at LabTwin is to enable scientists to have access to information at the point of experimentation and be able to document (their findings) without effort. A liveable and interactive process.”