IDBS BlogData-driven transformation in the lab

IDBS Blog | 30th April 2024

Data intelligence requires an intelligent, data-driven approach to digital transformation

By Unjulie Bhanot, Product Marketing Manager (Process Development & Manufacturing), IDBS

In BioPharma, there is a continuous effort to speed up drug discovery, accelerate clinical trials, quicken regulatory filing and shorten time to market. The BioPharma industry is increasingly turning to digital tools like artificial intelligence (AI) and machine learning (ML) to be a cure-all toward achieving these outcomes – some of which may prove significant. Consider that it typically takes 12 to 15 years to bring a drug to market, according to Boston Consulting Group. The firm says AI-driven R&D could help cut 25% to 50% of the time and cost of bringing drug candidates to the point of human testing.

However, digitalization can be a double-edged sword. While the technology creates opportunities for enhanced data capture and access to higher-quality data in the research lab, the systems also add complexity to the regulatory decision-making landscape.

It is important to recognize this dichotomy as you begin your digital journey. This endeavor must be taken slowly. In other words, you learn to walk before you run. 

Regulatory authorities acknowledge the importance of taking a methodical approach to using AI-generated data for quality decision-making. This past December, the European Medicines Agency (EMA) and the EU Heads of Medicines Agencies (HMA) got together to publish the first full release of the Data Quality Framework for EU Medicines Regulation. The publication aims to provide definitions, principles and guidelines to characterize, assess and assure data quality for regulatory decision-making.

Enabling data-driven transformation

A Deloitte survey of biopharma development teams showed that >51% of respondents were yet to adopt cloud technology in their digital transformation journey and their day-to-day work, including recording process development work. 57% admitted that they tend to follow the path of competitors after they have proven the value of technologies. This unwillingness to take the risk first delays progress, and this can lead to encountering problems such as late biologics license application, ceasing projects and repeating studies. Relying on disparate, disjointed legacy systems is, simply, inefficient.

Building a more efficient, data-driven, integrated process that drives operational efficiency is a better solution. But such a transformation requires having the essential building blocks. Ken Forman, Lead Product Manager, Manufacturing at IDBS, explains in a recent Lab Manager article that turning lab data into intelligent data using advanced automation is indeed possible – if you know how to make AI/ML digital transformation perform to desired expectations. This progression must be systematic and may be achieved in five well-planned steps.

  1. Ensure your data is complete

Let’s begin with the data. Forman says rather than archiving the data from a failed experiment, ML can yield useful information from that data. ML models can perform analytical reviews on trial parameters and ultimately make determinations based on failed and successful experiments. Laboratory data is fertile ground for AI. Data-driven quality control can alert labs to instrument trends and deviations. Analyzing data can improve resource allocation and identify emerging patterns around degrading data or process integrity.

Your data should be complete, current and unique, with no missing entries. For instance, if you only rely on the sample-centric data generated from your laboratory information management system (LIMS), then you may be putting your faith in partial data. This would require gathering process data from other sources across your lab to create a more complete picture.

  1. Make data accessible

At least for the time being, automation cannot completely replace humans. The research lab is no exception. If you are relying on disparate, siloed systems to store your data, retrieving high-quality data can be difficult. Additionally, that data must be intelligible to humans and automation. Forman says that he has witnessed many companies begin to funnel data from disparate systems into a single data source. While this sounds good in theory, he says that this process performed in retrospect can be resource-intensive. Instead, consider implementing a system that channels all data to one location as you go along while maintaining data integrity. 

  1. Capture and curate high-quality data

A key piece of your project should focus on the implementation of a digital backbone. IDBS Polar offers a digital data backbone that connects the pieces of drug development and can help navigate your digital journey to data-driven transformation. The BioPharma Lifecycle Management (BPLM) platform is ideal for data capture at the point of execution, process optimization and advanced analyses, data visualization and reporting. 

The cloud-based platform curates the data needed for process design and optimization as well as scale-up and tech transfer. The overall benefit is capturing valuable, high-value quality data. As a result, effective lifecycle management can bring biological products to market faster.

Platforms like IDBS Polar that align, structure and contextualize data from project nose to tail make it easier for users to perform sophisticated data analysis. Forman says: “The synergy between AI/ML and low-/no-code tools ensures that high-quality data is accessible and actionable, enabling users with varying levels of expertise to contribute to data-driven decisions.” 

  1. Ask the right questions

To achieve data-driven transformation, Forman says this begins with the “right data and the right people asking the right questions.” Bench scientists are uniquely positioned to ensure that the right questions are being asked, such as: Which data matters? How should the data be organized? The responses to these questions will be unique from lab to lab. 

Labs that partner with industry experts who understand process development will often have the data science skills to help ensure the right questions are being asked. Additionally, data experts can help ensure that the right data is captured in ways that can trigger relevant business and scientific questions for desired outcomes.

  1. Create a shared digital foundation

Forman concludes that true data-driven transformation is within reach when an organization has a shared “foundation of digital literacy around how AI and ML models work.” He adds that this foundation should stress the importance of high-quality data.

He says: “True data intelligence requires an intelligent approach, from beginning to end, with high-quality, well-organized data supported by knowledgeable, thoughtful humans.”

Give data intelligence time to grow

CRO labs are still at the forefront of implementing AI in clinical development (Life Sciences Review). The McKinsey Global Institute has estimated that the technology could generate $60 billion to $110 billion a year in economic value for the pharma and medical-product industries, largely because it can boost productivity by accelerating the process of identifying compounds for possible new drugs and speeding their development and approval.

Forman agrees that with the advancement of automation and new drug modalities, modern labs generate more data than ever. “Lab leaders want to make data actionable or achieve data intelligence from their data pools,” he says. “They know their laboratory data could help their businesses better — they just need to harness it.” And take it slowly.


About the author

Unjulie Bhanot, Product Marketing Manager, IDBS. Unjulie Bhanot, Product Marketing Manager (Process Development & Manufacturing), IDBS

Unjulie Bhanot is the Product Marketing Manager for Process Development and Manufacturing at IDBS. With over 10 years of experience in the Biopharma informatics space, she has led the strategy and development of IDBS’ Bioprocess solutions and was instrumental in the launch of IDBS Polar™ to market. 

She joined IDBS in 2016 and spent over three years as part of the Global Professional Services and Solutions Consulting teams, where she was responsible for presenting business and technical value of IDBS solutions to customers. In 2019, she assumed a leading role in the Product & Strategy team, establishing enterprise solutions for Biopharma, and closely continues her relationship with this domain today. 

Prior to joining IDBS, Unjulie worked as an R&D scientist at Lonza Biologics and UCB. She received a BSc in Biochemistry and MSc in Immunology from Imperial College London. 



  1. European Medicines Agency and EU Heads of Medicines Agencies. (n.d.). Data quality framework for EU medicines regulation. Evidence Base Online. Retrieved from []
  2. Bloomberg. (2024, January 31). Big Pharma turns to AI to speed up drug research and development. Bloomberg Newsletters. Retrieved from []
  3. Deloitte. (n.d.). Biopharma digital transformation: From AI to genomics. Retrieved from []
  4. CRO Life Sciences Review. (n.d.). AI reshaping clinical development: The CRO landscape in 2024. Retrieved from []
  5. McKinsey & Company. (n.d.). What’s the future of generative AI? An early view in 15 charts. McKinsey Explainers. Retrieved from []
  6. Lab Manager. (n.d.). Digital maturity requires data maturity. Retrieved from [
  7. IDBS. (2024, February). Digital maturity requires data maturity. Retrieved from []
Check out our Blog section for more like this

Flere nyheder