3 Hurdles to Accelerating Scientific Discovery with Data

Global life science organizations have faced significant economic challenges in 2023, including labor shortages, inflation, and competition from emerging biopharma companies. To overcome these challenges and optimize for future growth, business units within these organizations are increasingly investing in technologies aimed at automating and accelerating arduous and costly workflows.

However, to truly accelerate drug discovery amidst the hundreds of terabytes of data generated each year, life science organizations must leverage collective knowledge at every step of the therapeutic journey. Despite ongoing investments in digital technologies to progress this effort, the biggest barrier to success has been data management and data governance, including an inability to get the right data to the right users at the right time.

But there are also cultural, technical and operational challenges that can make it difficult to get the most out of your data. We outline why below.

1. The culture of science and data silos

Nearly half of senior decision-makers working for life science organizations report that data silos derail the efficiency of cross-functional collaboration in their organization. This siloed data is either not stored or not stored in a manner that democratizes its use. 

Data silos represent one of the greatest contributing factors to the high cost of drug discovery. The inability of researchers to access current and legacy data — and the untapped insights and potential they might provide — often leads to redundant work, prolonged timelines and higher failure rates.

2. The technology quandary

Despite billions of dollars in digital investments over the past few years, life science organizations still trail most industries in digital maturity. With 65% of molecules in the global R&D pipeline coming from emerging biopharmas, large organizations recognize the importance of harnessing technology to transform traditional workflows that cannot keep pace with the industry. Examples include AI-enabled early drug discovery, NLP-driven clinical and operational workflows, and smart factories.

Still, for technology to achieve these business outcomes, the data it relies on must be clean, available, consumable and have integrity. Disparate, decentralized data sources and poorly interconnected systems abound within the life sciences industry. From instrument data to omics datasets to imaging content, most data generated inside biopharma is unstructured, unconnected and difficult to access. 

Despite the promise of the power of artificial intelligence, most organizations are nowhere near ready to harness the power of AI due to challenges with computing, storing, accessing and securing the underlying data.

3. The truth about operations

"Process" is the all-important third leg of the people and technology stool. Operational leaders in life science organizations are tasked with ensuring that the aggregation and storage of existing data are executed in a manner that enables advanced analytics and insights crucial to accelerating therapeutic discovery. They must also simultaneously implement processes that ensure newly generated data is optimized for use and reuse with protocols around data format, data acquisition, data storage, data quality and data sharing.

Next steps

Based on our experience, we would focus on creating predictable, scalable cost structures with strong central governance controls. Here are some of the areas that we would investigate:

• Establish globally optimized dataset placement with well-documented transformation pipelines that centralize best practices in an enterprise data fabric to create hard and soft savings for additional innovation.
• Establish a centralized strategy for core foundational models to be shared and customized by multiple teams — making it easier to harness the potential of many.
• Design, deploy and support comprehensive security ecosystems to ensure all services provide the confidentiality, integrity and availability necessary to fulfill operational requirements.
• Test features in a secured custom lab environment where life science organizations can test and validate competing architectures and solutions.
• Based on geographic footprint, create a global integration, logistics and distribution plan to rapidly receive, configure, and deploy products and services to client locations around the world.

No matter where you are in your journey to data maturity, a focus on data management, analysis and planning are key components to help you achieve your drug discovery goals.