How Collaboration Can Help Research Go Further

The following article is an opinion piece written by Nick Portch, Equinix. The views and opinions expressed in this article are those of the author and do not necessarily reflect the official position of Technology Networks.

The life sciences sector is experiencing a drop in funding globally, and as a result, labs are looking for ways to make research go further and extract insights from legacy data using modern tools such as artificial intelligence (AI). So far, this approach has led to some promising early results.

However, due to privacy laws and a focus on protecting intellectual property (IP), the sector is guilty of repeating trials already completed by other researchers. This duplication of effort is often unintentional, and is a result of data being stored in silos where it cannot be easily accessed for further study by industry peers. This containment of data limits humans but also AI; having a smaller pool of training data limits the potential AI can reach.

An answer may lie in greater data collaboration. Research in life sciences is heavily reliant on data analysis and insights, and one way to stretch budgets further and multiply the impact of research results is to collaborate and combine insights. There are substantial roadblocks to achieving this, both regulatory, financial and technical, but the real challenge is awareness; life sciences researchers already have tools which enable compliant collaboration, but all too often adoption of those tools is low.

One technology-enabled approach to unlocking data collaboration in life sciences is the use of colocation data centers, which can ensure full regulatory compliance and robust security.

The scale of the challenge

A major challenge for collaboration in life sciences is the confidentiality of the data. This type of data is often the result of significant business investment and is often derived from strictly regulated sources such as medical diagnostics. The health sector is producing around 30% of all data generated globally, with the volume of data created growing 6% faster than manufacturing, 10% faster than financial services and 11% faster than media and entertainment.

Moving this data out of in-house storage and into colocation data centre facilities has created an opportunity for life science researchers. While data traditionally remained strictly within the confines of an organization to protect intellectual property, it also had the unfortunate consequence of stunting collaboration and limiting progress on drug discovery.

With colocation, new opportunities for efficiency are unlocked and broader insights can be captured as data sets grow. The result: accelerated time to outcome on potentially life-saving drugs and diagnosis. 

As well as expediting discovery, data center operators with colocation offerings have also developed dedicated departments to stay abreast of legislation and develop industry-specific solutions, ensuring compliance with key regulation such as General Data Protection Regulation (GDPR) and European Health Data Space (EHDS). 

Pressures on the pharma industry

Data sharing can be a controversial and challenging issue, especially between competitors in the pharmaceutical industry who fastidiously guard their IP. But the forces that encouraged such intense competition in the first place have abated. While patents notionally last for 20 years, the reality is that they now deliver effective market exclusivity for 7 to 12 years by the time the drug comes to market, and it can be as few as 5 years, despite the fact that developing a new drug can cost $300 million–$2.6 billion.

This backdrop of rapidly increasing drug development costs and need to accelerate post-patent drug development is driving competition in the field. The advantages of innovation and acceleration to market launch outweigh the customary need for secrecy, while the new therapies also stand to benefit patients through lives extended and enhanced.

We have already seen the benefits of collaboration in the time compression of drug development, and we expect continued improvement thanks to AI and deeper data-sharing practices which enable drug development to be completed within four to eight years.

In the future, the potential of more advanced AI and the impact of quantum computing might even reduce drug development to months. If you plot the current rate of development on a graph, AI-driven increases in efficiency could break through the current plateau, allowing new therapies to be developed in under six months.

The collaborative data approach has the potential to save, by our calculations, around $300m a year, while accelerating cures for major diseases.

Calling Dr. AI

Data analytics from research into life sciences will inevitably move out of hospitals or research institute-based IT rooms and into data centers. The sheer scale of the data mountain, the need for intensive compute and the requirement for regulatory compliance increasingly preclude on-site data processing and analytics.

An additional upside of this transformation, if data is shared in a co-located data center, is that AI can be implemented far more effectively on a larger pool of training data. Accelerated discovery of new therapies is the remarkable potential for this approach.

Uploading an MRI scan of a patient could go far beyond assessment of the individual patient and instead contribute to a much broader knowledge base which researchers and AI models can assess over time. Any results unearthed by AI models will be verified after the fact by trained specialists, ensuring quality of care while streamlining workflows for medical professionals and researchers.

Optimizing the transfer of data between researchers has several confirmed benefits, improving efficiency for researchers, reducing costs and reducing development times substantially.

Now that the technology and collaborative frameworks are in place to share this data cost effectively, the industry must take advantage of this opportunity to benefit researchers, the firms investing in research and most importantly, the patients. These benefits above are already guaranteed, but the most exciting quality of data is that once enough is aggregated, novel benefits that we are unable to predict today will emerge.