How Life Sciences are steadily becoming Data Sciences

Whereas life sciences typically revolve around understanding the biological processes that make up the human body and mind, and how this can be impacted in disease and health conditions, it is clear that besides biology and chemistry, data sciences are becoming increasingly important in this sector. Despite the great promise, several critical success factors and barriers can be identified.

The advent of big data analytics, artificial intelligence (A.I.), real-world evidence, … have significantly impacted the way things are getting done in life sciences. Not only at the level of discovery and early research & development, but even in strategic decision-making or monitoring of (bio)manufacturing, data has taken center stage in current business practice. Having accessible and reusable data is often considered as a key growth engine for individual companies, but the potential benefits thereof lie far beyond the boundaries of any single entity. Nevertheless, it remains an intricate balancing act to collect and (re)use data in an ethical, correct way while also maximally leveraging its potential.

Data obviously holds a big promise to all stakeholders in the life sciences ecosystem. However, at the same time, generating real insights from different types and disparate sources of data requires dedicated skillsets and approaches that might not be widely available yet. Moreover, with data being generated in different settings (e.g., clinical trials, primary care, …), with different devices (and with different intentions), data integration and linkage needs to be established at various levels, which is not trivial.

To work with data, one first needs to better understand the different concepts surrounding data. First, there might be very different types of data needed to solve a particular question or address a particular need. For instance, when considering the best treatment options for an individual patient, one might want to combine different omics data with lab results and other elements from the patient record, whereas in early discovery, one would be more interested in discovering patterns in toxicity testing data.

Second, depending on the intent, it might be more important to have high volumes vs high quality, or assuring data integrity. Especially in those cases where regulatory compliance is important, the traceability and veracity of data needs to be assured at all times.

Once you have established which data are relevant for your specific case, the next step is to determine which data sources you will want or need to include. These might be strictly internal to the company, or sourced from a wide array of data partners, either via data sharing agreements, or third parties offering data-as-a-service or platform-as-a-service or other options. Again, data sharing in itself poses different challenges, among which establishing the right data sharing agreements and concomitant business models, safeguarding data security and privacy, and establishing the right levels of trust among the different stakeholders are deemed key.

Subsequently, data will need to get further processed, analyzed and visualized to generate answers and insights on the particular case at hand. This not only requires analytical capabilities, but also storytelling to support the correct interpretation on the outcomes.

In addition to the different barriers listed above, it should be noted that the field of data technology is rapidly evolving (e.g., data lakehouses, data fabrics, graph, solid technology), requiring different stakeholders in the ecosystem to continuously adapt their strategies and practical set-ups towards establishing data-driven healthcare.

The opportunity of data has now also been widely recognized by governments, with a.o., the European Commission recently announcing the creation of a Knowledge Center on Cancer, with a main aim to consolidate, track and provide access to cancer data, as an aid to supporting research and improving treatments. In Flanders, the spearhead cluster on health, Flanders.healthTech, was launched in June 2021, having data as a foundation and connector between its key themes.

With digital transformation only just really beginning to start and/or accelerate in many life sciences companies, and various initiatives co-creating the right infrastructure and tools for making data more readily accessible and reusable, it will be an exciting journey to really unlock the power of data in life sciences in the years to come.

Are you curious to know more about what the Flemish Life Sciences ecosystem has to offer in this domain? Come join us for the sessions on ‘Understanding data (13-14h)’ and ‘Data Availability & Use’ (moderated by Modis, 15.30-16.30h) at the Knowledge for Growth conference held in Ghent, Belgium today!