Over the last decade, the role of bioinformatics in drug discovery has been undergoing a major transformation. Bioinformatics expertise focuses on the development of computational tools and approaches to make sense of biological data. It has gone from being a good-to-have resource to a mission-critical asset for pharma and biotech companies across the globe.

Transforming Role of Bioinformatics

This transformation manifests itself in many different ways across individual companies and the industry at large. Companies like Benchling, a data management and collaboration platform for the life sciences industry, that are addressing major pain points for scientists are being heralded as the new wave of unicorn startups.

Vas Narasimhan, the CEO of Novartis, has emerged as a prominent voice sharing his commitment to making Novartis ‘a focused medicine company … powered by data science and digital technologies’

Flagship Pioneering demonstrated its commitment to digital technology and data-enabled capabilities recently. They hired CIO of Novartis as the Chief Information and Digital Officer to further their goal of creating ‘first-in-category life sciences companies’. Closer home at Elucidata, we have had early-stage biotech companies partner with us for our bioinformatics capabilities even before hiring their first employee. These movements are not limited to startups and biotech companies. CEOs of many major pharma companies have spoken about the role that big data and artificial intelligence (AI) is playing and will continue to play in shaping the pharma industry.

Driving Forces Behind the Transformation

Advances in instrumentation and experimentation methods have resulted in a dramatic increase in the volume, variety, and velocity of the generation of biological data and simultaneously decreased the cost associated with generating this data. Growth in gene sequencing data is discussed most extensively. It is expected to outgrow data generated by Twitter and YouTube by 2025. Advances in mass spectrometry methods and phenome centers coming up in different parts of the world to support metabolic phenotyping.

It has resulted in a rapid increase in the generation of other kinds of omics data. For e.g., metabolomics that studies metabolites (small molecules) participating in cellular reactions. Today, biological big data is among the fastest growing categories of data. Governmental and non-governmental bodies have worked extensively on funding and creating public data sets. For e.g. The Cancer Genome Atlas, Metabolomics Workbench. These have the potential of transforming the way we understand diseases and consequently, develop cures.

Implications for the Bio-Pharma Industry

The focus is now shifting towards processing this data more effectively and deriving relevant insights to realize its full potential. Biological big data has specific relevance for the development of drugs which has been plagued by long timelines (~15 years), high costs (~$2B) and low odds of success (only 1 in 20 clinical trials are successful). Bioinformatics is becoming critical to better understand the underlying mechanism of disease and to develop safer drugs in a faster and more effective manner. Some bio-platforms allow scientists to process and generate insights from large biological datasets. They enable the focus to shift from pursuing a specific gene or mutation and applying trial and error methods for finding drugs, to developing a more holistic understanding of how the disease functions. Consequently finding a drug candidate that is more likely to succeed in a clinical trial.

This is not to suggest that big data and more specifically AI should be seen as a panacea that will transform the industry overnight. There has been valid criticism from different quarters that ‘AI’ is often applied without incorporating relevant domain expertise. This creates a ‘black box’ solution. In such cases, the ultimate users i.e., the scientists provide limited input and hence have limited engagement with the product. The focus should be on generating better quality data instead of obsessing over the application of AI.

At Elucidata, we are committed to keeping the scientists at the center of our design process. We enable them to use the data they generate from their experiments as well as leverage data from public data sets in the most effective way. We also focus on combining deep domain expertise with strong engineering capabilities. We’ve created a robust platform that meets the computation needs of drug development in the 21st century.

The companies developing drugs in the future will look very different from the companies of the past. Traditional methods focused on experimentation in labs will continue to create major value. But methods focused on bioinformatics capabilities will become increasingly important complements to traditional methods. We are excited to offer our cloud-based platform, Polly to labs across academia and industry. We support end-to-end data processing and analysis for target discovery. This way, we play our part in empowering the next phase of drug development.