Part 2 of this two-part article analyses the new data-driven decision-making approaches that are making a difference to life science R&D.Big data is one of the hottest topics in R&D as its potential to accelerate successful drug development becomes more widely recognized. The data deluge is challenging life science companies’ capabilities to apply powerful analytics that uncover usable insights. While there is no technological silver bullet that aggregates, mines and extracts information, a combination of approaches will streamline and improve R&D.
Data-driven decision-making
The power underlying big data is its ability to unlock the intrinsic value of data-driven decision-making. For example, improving target selection requires mining diverse sources of evidence, including experimental, clinical and published data. Robust analytical solutions can process multiple types of data sets and use sophisticated entity recognition and pattern matching software to identify meaningful associations between targets and molecules. These insights can identify novel relationships between entities and may increase the probability of new drug discoveries.
In R&D, no single technology provides all the answers. Rather, a combination of research modalities provides more comprehensive and reliable information. For example, integrating information from DNA/RNA screens, pathway analysis algorithms and text mining enables scientists to build more reliable molecular networks and gain new insights into disease mechanisms and potentially new therapeutics. Ultimately, big data analytics powered by expert capabilities in data aggregation, structuring, normalization and integration is generating data-driven decisions that successfully impact patient care.
Big data challenges
Life science stakeholders, including pharma/biotech, academia, government, diagnostic companies and medical centers, are now inextricably linked to an ecosystem that is driven by big data and new technologies. Personalized patient care and treatment decisions are being informed by big data, which is facilitating the application of comprehensive genomic, pan-cancer diagnostics.
What will it take to expand the scope of personalized medicine? Consider this: The Journal of the American Medical Association recently published an exploratory study of whole-genome sequencing to detect clinically meaningful genetic variations among 12 adults. The investigators assessed the prevalence of rare, potentially pathogenic genetic findings, genetic risk of disease, genetic drug response predictions and the burden/cost of clinical follow-up. The results were sobering. The authors found incomplete coverage of inherited disease genes, low reproducibility of clinically relevant genes and disagreement regarding the significance of the findings. These limited results underscore the opportunity that big data offers to help the scientific community better understand patient-specific information, draw appropriate clinical conclusions and bring viable targeted therapies to market.
Big data challenges revolve around structure, quality, actionability and standardization:
- Data structure and quality: Most medical data is unstructured, yet clinically relevant. Also, cost-effective and efficient sequencing is a major driver of personalized medicine. But, sequencing variability can compromise data quality and analysis
- Data actionability: Identification of faulty proteins that are disease harbingers can lead to early detection, guide treatment and determine relevant lifestyle modifications. Yet, in oncology, less than five percent of cancer patients currently participate in clinical trials. Therefore, there is a wealth of data that can be extracted and analyzed from the remaining more than 95 percent of non-participating patients.
- Data standardization: Current efforts to establish uniform analytical protocols will enable R&D to more effectively progress from data generation to analysis. Data standardization will allow organizations to integrate information from multiple and diverse sources in order to provide more accurate answers to R&D queries. In addition, standardized data will enable effective R&D collaboration by facilitating information exchange and utilization among internal and external stakeholders
Looking ahead
Life science stakeholders are at the forefront of tackling these big data challenges and are using real-world data to inform and transform patient care. An accelerated influx of big data is being driven largely by patient-centric data sources, including electronic health records, mobile diagnostics and monitoring applications.
In parallel, new sequencing initiatives aimed at better understanding disease mechanisms continue to launch, producing even more data that needs to be integrated, standardized and analyzed. For example, the U.S. National Institute of Allergy and Infectious Diseases recently awarded a five-year, $25 million grant to the J. Craig Venter Institute to establish a genome center for infectious diseases. This center will apply next generation sequencing and other technologies to gain new insights into infectious pathogens, including viruses, bacteria and parasites, and to characterize the genomic variations in infectious diseases.
In conclusion, traditional data frameworks and tools cannot deliver adequate solutions to today’s research and clinical challenges. To increase R&D productivity and to improve patient care, life science stakeholders need better data management systems and advanced analytics to successfully unlock the power of big data and bring new therapies to patients.
Filed Under: Drug Discovery
