Contributed by Karen Madden, PerkinElmer
PerkinElmer is excited and proud to be a sponsor of this year’s Data Science Bowl, a competition that closely aligns with our commitment to innovating for a healthier world and providing novel solutions that enable scientists to make breakthrough discoveries faster.
This year’s challenge highlights the potential of high content screening (HCS) and its ability to accelerate the understanding of disease and development of new treatments – something that is close to home for us as we work daily to perfect the power of HCS systems and software. While HCS technology has been around since the late 1990s, its usage began to rise in 2011, when an article published in a well-known scientific journal stated that almost 40% of first-in-class drugs approved between 1999 and 2008 were discovered using phenotypic approaches. This is largely due to the improved imaging capabilities of HCS systems that allow researchers to rapidly screen a wide variety of molecules (including small molecules and biologicals), and measure phenotypic changes in relevant cellular models at a relatively low cost and high-throughput. Furthermore, the predictive quality of the biological models used in these screens has significantly improved, especially with the use of more complex and physiological relevant models including primary cells, induced pluripotent stem cells (iPSCs), co-cultures and now even 3D cultures and “organs”.
While HCS has become widely adopted, a large number of researchers struggle with the deluge of data generated from these screens and the data analysis, with most experiments utilizing only a small amount of information present in the image data, leaving much information undiscovered that might be important to making a new discovery. So ironically, many HCS experiments are in fact not truly “High Content”. Through improving the data workflow, along with image analysis (using artificial intelligence to detect a range of nuclei), we hope to see a solution developed that is able to take full advantage of the rich volume and diversity of the multiparametric data generated from high content screens so HCS will reach its full potential, hopefully leading to faster discovery and enabling new applications such as image-based precision medicine. We look forward to seeing the results of the competition, which will undoubtedly be a critical step forward in helping tackle some of today’s most serious diseases.
 Swinney, D. C., & Anthony, J. (2011). How were new medicines discovered? Nature Reviews. Drug Discovery, 10(7), 507–19. http://doi.org/10.1038/nrd3480
 Scannell, J. W., & Bosley, J. (2016). When Quality Beats Quantity : Decision Theory , Drug Discovery , and the Reproducibility Crisis.
PloS One, 2, 1–21. http://doi.org/10.1371/journal.pone.0147215
 Singh, S., & Carpenter, A. E. (2014). Increasing the Content of High-Content Screening : An Overview. http://doi.org/10.1177/1087057114528537
 Chia, S., Low, J., Zhang, X., Kwang, X., Chong, F., Sharma, A., … Dasgupta, R. (2017). Phenotype-driven precision oncology as a guide for clinical decisions one patient at a time. Nature Communications, 8. http://doi.org/10.1038/s41467-017-00451-5