Necessity is said to be the mother of invention, but the jury is still out on the success of the desperate and rushed phenotypic screens that were done a few months ago to find the first effective pharmacologic treatment for COVID-19 disease.
The highest profile case of these was, of course, the use of hydroxychloroqine for COVID-19, in which the phenotypic screen had an n of 1 and was done directly in humans. Not something you see every day, nor something you would want to. More reasonably, testing drugs active against other viruses on the COVID-19 virus in a laboratory dish seems more the way to go. Indeed, there seems to be some weak benefit to treatment of COVID-19 with Gilead’s remdesivir. A host of other phenotypic screens were done and are ongoing, including additional shotgun trials of other existing drugs directly in humans. Overall, the COVID-19 emergency has been a crash course in phenotypic screening.
Well, why not take advantage and dive deeper into the science and technology of phenotypic screening for drugs? It's a thriving and diverse world of assay development and data analysis. A two-day virtual conference this week at the New York Academy of Sciences looks specifically into addressing some of the challenges of phenotypic drug discovery using state of the art computational biology and chemistry approaches. The biggest of these is so-called target deconvolution, where, once you find the drug that cures the disease in the dish, you have to find out what gene products it is actually working on in the cell (its targets) to bring about its beneficial effect. That is a thorny problem, but is required for drug optimization for clinical use usually. Big data, such as the readouts from high throughput phenotypic assays, which can include complex image data, as well as transcriptomics, proteomics and other data on the disease and the assay can be leveraged to meet the deconvolution challenge. GeneCentrix is an event sponsor and will be there to put in our two cents with our paper: "Historeceptomics Improves the Yield of Phenotypic Screens." Join us!