Thursday, April 24, 2014

Data rich, analysis poor

A recent research (1) that shows studying the effects of drugs on yeast could inform drug interaction effects on humans, is revealing. Life sciences companies have lagged in the use of rich data they have collected from past studies. Although the yeast studies are potentially revealing, there is significantly more information hiding in the data from past experiments on humans.

Drug interactions are important from multiple perspectives – some indicating a reduction in efficacy and others showing enhanced toxicity. The yeast studies show interactions between drugs at very high levels, albeit in a different biological entity. The tendency of life science companies to collect “hard data,” holds them back. Every other industry is using data better to predict, act and optimize. It is ironic that the industry that thrives on data is bringing up the rear of analytics.

More importantly, we are fast approaching a regime in which personal efficacy and toxicity could be reliably predicted by individual. One could slow down innovation but it is unlikely that big pharma can escape the coming revolution in data analytics.

(1) Large-scale identification and analysis of suppressive drug interactions, Lunenfeld-Tanenbaum Research Institute