Posted on January 26th, 2019 by admin in Pharma R&D
The process of discovering a viable drug candidate molecule is often time-consuming and research-intensive. A lab must test thousands of compounds every day — and a hit rate of just 1 percent is considered above average.
Even after a potentially viable molecule has been discovered, a pharmaceutical developer must invest significant time and resources into in vitro and in vivo testing to determine whether the molecule is worth the larger investment of a clinical trial.
To help streamline this process, many pharma developers are turning to predictive analytics. This optimizes the process of discovering new candidate molecules and selecting the ideal candidates in which to invest.
Predicting Drug Behavior
In the past, drug interactions and toxicities could be difficult to predict without extensive in vivo testing. Today, on the other hand, many developers leverage multi-source data to create intricate mathematical simulations of a compound’s interactions with specific chemical systems in the human body. Starting from a library of 10,000 or more compounds, researchers can use these rapid computer simulations to narrow down their list to a mere 10 compounds and then focus their attention on those molecules.
Multiple Data Sources
A developer doesn’t need an in-house supercomputer to perform this type of compound analysis. As the world’s library of digitized clinical data continues to grow, developers can access a wealth of research findings from chemical studies, clinical trials and marketing reports around the globe. Specialists can then use their own in-house analytics software to analyze this data and predict the likely outcome of their own research projects.
Outsourcing to Partners
With the spread of cloud connectivity and software-as-a-service (SaaS), a growing number of companies — including IBM — are collecting and analyzing data on millions of compounds and generating vast libraries of reports on candidate molecules. Pharma companies can even cross-reference this data with their own in-house research in order to predict the pharmacokinetic behavior of newly discovered compounds.
From computerized molecular modeling to clinical trial forecasting, predictive analytics are transforming libraries of chemical and clinical data into actionable insights for pharma developers. These insights enable pharma firms to target their investments more precisely and deliver new drugs to market more efficiently than ever before.
This is a follow up piece to my original post on early stage drug discovery, exploring how non-conventional and preemptive actions can save an immeasurable volume of time, money and resources entailed to get to everyone’s ultimate end goal: Avoid Pitfalls and Costly Mistakes in Pharma R&D
Shawn Hooks is an account manager at Elsevier. He works with pharmaceutical companies, biotech companies and other life science corporate customers to help them optimize their R&D efforts by consulting on their research and workflow needs.
R&D Solutions for Pharma & Life Sciences
We’re happy to discuss your needs and show you how Elsevier’s Solution can help.