Skip to main content
Premium Trial:

Request an Annual Quote

AstraZeneca Says Its SysBio Approach Leads to More Predictive Disease Models

Premium

SAN FRANCISCO — As part of their drug-discovery efforts, scientists at AstraZeneca have developed and are using multiparametric high-content toxicity and phenotypic assays that integrate lab automation, image and statistical analysis, and in silico modeling, a company official said at a high-content screening conference held here this week.

The approach may have several advantages over more traditional, target-based drug-discovery models, Neil Carragher, an associate principal scientist at AstraZeneca R&D’s advanced science and technology lab, said at the meeting.

In its methodology, “no assumptions are made about the target, compounds are categorized according to the mechanism of action, compounds with a novel mechanism of action can be more easily identified, a large target space can be tested in each assay, compounds that target multiple pathways can be identified, and new targets can be identified,” Carragher said.

Carragher discussed AstraZeneca’s use of high-content phenotypic and toxicity assays during a presentation at Cambridge Healthtech Institute’s Sixth Annual High-Content Analysis Conference.

The AstraZeneca investigators ran cell-cycle, -morphology, and -apoptosis assays to develop phenotypic fingerprints, said Carragher. During his presentation, he used as a toxicity example the development of an assay looking at steatosis, or the abnormal retention of lipids within a cell, which he said is a key pathology for AstraZeneca’s drugs in development because it results in perturbation of liver enzymes, inflammation, fibrosis, and ultimately, loss of liver function.

He said the he and his colleagues asked, “’Can we use fluorescent neutral lipid dyes and automated imaging to predict in vitro steatotic compounds?’” According to Carragher, the answer was yes.

Carragher said that his group uses Definiens’ Cognition Network Technology because it is context-based and identifies objects rather than just examining individual pixels. It then makes inferences about those objects by looking at them in context.

Other tools and technologies used by Carragher and his group are Molecular Devices’ ImageXpress and GE Healthcare’s IN Cell Analyzer 3000 for image acquisition and storage; Photonics’ IncuCyte incubator imaging system, Chipman Technologies’ Cell-IQ, and ECI’s TaxiScan for kinetic and 3D cell assays; Olympus’ OV 100, CRI’s Maestro, and VisEN Medical’s VisEN FMT for in vivo and ex vivo fluorescent imaging; and Accelrys’ Pipeline Pilot for automated data processing.

Carragher said that in the future, AstraZeneca’s advanced science and technology laboratory will focus more on an approach that combines experimental knowledge gleaned from in vitro and in vivo biological models, perturbation experiments using things such as RNAi and small molecules, and high-dimensional genomic, proteomic, and phenotypic data with computational models derived from external data, cluster analysis, and pathway, network, and system generation.