Updated Accelrys Discovery Studio Targets Protein Aggregation

By Kevin Davies 

July 13, 2011 | With the latest release of its flagship modeling and simulation software, Discovery Studio, Accelrys is hoping to boost scientific collaboration and efficiency, as well as address a major challenge in the development of biotherapeutics.   

“The whole process of collaboration and communication is becoming more and more of a challenge,” says Adrian Stevens, the UK-based senior product marketing manager for Accelrys’ life sciences portfolio. “Within Accelrys, we have two opportunities. First, take new science to market and attract new customers in biotherapeutics, to give them a more effective tool for prediction. The new part is putting this on the Pipeline Pilot platform.”  

The new release incorporates what Accelrys says is the first commercially available software for predicting protein-protein aggregation to advance biotherapeutics research. It enables protein engineers to identify the location of regions on antibodies prone to aggregation and then predict substitutions to improve molecular stability.   

“With its flexibility and improved visualization capabilities, Discovery Studio 3.1 enables computational experts to develop sophisticated predictive models and communicate them clearly to scientists who need them to identify leads and optimize candidates,” comments senior VP/chief science officer, Frank Brown.   

Advances in Aggregate   

The launch of Discovery Studio 3.1 comes in the midst of severe industry challenges, including the patent cliff and multiple industry mergers. A silver lining is the growing attractiveness of biotherapeutics such as monoclonal antibodies in pharma pipelines. “It’s been quite a resurgence,” says Stevens. “The average pharma pipeline is heavily loaded with biotherapeutics candidates.”  

However, a major problem in biotherapeutic development is that of protein aggregation, which increases the rate of protein degradation. “Once the protein aggregates out, it will trigger a massive immunogenic response. That’s not suitable for delivery into patients,” says Stevens. “These molecules are typically stored at high concentrations, but you can’t have it crashing out of solution. With a synthetic analogue, you’re inadvertently increasing the propensity to aggregate.” Stevens says one pharma customer equated the cost of aggregation to $300,000/day.  

Aggregation occurs because of hydrophobic (water repelling) residues on the protein surface. But some of those residues might also be the crucial driving force for protein-protein recognition. To address this problem, Accelrys has in-licensed what Stevens calls a “spatial aggregation propensity code” from the MIT lab of chemical engineer Bernhardt Trout. “We’ve taken the academic scripts and code, put them into a GUI for accessibility, and included it into the design workflow.”  

Trout’s group has had a fruitful four-year collaboration with a large pharma (Stevens declined to name the company) to identify the molecular basis of aggregation. “It’s the only example looking at experimental validation that has been subjected to peer review to identify [aggregation] regions on a protein,” says Stevens. From there, it’s possible to computationally rank each residue’s relative propensity for aggregation.  

“The aggregation site could be in a region that is part of the protein-protein recognition domain (epitope),” says Stevens. “You don’t want to knock out the potency or create an unstable antibody in the process. You want to say, ‘Here’s the binding site, here’s the aggregation site, let’s do the mutagenesis work… By having it in this workflow, you now have the tools available.”  

The tool allows various levels of “smoothing,” says Stevens. For example, a researcher can examine a region of aggregation, zoom into the “rogue residues” as well as their location and context on the protein surface it. “Sometimes you find you need to keep the residues but you have to tune residues around it,” he says. By passing the MIT code onto commercial software developers, Stevens says Accelrys has been able to optimize the code. “Instead of 24 hours, a calculation only takes two minutes,” he claims.   

Another challenge is that of collaboration and data sharing across vast geographic regions. “We need efficiency enhancements,” says Stevens. “There are downstream challenges in the in silico market, challenges that require scalable enterprise environments.” Stevens says the integration of the Pipeline Pilot platform addresses many security issues for pharma, and provides better scalability for the IT organization. “We’re bringing in science from Discovery Studio to an enterprise-ready product,” says Stevens.  

Other features of the new Discovery Studio include improved tools for automation. “Because of the glut of data and the need for speed and distribution, the whole process of automation becomes more prevalent,” says Stevens. “But that’s only good if you have experimental validation data.”   

An updated Visualizer program now has an ActiveX plug-in  that, coupled with the “storyboard” functionality introduced in the previous iteration of Discovery Studio, allows computational modelers to share interactive 3D visualizations of biomolecules and insert dynamic molecular visualizations into web pages and Microsoft Sharepoint software. The storyboard feature also includes a movie export capability.   

Today’s announcement coincides with Accelrys’ participation at 3DSIG 2011, a structural bioinformatics and computational biophysics meeting in Vienna, held in conjunction with the Intelligent Systems for Molecular Biology conference.   

“We’ve already had very positive response from key customers,” says Stevens.