|
Nov. 24, 2004 — If you’ve ever used a desktop publishing program, then you know the concept of “drill down.” You start at the highest level — the view of an entire page — and you drill down into individual parts to work on them. Double-click the photo to edit it, for example.
Molecular modeling works the same way — sort of. Trouble is, the laws of physics change when you move from software models of a few molecules to the macroscale schematic of the biosensor that uses it. Designers are forced to use different and unintegrated software tools.
|
Now one software vendor is aiming to change the situation. Accelrys Inc. (Nasdaq: ACCL), a San Diego provider of molecular modeling software, announced the launch of its nanotechnology consortium.
Launched in the summer of 2004, the consortium’s broad goal is to accelerate the development of software for designing nanomaterials and nanodevices. More specifically, it wants to build software that can model systems with tens of thousands of atoms — one of a number of features that would provide the missing link between atomic scale and macroscale design.
The first Nanotechnology consortium meeting took place in early October in Manhattan. The three-day event brought together scientists with an interest in driving forward the development and application of software tools.
“The whole notion is to engage a selection of people embroiled in trying to use a new technology,” said Scott Kahn, Accelrys’ chief science officer. “If you allow that group to constantly review and prioritize what needs to be developed…it’s an extremely efficient way to make advances.”
Charter members of the consortium include Corning Inc., Fujitsu, e2v Technologies, Imperial College, London, and Uppsala University, Sweden. Kahn expects to ultimately attract anywhere from 20 to 40 companies. Each company pays a fee to participate and gain access to the software that is developed.
The consortium is run in three-year segments. After each segment, a review is performed and a decision is made on whether to continue.
The explicit goals for the first segment are, first, to create a broader model library of nanostructures and expand the number of structural features that can be modified. The second goal is to create new algorithms that can handle models featuring tens of thousands of atoms.
The second goal is of particular interest to Richard Gilbert, the principal scientist working on biosensors at e2v, one of the consortium participants. With the current software, he said, computation time increases exponentially as more atoms are added to a system. When it comes to modeling large nanoscale systems, it would take months or a year to do the computation. The idea is to make the time-required curve increase in a linear rather than exponential fashion as more atoms are added to a system.
Gilbert’s group is building a proteomic lab-on-a-chip device intended to help pharmaceutical companies develop new drugs and provide medical personnel with on-site, real-time clinical diagnostics.
It’s an area, Gilbert said, “where engineering and computational chemistry meet.” In other words, in order to rationally design the microfluidic structures that perform the chemistry, his team needs an atomic scale model of the fluids within.
By participating in the consortium, so the theory goes, Gilbert’s team gets one of the first cracks at using the new tools that are developed, while Accelrys gets a new feature set it can offer to other customers at a later date.
“This is a very interesting and well thought through business model for Accelrys,” said Gregg Zank, the chief technology officer and executive director of science and technology at Dow Corning Corp. Dow Corning declined to participate because the consortium’s goals “don’t appear on our critical path.”
But, Zank said, the Accelrys consortium model is likely to succeed in advancing the tools’ capabilities in response to customer needs and in a way that shares the cost across industries.
The challenge for Accelrys, he said, would be to make sure individual customer requirements don’t pull the work too far in any one direction. He said a diverse group of participants spanning materials and device companies and academia — such as the one Accelrys has assembled — would help it ascertain what features are best suited for eventual inclusion in its commercial software.