As in the desktop-computer revolution, the silicon chip is playing a key role in downsizing expensive and arcane technology into low-cost, easy-to-use systems. Progress in that direction is creating some unusual mixes of technology, such as minute RF tags, described at the conference by developer PharmaSeq Inc. (Monmouth Junction, N.J.), that combine optical fluorescence with wireless chip techniques to tag and analyze biochip samples.

Stefano LoPriore, STMicroelectronics' business manager for biochips, touted the company's recent technology partnership with MobiDiag (Helsinki, Finland) to create a workstation-sized infectious-disease analysis system. And Nanosys Inc. CEO Larry Bock gave a keynote at the conference on the significance of nanotechnology to the emerging field of bioarray technology.

ST's polymerase chain-reaction-on-chip system, announced at last year's Chips to Hits conference, is a good example of the type of heterogeneous integration required in this new field. The chip contains microfluidic channels and reaction chambers heated with electronic resistors. DNA samples are amplified on chip in the chambers and piped to a DNA sample array for optical analysis.

"We want to transfer the complexity of large-scale labs onto these chips and use volume manufacturing to reduce cost," LoPriore said. Once it is in volume production, the MobiDiag system will replace lab equipment costing more than $10,000 with a small-scale unit costing only a few thousand, he said.

Equally significant is the short response time of the diagnostic system. Today, a patient with an unspecified infection needs to take a broad-spectrum antibiotic until a diagnosis is obtained that allows a switch to a narrow-spectrum antibiotic. With the new system, the doctor would know immediately which pathogen to target. "This could save billions of dollars per year, just in the cost of antibiotics," LoPriore said.

The initial system is the first step to taking molecular biology out of the lab, but eventually LoPriore would like to see point-of-care systems that could be used in a doctor's office.

A critical point in the development of any microarray system is some means of observing and acquiring data about molecular interactions. Most approaches use optical schemes, since light is precise and has minimal interference with the reactions. But a novel approach described at the conference by Wlodek Mandecki, president of PharmaSeq, uses tiny wireless transponders that are powered by light. The RF silicon chips measure only 250 microns on a side and contain radio circuitry, an antenna, a 64-bit memory and a photocell as power source. Seventy thousand of them can be manufactured on an 8-inch wafer.

The chips are designed so that data about molecules on their surface is broadcast to a data analysis system. Since the system is powered by light, it is also possible to include the usual optical methods, along with the wireless data acquisition, to capture information.

For combinatorial drug experiments, for example, the chip's memory could be programmed with an identification number representing a specific molecule, such as a protein or a fragment of DNA. When a match occurred, the appropriate transponder would broadcast the ID number, and the signal would be picked up and recorded by the reader.

RF techniques also figure in Nanosys Inc.'s plans to develop a large-area electronics technology using some nano-engineered structures. The project kicked off recently with a Defense Advanced Research Projects Agency grant to develop the technology in conjunction with research groups at the University of Texas at Dallas, Penn State University and printed-antenna array specialist Sciperio Inc. (Stillwater, Okla.). The goal of the project is to devise a nano-engineered material that could be easily coated on flexible substrates and that would perform as well as single-crystal silicon.

That capability could be applied widely, which is the design philosophy at Nanosys. Larry Bock, the company's CEO, has compiled a long track record for biotechnology startups and believes that nanotechnology will make fundamental contributions to the biotech revolution.

Bock expects nanomaterials to create high-performance substrates for biological arrays that would aid in the identification of complex biological molecules. Also, engineered biological surfaces could play a role in chemical and biological sensors. Physical effects could be created in nanoparticle tags for molecular recognition.

Another development reported at the conference was a technique for building artificial lipid films on a substrate. Jay Groves, a researcher in the chemistry department at the University of California, Berkeley, described a membrane array technique that is being put to use in drug discovery systems. The films will help molecular biologists understand the critical functions of cell membranes.

Protein identification is showing some progress, following a development path blazed by DNA identification arrays. Protein identification is crucial for monitoring the current state of cells. While DNA can predict a propensity for a certain disease, it is only by monitoring the production of proteins that the actual course of the disease can be tracked.

The Food and Drug Administration is working on a complex tracking technique at its Center for Biologics Evaluation. Researchers have developed protein arrays that can track multiple protein expression pathways simultaneously, a capability considered critical for understanding how certain diseases develop. At the conference, Emanuel Petricoin, co-director of the FDA-NCI (National Cancer Institute) Clinical Proteomics Program, described how he tracked dozens of pathways simultaneously to monitor the progress of molecular-based intervention drugs in treating cancer.