SYDNEY — A technique of designing molecules to produce a particular electromagnetic frequency, pioneered by an Australian electrical and biomedical engineer, has resulted in a patent on one potential drug treatment and a patent application on another.

Irena Cosic, deputy head of the electrical and computer systems engineering department at Monash University, in Melbourne, used this approach to design a potential vaccine for HIV, which has since been constructed and tested at the Universite de Lyon, in France.

Otherwise completely unrelated to HIV, the molecule has proved sufficiently promising in stimulating the immune system into recognizing HIV to be patented, with the patent shared between the French government and Monash University.

The Serbian-born Cosic, who came to Australia in 1989, told BioWorld International she had designed another set of molecules that is now the subject of a patent application.

However, beyond saying the molecules were involved in the process of purifying other molecules, she declined to give any further details until the patent application is finalized.

The first step in Cosic's technique involves finding estimates of the energy of the most loosely bound electrons in each of the 20 naturally occurring amino acids that make up proteins. Those energies are then graphed against the spot each amino acid has in the protein chain — that is, electron energy is graphed against distance along the protein chain.

All she got initially was a squiggle, but believing some sort of order lay underneath the mess, Cosic analyzed it as if it were a radio signal.

Using a standard technique known as a Fourier Transform that breaks down a messy electromagnetic wave form into a series of standard wave forms, she found that each group of molecules — such as hemoglobins, lysozymes, mioglobins and fibroblast growth factors — had its own characteristic "frequency."

She also found proteins that interacted, such as receptors and the matching binding molecules, had the same characteristic "frequency."

Did this theoretical exercise have a basis in reality? Cosic calculated the energies involved in the movement of electrons along the chains of amino acids, as such movements produce electromagnetic waves, and found they were typical of visible light. She then calculated the frequencies for proteins known to be sensitive to light and found her calculations matched the proteins' known, sensitive frequencies.

To be any good the theory has to make predictions that are fulfilled, so Cosic designed an amino acid sequence that matched the frequency of the HIV molecule but was unrelated to HIV. As noted, that design proved promising enough to be patented.

Another design copying the frequency of a fibroblast growth factor, a protein which stimulates the growth of solid tumors, showed some initial promise but has now been abandoned.

Cosic is still refining her technique, and sees her characteristic frequency approach as "another level" of molecular recognition on top of the traditional lock-and-key theory of binding sites on each molecule fitting together.

She believes the molecules are recognizing each other on the basis of their frequencies, "like tuning in to a radio station," in addition to the less precise recognition of binding sites.

A number of scientists still remain skeptical of the approach, but her technique is winning recognition, Cosic said.