Auckland man Renton Innes has a knack for folding proteins. Since the 38-year-old cut the time spent on another of his pastimes, Thai kick-boxing, he has been sitting at his PC for up to 18 hours a day pulling strands of amino acids this way and that.

He is no molecular biologist - his highest science qualification is School Certificate - yet his ultimate aim is to save lives by manipulating the building blocks of life.

Innes is one of thousands of people around the world who play Foldit, an online game designed to help scientists predict protein structures. In teams or individually, they compete to solve puzzles posed for them by researchers at Baker Lab at the University of Washington in Seattle.

The game was created by David Baker, Baker Lab's principal investigator and professor of biochemistry at the university, who is an occasional player. "I think you have to have strong three-dimensional visualisation skills and mine aren't terrifically good," he says.

Innes, however, seems to have a natural bent for protein folding. Combined with the competitiveness and cunning that helped him survive numerous kick-boxing bouts without visible scars, it has made him one of Foldit's top performers.

Playing under the names Aotearoa and Renton since late 2008, he consistently ranks in the top couple of dozen players. The spur to his involvement was the cancer death of his mother.

"When she died of a brain tumour I was left thinking what can I do to try to fix this cancer." His online research into the disease gave him a crash course in proteins, and led him to the website.

"I came across this game with which you could win a Nobel Prize for finding cures for cancer and other diseases like Aids."

All of life is dependent on proteins - organic compounds made of amino acids in a genetically coded sequence. The 20 different amino acids form thousands of proteins that play a part in everything from cell structure to metabolic processes, but also diseases like Alzheimer's, cancer and the common cold.

Simple proteins may have 100 or so amino acids and large ones 1000 or more. They fold themselves into three-dimensional shapes according to the most stable bonds that can be formed between the atoms of adjacent amino acids.

The final shape could be one of countless possibilities, but it will determine protein's biological role.

Not that Innes knew any of that before discovering Foldit. "I just started communicating with scientists from around the world who had also joined up."

He cut his teeth on a protein from scorpion toxin, learning how to find the most stable shape by pulling the worm-like structure in different directions, knowing he was on the right track as it turned green on his computer screen, indicating maximum stability.

Innes has since gone on to predict the shapes of about 250 proteins and is familiar enough with particular amino acid sequences to be able to take a stab at what a protein does.

What do the researchers hope to discover? In the first instance, says Baker, they want to learn from the game players' techniques to improve software used for predicting protein structures.

"But more recently we've been trying to solve real-world problems using Foldit. We're trying to design new proteins that do all kinds of things from blocking the flu virus to producing new answers to energy problems."

The US Department of Energy clearly sees their efforts as more than a game, providing Baker Lab with funding for creation of proteins that might act as catalysts in biofuel production.

"We don't have any results yet in terms of a Foldit player curing the common cold or finding a new way to fix carbon dioxide, but that's what we hope will happen," Baker says.

There has been one recent possible breakthrough.

A Russian player, Vakobo, designed a protein that looks promising as an H1N1 flu virus blocker. Innes says a virtual cheer went up among players at Vakobo's discovery, but Baker is more cautious.

"It's a very long road from the design of a protein that binds to H1N1, for example, to an actual medicine you can buy."

Turning a computer-designed protein into the real thing requires synthesising a gene which is then inserted into a bacterium that acts as a factory for production of the protein.

"It's great that people all around the world can work together to try to solve problems that affect everybody. I'm a big believer in the power of the creativity of human thinking," Baker says.

Innes is too. "This game gives back real results. The biggest buzz I could get from this ... would be to save one life."

Anthony Doesburg is an Auckland technology journalist