A Chinese University researcher said a new computing model could help the authorities make better choices in selecting the strain of flu they vaccinate people against.

Health officials have to choose which vaccine to order ahead of each flu season.

Maggie Wang, who developed the new model, known as beth-1, with colleagues from the Faculty of Medicine, said it could predict how the virus would change in the future.

She said the model would give authorities early warnings of emerging virus variants, particularly for the H3N2 virus, the H1N1 virus and the influenza B virus.

"This information is particularly important for the WHO (World Health Organisation) because each year or each season, the next season's influenza vaccine strain should be determined earlier, that is almost nine to 12 months ahead of the target season. A prediction model can provide critical information to inform the decision makers of the optimal vaccine strain to select," she said.

"For pharmaceutical companies, the forecasting model is also very useful because the manufacturing of the influenza vaccine has a really compact timeline. So earlier prediction can give more time and the buffer for the manufacturers to prepare the vaccine development and manufacturing."

Wang added that their research team has already been working closely with institutions in the mainland to conduct animal experiments for manufacturing more effective vaccines using the beth-1 model.

"We did the prospective validation using mice model. That is, we predict the next season, the future season's representative viral strain, and then use this strain to make the vaccine to immunize the mice, and we also immunize another group of mice using the current vaccines," she said.

"And after a few months, when the target season comes, we collect the circulating viruses from the Prince of Wales Hospital, and then use this circulating virus to challenge the immunized mice, and compare how strongly antibody response in these two groups of mice. The outcome is that our model-predicted strain generates significantly higher, or non-inferior neutralization titer, antibody response against the circulating viruses."