AI can help generate personalized cancer vaccines in an hour

Now it takes quite long to build personalized vaccines because computing of how a vaccine should look like uses matrix methods, Alexander Gintsburg said.

The use of artificial neural networks could bring down the duration of computing required to create a personalized cancer vaccine, which is currently a lengthy process, to less than an hour, said Director of the Gamaleya National Research Centre for Epidemiology and Microbiology Alexander Gintsburg

“Now it takes quite long to build [personalized vaccines] because computing of how a vaccine, or customized mRNA, should look like uses matrix methods, in mathematical terms.

We have involved the Ivannikov Institute which will rely on AI in doing this math, namely neural network computing where these procedures should take about half an hour to an hour,” said Russia’s vaccine chief Alexander Gintsburg.

Earlier, the researcher explained that an experimental base of sequences of between 40,000 to 50,000 tumours with identification of antigen compatibilities that are converted into protein or RNA in a patient will be needed to train AI.

This would help determine whether or not this combination can be used for the individual, he said.

The vaccine was developed in collaboration with several research centres. Preclinical tests have already been completed, General Director of the Radiology Medical Research Centre of the Russian Ministry of Health Andrey Kaprin has said.