Decoding malaria

The fight against malaria could be given a boost by a computer programme that claims to identify strands of the virus by matching them against known species.

The software, developed by researchers at the Machine Learning Group of Cambridge University’s Department of Engineering, uses a ‘comparative gene finding’ technique to help map the way in which the disease works and changes in different environments.

Dr Karsten Borgwardt explained: ‘We want to be able to give a label to every part of the genome sequence as we move along it. Finding the most likely label for each part of the sequence is basically a mathematical problem — it depends on probability.

‘Machine learning and statistics can help scientists to label the sequence and establish how the parasite works. Because not every parasite is the same, they can then spot the genes that are only present in one particular malaria species.’

Work by the team has already been significant in decoding the genetic make-up of a malaria parasite known as Plasmodium knowlesi. Previously thought to be the cause of malaria in monkeys, the strain is currently emerging as the fifth malaria parasite affecting humans.

Speaking of the future of the software, Borgwardt said: ‘Our algorithm has to be able to adapt to each genome you want to decipher. New mathematical questions are being generated all the time as more comparative gene finding gets underway. By answering them, we may ultimately be able to help to solve a whole range of biological problems.’