A spam filter for HIV is in the works
This article was originally on a blog post platform and may be missing photos, graphics or links. See About archive blog posts.
HIV: a deadly virus that kills an estimated 5,000 people a day.
Spam: annoying emails that infiltrate your inbox and try to get you to shell out for erectile dysfunction drugs, credit cards and international scams purportedly involving Nigerian princes.
Could these two things possibly have anything in common? According to Microsoft researcher David Heckerman, the answer is yes.
Heckerman is the inventor of Microsoft’s spam filter that protects Hotmail, Outlook and Exchange clients from deluges of unwanted email, but for the last seven years he’s been working on designing a vaccine for HIV.
He said it’s not so strange that he shifted his attention from protecting email systems to protecting body systems. He is a doctor, and besides, fighting spam and fighting HIV are not as different as you might think.
‘We have an adversarial situation going on between spam filters trying to block the spam and the spammers changing and mutating,’ he said in an interview with The Times. ‘And in the case of HIV, we have the immune system fighting the virus and HIV mutating to try to get through.’
Heckerman said the key to fighting spam and HIV is the same: Find the part that absolutely can’t mutate -- what he calls the Achilles’ heel -- and attack there.
‘In the case of spammers, they want to extract money from you. That’s what they can’t avoid. So our spam filters, at least in part, focus on that,’ he said.
Now he and his team are trying to find the Achilles’ heel of the virus that causes AIDS.
‘It mutates a lot, but it can’t mutate to where it stops functioning,’ he said. ‘If it does do that, we win.’
The work now is to find the places where if the virus mutates there, it dies out. To find it, Herckerman and his team are using the needle-in-a-haystack approach -- crunching enormous amounts of data with the help of thousands of computers in order to find clues to what might work.
‘I think it is a solvable problem, but we have a lot of work left to do,’ said Heckerman. ‘But I’m working on this every day, and I’m hopeful.’
Want more information? Microsoft Research has put together this video.
-- Deborah Netburn