Microsoft Aims to Apply Spam Study to HIV Researchers at Microsoft are studying similarities between HIV and e-mail spam. The two are similar in one important respect: Both mutate as they spread. Scientists hope they can apply studies of how spam can be stopped to developing an AIDS vaccine.
NPR logo

Microsoft Aims to Apply Spam Study to HIV

  • Download
  • <iframe src="" width="100%" height="290" frameborder="0" scrolling="no" title="NPR embedded audio player">
  • Transcript
Microsoft Aims to Apply Spam Study to HIV

Microsoft Aims to Apply Spam Study to HIV

  • Download
  • <iframe src="" width="100%" height="290" frameborder="0" scrolling="no" title="NPR embedded audio player">
  • Transcript


Researchers haven't yet come up with an AIDS vaccine but they're approaching the challenge from all directions.

David Heckerman and his colleagues at Microsoft were inspired by spam that most of us are just annoyed by those unwanted e-mails. But Doctor Heckerman finds inspiration in them. He says the study of spam may prove useful in developing an AIDS vaccine.

Doctor Heckerman is at the Microsoft campus in Redman, Washington. Welcome.

Dr. DAVID HECKERMAN (Manager, eScience Research Group, Microsoft Research): Thanks, James.

HATTORI: Doctor Heckerman, you're both a physician and a computer scientist. Now we know Microsoft is in the forefront of battling spam but does that have to do with HIV?

Dr. HECKERMAN: Well, it turns out our immune systems are a lot like spam filter, they're constantly watching for bad things such as viruses and other bugs and they usually stop them in their tracks. But just as some clever spammers mutate their e-mails to evade spam filters, HIV mutates itself to evade our immune system.

HATTORI: So there are similarities between how spammers constantly change their tactics to evade these spam blockers and how the AIDS virus changes or mutates when it's in the human body?

Dr. HECKERMAN: Exactly. For example, spammers will replace the letter I in Viagra with the number one, which looks like an I to us but it makes the spam filter miss the message. HIV actually changes itself to avoid the attack of our immune system. In the case of spam, what we tried to do is figure out what spammers are going to do next and prepare our filters for those changes. Similarly with HIV, we're trying to find HIV's escape pass using complex computer algorithms to prepare our immune system for those changes. And the way we prepare our immune system is with the vaccine.

HATTORI: Yeah. So that's the goal of the research being done by you and your colleagues - to eventually come up with something that will lead to a vaccine.

Dr. HECKERMAN: Exactly.

HATTORI: What is it about HIV that makes it so difficult to combat - so difficult to find the vaccine?

Dr. HECKERMAN: In one way, solving the HIV problem is a bit easier than solving the spam problem. In that, these dastardly spammers come up with some very tricky mutations whereas, fortunately for us, HIV is not so clever. However, there are some very tricky problems with dealing with HIV that makes it even harder to deal with than with spam. For example, with spam, we get to watch it mutate on minute-by-minute basis. With HIV, we typically only see a single snap shot of the virus after it's had a long time - maybe a year - to evade the immune system. And to find these hidden paths of HIV mutation, we have to run computer simulations - much like weather simulations - to see which mutational paths best fit the small amount of data that we can see. To run a single analysis would take one year on a single computer. But thanks to the super computers we have at Microsoft, we're able to do that in a fraction of a day.

HATTORI: What else makes it difficult to find this vaccine?

Dr. HECKERMAN: Well, one issue that makes both fighting spam and fighting HIV tough is that they are different in different parts of the world. And what makes the HIV problem even tougher is that each of us has a different immune system, so that HIV mutates differently in each of us.

Because of this variability, we may not be able to build a single vaccine that works for everyone. But perhaps we can build a vaccine that will cover most people in a large region. To do this, we've developed computer algorithms that exploit the commonalities among the different kinds of HIV and the commonalities among our different immune systems.

HATTORI: Will your research be used, eventually, in an actual vaccine trial soon?

Dr. HECKERMAN: Yes. We're putting together some promising vaccine designs based on the new information we've gained from our computer simulations. And the next step will be to put these designs to the test in clinical trials. It'll be years before we have definitive results from those trials, but I'm optimistic.

HATTORI: Dr. David Heckerman, a researcher at Microsoft. Thanks for talking with us.

Dr. HECKERMAN: Thanks, James.

Copyright © 2007 NPR. All rights reserved. Visit our website terms of use and permissions pages at for further information.

NPR transcripts are created on a rush deadline by Verb8tm, Inc., an NPR contractor, and produced using a proprietary transcription process developed with NPR. This text may not be in its final form and may be updated or revised in the future. Accuracy and availability may vary. The authoritative record of NPR’s programming is the audio record.