IBM uses serious science to work with the world's most complex system: the human body
Genetic regulator motifs. Mining and modeling of biological networks. Structural studies of rhodopsin. It's all in a day's work at the IBM Computational Biology Centre (CBC).
The CBC was formed in 1995-a small group that was engaged in exploratory research. But it quickly became vital to IBM's business in healthcare, life sciences and high-performance ("deep") computing. Today, the 40 full-time researchers have extensive backgrounds in computer science and mathematics. Their projects might sound academic in nature, but they tackle real-world problems such as the H1N1 virus and paving the way to personalised medicine.
In 1918, an estimated 50 million people died in a worldwide pandemic. Fifteen years later, the virus behind it was finally identified as influenza.
In the spring of 2009, a new influenza strain was isolated from humans in Mexico. It spread so quickly that the World Health Organisation identified this new flu strain (known as H1N1) as the cause of a worldwide pandemic by that June. This was the first officially declared flu pandemic in 41 years.
Influenza is a formidable opponent, with mutations that occur gradually over time as well as erupt in sudden new strains. The U.S. Centres for Disease Control and Prevention (CDC) has indicated that the H1N1 swine flu has an RNA genome that contains five RNA strands derived from various swine flu strains, two RNA strands from bird flu strains, and only one RNA strand from human flu strains.
So how do you combat such a complex, constantly changing opponent?
The challenge is not new to IBM. In 2006, IBM, the World Health Organisation and the CDC formed the Global Pandemic Initiative, a steering committee with a mission to explore the use of advanced analytical and computer technology as part of a global preparedness program for responding to potential infectious disease outbreaks around the world.
Today, IBM scientists are looking at H1N1 to create models of the virus's genetic and protein sequences and use that to predict the mutations the virus is likely to make. They want to answer questions such as: How might the virus bind to receptors and infect people? And for each of the possible mutations, will the antibody generated by previous exposure or vaccines recognise the new virus and combat it? By modeling the virus's behaviour, scientists hope to determine whether that mutation would be a new strain that escapes all previous vaccines and exposures-and therefore would be a new threat.
Through computational biology, IBM researchers hope to answer that question. And apply that same methodology and expertise to a variety of other situations, including HIV, bacterial infections and resistance to antibiotics.
The human genome is the genetic blueprint for life. More than mere indicators of physical attributes such as eye colour, these microscopic strands of millions of nucleotides also contain clues to which challenges the body will face on its journey, such as Alzheimer's, cancer or heart disease.
Interpreting this blueprint could change modern medicine. With genetic data in hand, doctors could create personalised treatments for patients and recommend changes in diet and other behaviours. But these days, sequencing one individual genome is an expensive undertaking, costing hundreds of thousands of dollars. That's why inexpensive DNA sequencing is a holy grail for geneticists and advocates of personalised medicine.
IBM has embarked on its own search for a technology that is capable of making personalised genome sequencing more affordable-from hundreds of thousands of dollars to just about $1,000. And because IBM is a technology company, it made perfect sense to start with a microchip.
A team of IBM scientists from the fields of nanofabrication, microelectronics, physics and biology have created a new chip that could become the basis for faster inexpensive, personal genetic analysis. The "DNA transistor" is built by drilling tiny nanometer-size holes-about 100,000 times smaller than the width of a human hair-into silicon chips, then passing DNA strands through them to eventually read the information contained in their genetic code.
This miniscule opening is called the nanopore, and critical to its success is the ability to control the speed at which DNA moves through the nanopore. IBM researchers are developing a method to essentially ratchet the DNA molecule using gates of electricity, progressing it one base at a time.
By cyclically turning on and off these gate voltages, researchers have shown theoretically and computationally, and expect to be able prove experimentally, the plausibility of moving DNA through the nanopore at a rate of one nucleotide per cycle-a rate that IBM scientists believe would make DNA readable.
What about privacy?
Of course, when you know your predisposition to certain conditions, there is a risk that someone will hold it against you. An employer. A health insurer. A hospital.
IBM saw this coming. In the fall of 2005, IBM revised its own corporate privacy and equal-opportunity policies to reflect the corporation's intention to handle information about an employee's genetics with a high regard for its privacy, and also to refrain from using genetic test information to discriminate against a person in the employment context. Three years later, the United States signed into law the Genetic Information Nondiscrimination Act (GINA) that protects Americans against discrimination based on their genetic information when it comes to health insurance and employment.
Few things are as clear-cut as a candy bar. But there is a lot of science behind something so simple and sweet.
Over the past several years, the cocoa industry has been hit with a series of destructive fungal diseases that have cost the world's growers an estimated US$700 million in losses every year. IBM Research, the US Department of Agriculture and Mars, Incorporated (US) are teaming up and going straight to the source. Researchers are using IBM's computational biology technology and expertise to develop a detailed genetic map, identifying the specific genetic traits that produce higher cocoa plant yields and resist drought or pests.
But like any sweet treat, the results of this research will be better when shared. Mars will make the genome information available for free through the Public Intellectual Property Resource for Agriculture (PIPRA), which supports agricultural innovation for both humanitarian and small-scale commercial purposes.
Here to help
Get the answers you need
Visit a smarter planet
Instrumented. Interconnected. Intelligent. Inevitable.