Unfolding the mystery of proteins

Share this post:

U.S. Patent #8,423,339: Visual analysis of a protein folding process
Patent Issued: April 16, 2013
Inventors: Laxmi Parida, Ruhong Zhou

What is patent #8,423,339? By Laxmi Parida, Manager, IBM Computational Genomics Group.

Proteins do just about everything in our cells. And those functions, from hormones transport to muscle contraction and much more, are determined by the order and arrangement of protein building blocks, amino acids. Patent #8,423,339 gives scientists a way to break down that folding process – which happens in the fraction of a blink of an eye – to understand why proteins fold (and mis-fold) into certain shapes.

In nature, amino acids in a cell bind together and form a specifically shaped protein. How does this happen? We’re still trying to figure that out. It even has its own paradox, Levinthal’s Paradox, stating that the folding can’t be random (even though it seems that way) because it’s too complicated.

We know what proteins are made of, but since we can’t see how proteins fold in nature, we have to simulate them. And we’ve been simulating protein folding since before Blue Gene/L in 1999. Our patent uses the data from these simulations to find and understand the paths the amino acids take during the folding process.

The tough thing about finding patterns in the process of protein folding is that everything in the data could potentially be a pattern. So, we have to hunt for informative combinations of values that surface again and again – not necessarily in tandem, in the process of folding.

Visualization of the varying landscape of patterns as a protein folds

These simulations that create 3D models generate massive amounts of data. But our pattern discovery technique works at the level of the simulator, from workstation to supercomputer. It efficiently combs the data to find these invariant combinatorial subsets, which could provide valuable insight into the folding process.

Knowing how proteins fold or why they mis-fold will help us understand how cells work and don’t work. Say, in the case of bovine spongiform encephalopathy (mad cow disease). By understanding the process that created the mis-shapen prion, we could possibly cure or eradicate it, or any other such disease.

IBM has led in total U.S. patents for 21 year in a row in 2013. Read more about patents like #8,423,339, here.

More stories

A new supercomputing-powered weather model may ready us for Exascale

In the U.S. alone, extreme weather caused some 297 deaths and $53.5 billion in economic damage in 2016. Globally, natural disasters caused $175 billion in damage. It’s essential for governments, business and people to receive advance warning of wild weather in order to minimize its impact, yet today the information we get is limited. Current […]

Continue reading

DREAM Challenge results: Can machine learning help improve accuracy in breast cancer screening?

        Breast Cancer is the most common cancer in women. It is estimated that one out of eight women will be diagnosed with breast cancer in their lifetime. The good news is that 99 percent of women whose breast cancer was detected early (stage 1 or 0) survive beyond five years after […]

Continue reading

Computational Neuroscience

New Issue of the IBM Journal of Research and Development   Understanding the brain’s dynamics is of central importance to neuroscience. Our ability to observe, model, and infer from neuroscientific data the principles and mechanisms of brain dynamics determines our ability to understand the brain’s unusual cognitive and behavioral capabilities. Our guest editors, James Kozloski, […]

Continue reading