Similarly, Stanford has also made some exciting progress with regards to Huntington’s disease, and this year the Journal of Molecular Biology published a paper from Stanford called ‘The Predicted Structure of the Headpiece of the Huntingtin Protein and Its Implications on Huntingtin Aggregation,’ which again showed the potential for collaboration between computational and experimental labs.
Pande explained that Judith Frydman’s experimental lab, also located at Stanford, had a new insight into a feature of Huntington’s called the PolyQ repeat. "The Q is a letter that signifies a particular amino acid – glutamine," says Pande. "In Huntington’s disease, people have stretches of 35 or 37 of these Qs, and the longer the stretch, the more likely you are to get Huntington’s at an earlier age".
However, Pande says that it was Frydman’s "insight to look into the protein that comes right before the PolyQ repeat’ and ‘found that the aggregation properties were very different". This is where Folding@home came in. "She had some hypotheses and experimental data to suggest why," says Pande, "but the simulations yielded more structural data and insight in terms of the nature of the part that comes before the PolyQ part – the so-called N17 part."
At this point, it’s worth pointing out that computational research isn’t intended to replace experimental protein research. Experimental tests are still needed for validation and accuracy, among many other factors. However, Gruebele points out that computational research also provides additional benefits over experimental research, such as detailed pictures of the proteins that can’t be seen in experimental tests, and Gruebele is confident about their accuracy.
He explains, "There were plenty of people who looked at the 2002 paper and asked, 'How can we really be sure that the protein is doing what those pictures show?' The paper has a figure with pictures of several of those pieces of protein confirmation data, which Pande plotted. However, Pande has performed this multiple times now, and with other people’s data. There’s now no reason not to trust what the experiments can’t see, but you can see on the computer. When Vijay has a paper that says these are the seven structures or clusters that the protein populates that tend to be important for the folding process, that’s basically what’s really going on."
In less than a decade, Folding@home has been through its teething phase as a small start-up project that needed to be proved and tested, and has now moved on to be useful for medical science. "That’s the part I’m probably most excited about," says Pande, "because the work that Grubele and Dyer perform is very important for validation of the general technique, but our real goal is to make an impact on something like Alzheimer’s."
What does energy mean in the context of protein folding? Professor Scheraga explains that "in physics, or in the natural world, systems like to achieve the lowest free energy. In other words, if you’re holding a ball, and you let it go, it will go down rather than up, as it’s attracted by gravity and it lowers its energy. It has high energy when it’s in your hand, but as soon as you drop it, it goes down and lowers its energy".
David Baker (left in the picture), the principal investigator of the Rosetta@home project, explains that "the structure to which a protein folds is the lowest energy structure for an amino acid sequence. Rosetta@home tries to find, given the amino acid sequence of a protein, its lowest energy state".
As well as having a standard screensaver that shows the protein folding in action, Rosetta@home also has a game called Foldit. The game is a part of a project to determine whether people’s pattern-recognition abilities could potentially make them better at simulating protein folding than a computer. "So far, we have preliminary results that show there are classes of problems where people can improve on computers," says Baker, "but the jury is still out."
Like Folding@home, Rosetta@home is actively looking to find solutions for a number of diseases, including anthrax and malaria, but one of their more interesting projects is a vaccine for HIV. "We’re designing proteins that mimic the critical parts of the virus," Baker tells us. "The reason why AIDS is such a nasty disease, among other things, is that the virus can change very rapidly. So you get infected by the virus, and your body tries to fight it off, making antibodies against it, but it changes very rapidly, so your antibodies don’t recognise the altered virus and it can keep evading your immune system."
Baker thinks that you could potentially get around this by designing new proteins. "You might think of the HIV virus as being very hard for the immune system to deal with," says Baker, "but it has an Achilles heel, which is that there are a couple of regions on the virus that can’t change. The region where it gets into cells, for example, just can’t change. So what we’re doing is designing proteins that look exactly like these Achilles heel regions of the virus. Normally, when a virus gets into you, you don’t make antibodies against these Achilles heel regions because they’re hidden away – they just aren’t very exposed. However, our idea is to prime the immune system by making these mimics, which look exactly like these critical regions, so that you already have antibodies against them. When you get infected by the virus, they’ll knock it out."
After running their tests via a distributed computing system, Baker says that the Rosetta@home team then attempt to "design proteins that look just like these critical regions of the virus". After that, the next step is to "actually make the proteins in a laboratory and verify that they look just like those critical regions of the virus". This is performed experimentally.
Pande explained that Judith Frydman’s experimental lab, also located at Stanford, had a new insight into a feature of Huntington’s called the PolyQ repeat. "The Q is a letter that signifies a particular amino acid – glutamine," says Pande. "In Huntington’s disease, people have stretches of 35 or 37 of these Qs, and the longer the stretch, the more likely you are to get Huntington’s at an earlier age".
However, Pande says that it was Frydman’s "insight to look into the protein that comes right before the PolyQ repeat’ and ‘found that the aggregation properties were very different". This is where Folding@home came in. "She had some hypotheses and experimental data to suggest why," says Pande, "but the simulations yielded more structural data and insight in terms of the nature of the part that comes before the PolyQ part – the so-called N17 part."
At this point, it’s worth pointing out that computational research isn’t intended to replace experimental protein research. Experimental tests are still needed for validation and accuracy, among many other factors. However, Gruebele points out that computational research also provides additional benefits over experimental research, such as detailed pictures of the proteins that can’t be seen in experimental tests, and Gruebele is confident about their accuracy.
He explains, "There were plenty of people who looked at the 2002 paper and asked, 'How can we really be sure that the protein is doing what those pictures show?' The paper has a figure with pictures of several of those pieces of protein confirmation data, which Pande plotted. However, Pande has performed this multiple times now, and with other people’s data. There’s now no reason not to trust what the experiments can’t see, but you can see on the computer. When Vijay has a paper that says these are the seven structures or clusters that the protein populates that tend to be important for the folding process, that’s basically what’s really going on."
In less than a decade, Folding@home has been through its teething phase as a small start-up project that needed to be proved and tested, and has now moved on to be useful for medical science. "That’s the part I’m probably most excited about," says Pande, "because the work that Grubele and Dyer perform is very important for validation of the general technique, but our real goal is to make an impact on something like Alzheimer’s."
Rosetta@home
It’s important to note that Folding@home isn’t the only distributed computing project looking at protein folding for medical science. Another similar project is the University of Washington’s Rosetta@home, which has a different goal but equally exciting medical applications. While Folding@home is concerned with simulating the physical process of protein folding at the atomic level, Rosetta@home’s purpose is to predict the structure of naturally occurring proteins, by trying to find their lowest energy state, and then attempting to design new proteins.What does energy mean in the context of protein folding? Professor Scheraga explains that "in physics, or in the natural world, systems like to achieve the lowest free energy. In other words, if you’re holding a ball, and you let it go, it will go down rather than up, as it’s attracted by gravity and it lowers its energy. It has high energy when it’s in your hand, but as soon as you drop it, it goes down and lowers its energy".
David Baker (left in the picture), the principal investigator of the Rosetta@home project, explains that "the structure to which a protein folds is the lowest energy structure for an amino acid sequence. Rosetta@home tries to find, given the amino acid sequence of a protein, its lowest energy state".
As well as having a standard screensaver that shows the protein folding in action, Rosetta@home also has a game called Foldit. The game is a part of a project to determine whether people’s pattern-recognition abilities could potentially make them better at simulating protein folding than a computer. "So far, we have preliminary results that show there are classes of problems where people can improve on computers," says Baker, "but the jury is still out."
Like Folding@home, Rosetta@home is actively looking to find solutions for a number of diseases, including anthrax and malaria, but one of their more interesting projects is a vaccine for HIV. "We’re designing proteins that mimic the critical parts of the virus," Baker tells us. "The reason why AIDS is such a nasty disease, among other things, is that the virus can change very rapidly. So you get infected by the virus, and your body tries to fight it off, making antibodies against it, but it changes very rapidly, so your antibodies don’t recognise the altered virus and it can keep evading your immune system."
Baker thinks that you could potentially get around this by designing new proteins. "You might think of the HIV virus as being very hard for the immune system to deal with," says Baker, "but it has an Achilles heel, which is that there are a couple of regions on the virus that can’t change. The region where it gets into cells, for example, just can’t change. So what we’re doing is designing proteins that look exactly like these Achilles heel regions of the virus. Normally, when a virus gets into you, you don’t make antibodies against these Achilles heel regions because they’re hidden away – they just aren’t very exposed. However, our idea is to prime the immune system by making these mimics, which look exactly like these critical regions, so that you already have antibodies against them. When you get infected by the virus, they’ll knock it out."
After running their tests via a distributed computing system, Baker says that the Rosetta@home team then attempt to "design proteins that look just like these critical regions of the virus". After that, the next step is to "actually make the proteins in a laboratory and verify that they look just like those critical regions of the virus". This is performed experimentally.
RELATED ARTICLES
MSI MPG Velox 100R Chassis Review
October 14 2021 | 15:04
Want to comment? Please log in.