Wednesday , November 13 2019
Home / canada / Investigation: XSEDE distribution on Stampede2 and Comet protein oligomers simulation speed – (Details)

Investigation: XSEDE distribution on Stampede2 and Comet protein oligomers simulation speed – (Details)


Using the proteins derived from the jellyfish, scientists collected sixteen complex protein structure consisting of two octamers set pressurization alone. This study can be applied to useful technologies, such as pharmaceuticals targeted collection of artificial energy, "smart" sensing and building materials, and more. Numerical simulations using XSEDE distributions on Stampede2 (TACC) and (Comet) SDSC refined measurement structure.

Erythrocytes strange. They take oxygen from our lungs and carry it throughout our body to keep us alive. A hemoglobin molecule in red blood cells to carry oxygen, changing its shape completely or nothing fashion. Four copies of the same protein hemoglobin open and close as flower petals, structurally related to react with each other. Using supercomputers, scientists are just beginning to develop proteins that self-assemble on the & # 39; to unite and resemble life-giving molecules such as hemoglobin. The scientists say that their methods can be applied to useful technologies, such as pharmaceuticals targeted collection of artificial energy, "smart" sensing and building materials, and more.

The science team has made the job of protein boost, which means that they have changed the subunits of proteins, amino acids, proteins, to give an artificially high positive or negative charge. The use of protein derived from jellyfish, the researchers were able to assemble a complex structure sixteen protein consisting of two stacked octamers boost one, the results of which were reported in the journal in January 2019 Nature Chemistry.

Next, the team used supercomputer simulations to check and report the experimental results. Supercomputer provision for Stampede2 at the Texas Advanced Computing Center (TACC) and comets in Supercomputer & # 39; yutarnaga San Diego (SDSC) were awarded to researchers through XSEDE, the Extreme Science and Discovery environmental technology, funded by the National Science Foundation (NSF).

"We found that taking the proteins that do not normally interact with each other, we can make the copies that are either very positive or very negative charge," said study co-author Anna Simon, a postdoctoral researcher at the Ellington Lab UT Austin, «The combination of highly positive and negative copies, we can make the protein collected in a very specific structural components, "Simon said. Scientists call their strategy of "boost protein assembly," where they drive a specific protein interaction by about & # 39; unity designed supercharged versions.

"We operated very well known and the basic principle of the nature of that opposite charges attract," said study co-author Jens Glaser. Glaser from the & # 39 is a researcher scientist Glotzer Group, Department of Chemical Engineering at the University of Michigan. "Group Anna Simon discovered that if they mix these charged variants of green fluorescent protein, they are highly ordered structures. It was a real surprise, "said Glaser.

Composed octamers structure looks like a braided ring. It consists of 16 protein – two interlocking rings of the eight that interact in very specific, discreet patch. "The reason why it is so difficult to design proteins that interact synthetic, making these sites and interact with them all lined up right so that they allow the proteins to assemble into larger, regular patterns really hard," explained Simon. They got around the problem by adding a lot of positive and negative charges engineer variants of green fluorescent protein (GFP), a well-studied "lab mouse" protein derived from the jellyfish Aequorea Victoria.

The positively charged protein, which they called blue fluorescent protein (Ceru) +32, have additional opportunities for interaction with negatively charged protein GFP -17. "Giving these proteins all of these features, these different places where they can potentially interact, they were able to choose the right," Simon said. "There were certain patterns and interactions that were there, available, and energetically favorable that we do not necessarily predict that will allow them to gather in these specific forms."

In order to get charged engineered fluorescent proteins Simon et al Arti Pothukuchy, Jimmy Gollihar and Barrett Morrow coded their genes, including the chemical tag, which is used for purification on portable pieces of DNA called a plasmid in E.coli, was then collected in that said protein E.coli, grew. Scientists mixed with proteins. They initially thought that the proteins can interact only with the formation of large, irregularly structured aggregates. "But then we went to see was this weird, funny peak around 12 nm, which was much less than the big lump of protein, but much more than one protein," Simon said.

They measured the particle size generated using a Zetasizer instrument in Texas Materials UT Austin Institute, and verified that the particles contained both Cerulean and GFP proteins Förster resonance energy transfer (FRET), which measures the energy transfer between different colored fluorescent proteins produce fluorescence in response to different energy of light, to see if they are close to each other. Negative stain electron microscopy identifed specific particle structure, the Group make David Taylor, associate professor of Molecular Biosciences at UT Austin. He showed that the particles consisted of 12 IA of the folded octamer consisting of sixteen proteins. "We found that they had these beautiful flower shape-like structures," Simon said. Co-author of Yi Zhou Taylor Group UT Austin increasing the resolution even further by using cryo-electron microscopy to reveal the details of the atomic level foregoing octamer.

Numerical modeling measurement clarified how proteins have been organized in a clear picture beautiful flower-like structure, according to Jens Glaser. "We had to come up with a model that has been quite difficult to describe the physics of charged green fluorescent proteins and to provide all necessary details of atomistic, but it was effective enough to allow us to simulate realistic time scale. With a fully atomistic model, it would take us a year to get a simulation with computer, but the computer was fast, "said Glaser.

They have simplified the model by reducing the resolution without sacrificing important details of the interactions between proteins. "That's why we used a model in which the form of the protein molecular surface accurately represented, as well as one that is measured from the crystallographic structure of the protein," added Glaser.

"What really helped us change the situation and improve on what we were able to get out of our cryo-EM simulation were given," said Vyas Ramasubramani, a graduate student in chemical engineering at the University of Michigan. "This is something that has really helped us to find the optimum configuration to put these simulations, which then helped us to check the stability of the arguments that we did, and we hope to move forward to make predictions about what we can destabilize or change this structure, "said Ramasubramani.

Scientists need a lot of computing power to do the calculations on the scale of what they wanted.

"We used XSEDE basically take these huge systems where you have lots of different parts that interact with each other, and it all out at once, so that when you start to move the system forward through some sort of time you can get an idea of for how it will evolve into several realistic time frame, "said Ramasubramani. "If you tried to do the same kind of modeling we have done on a laptop, it would have taken months, if not years, to really get close to understanding, not some sort of structure will be stable. For us, not being able to use XSEDE, where you could use basically 48 cores, 48 ​​computer units all at once, to make these calculations very parallel, we would have done it much more slowly. "

Stampede2 supercomputer TACC contains 4,200 Intel Knights Landing 1736 and Intel Skylake X computing nodes. Each node has Skylake cores 48, the main unit 39 Comp & #; computer processor. «Skylake nodes with supercomputer Stampede2 played an important role in achieving the efficiency, it was necessary for the calculation of these electrostatic interactions acting between the oppositely charged protein-effective manner," said Glaser. "Having a supercomputer Stampede2 was the right time for us to perform these simulations."

Initially, a team of scientists tested their simulation in the Comet system at SDSC. "When we were the first to find out which model to use, and whether this simplified model gives us good results, the Comet was a great place to try these simulations," said Ramasubramani. "The comet was a great break-in for what we do."

Looking at the broader picture of the scientific, scientists hope that this work is progressing understanding of why so many proteins in nature will oligomerization, or & # 39; combine to form a more complex and interesting design.

"We showed that we should not be very specific, pre-marked set of plans and interactions for these structures to form," said Simon. "This is important, because it means that, perhaps, it is likely that we can take other sets of molecules that we want to do oligomerization and generate both positively charged and negatively charged variants of the & # 39; to connect them, and specially ordered structure for them. "

Natural biomaterials, bones, feathers, and shells can be tough but easy. "We believe that overwhelmed the protein assembly to & # 39 is a simpler way to develop the kind of materials that have interesting synthetic properties without having to spend so much time or having to know exactly how they are going to advance to come together," said Simon. "We believe that the opportunity to accelerate the design and synthetic materials for the discovery and study of nanostructured materials of protein."

Source link