Cloud computing expands brain science

PICTURE: Franco Pestilli, Neuroscientist, Department of Psychology, University of Texas at Austin After

Credit: Franco Pestilli

People often think of human behavior in terms of what’s going on in the present: reading a newspaper, driving a car, or catching a soccer ball. But other dimensions of behavior span weeks, months and years.

Examples include a child learning to read; an athlete is recovering from a concussion; or someone who is 50 and wondering where all the time has gone. These are not changes that people see on a daily basis. They suddenly realize that they are older, healed, or have a new developmental skill.

“The field of neuroscience examines the brain in several ways,” says Franco Pestilli, a neuroscientist at the University of Texas at Austin (UT Austin). “For example, we’re interested in how neurons calculate and allow us to react quickly – it’s a rapid response requiring visual attention and motor control. Understanding the brain needs big data to capture all dimensions of the brain. human behavior.”

As an expert in vision science, neuroinformatics, brain imaging, computational neuroscience, and data science, Pestilli’s research has advanced the understanding of human cognition and brain networks over the past 15 years.

Pestilli likes to compare the brain to the Internet, a powerful collection of computers connected by cables that simultaneously keep many windows open and programs running. If the computer is in good condition but the cables are not, long-distance communication between computers in different parts of the brain begins to fail. This in turn creates problems for our behavior in the long run.

Pestilli and his team are also interested in how biological calculations change over longer periods of time, such as how our brains change when we lose our vision?

“We have shown that if you change the entrance to the eye, it can change the white matter of the brain, which is equivalent to the brain’s wiring system – just like computers are connected by cables, our brains have millions cables connecting millions of tiny computers called neurons. “

This research was published in Nature Scientific reports in March 2021. – The Platform Scientists Have To Do The Science They Want

New cloud technologies are becoming necessary to help researchers collaborate, process, visualize and manage large amounts of data at unprecedented scales.

A key aspect of Pestilli’s work began in 2017 when he received a BRAIN Initiative grant through the National Science Foundation (NSF) to launch At that time, he was an associate professor of psychological and brain sciences at Indiana University.

The computing platform provides a full suite of web services to support repeatable cloud searches. More than 1,600 scientists from all over the world have so far accessed the platform. allows them to upload, manage, track, analyze, share and visualize the results of their data.

Currently, the platform serves different communities of scientists, from psychology to medical sciences to neuroscience, and includes more than 600 data processing tools. integrates different expertise and development mechanisms to create code and publish it to the cloud, while tracking every detail that happens to the data.

“We have processed over 300,000 datasets to date – and we are serving many new users as the number of scientists accessing our platform has skyrocketed during the pandemic,” Pestilli said. “Lots of new people came to because they lost access to their physical facilities.”

The platform relies on a supercomputing infrastructure to run simulations on high performance computing (HPC) hardware. “National systems like Jetstream (Indiana University / TACC), Stampede2 (TACC) and Bridges-2 (Pittsburgh Supercomputing Center) are fundamental to what we do. We have received a lot of support from the Extreme Science and Engineering Discovery Environment ( XSEDE) funded by NSF. “ is also funded by collaborative awards from the National Institutes of Health (NIH) and the Department of Defense.

Aina Puce is Professor of Psychology and Brain Sciences at Indiana University. She is a self-proclaimed neophyte when it comes to, but she is a global expert in neuroimaging and the principal investigator of an NIH grant that supports the development of neurophysiological data management and analysis on the platform.

“I took the plunge to help Franco and his team extend the platform’s functionality to neurophysiological data,” Puce said.

“ allows us to start performing cutting-edge analysis, integrating neurophysiological data and MRI-based data. Studies include research that explicitly links brain structure to brain function, such as how information is. transported from one region to another and how the blood flow and electrical activity of the brain changes when performing particular tasks. ”

Soon, a suite of new tools will be available on to allow users to integrate EEG (electroencephalography), MEG (magnetoencephalography) and MRI (magnetic resonance imaging) data, which are “unique and will be extremely useful to both for science and society, ”she said.

Puce and his team are currently exploring brain activity by recording electrical output, both non-invasively from the scalp and invasively from the inside of the head. They also detect magnetic fields produced while a person is at rest and while performing tasks such as reading other people’s social messages.

“This is what we are bringing to for the first time,” Puce said.

Getting to Know Data Drives

The field of neuroscience is evolving from small data sets to large data sets. Bigger data sets mean scientists can extract statistically more powerful information from the information they collect.

From 1,000 subjects to 10,000 subjects to 500,000 subjects, the datasets keep growing.

For example, the Adolescent Brain Cognitive Development Study is one of the largest long-term studies of brain development and child health in the United States. The study collects data on more than 10,000 adolescent brains to understand biological and behavioral development from adolescence to young adulthood. In another part of the world, the UK Biobank contains detailed information on the health of more than 500,000 participants who have donated their genetic and clinical data for the benefit of science; 100,000 of these participants donated brain scans.

“As each new project grows,” said Pestilli, “the size of the dataset also increases, and as a result, the storage and compute needs change. We create datasets. of a size and impact that only supercomputers can With the recent advent of machine learning and artificial intelligence methods, and their potential to help humans understand the brain, we must shift our paradigm to data management, analysis and storage. “

Pestilli says neuroscience research can only survive if a cohesive ecosystem is built that will integrate the needs of scientists with the needs of hardware and software given the huge amount of data and next-generation questions to be explored.

He says many of the tools developed so far aren’t easily integrated into a typical workflow or out of the box.

“To impact neuroscience and connect the discipline to the most advanced technologies such as machine learning and artificial intelligence, the community needs a cohesive infrastructure for cloud computing and data science to bring all these great tools, libraries, data archives, and standards closer to researchers who work for the good of society, ”he said.

Fortunately, Pestilli has found a like-minded contributor who shares this vision in Dan Stanzione, the executive director of the Texas Advanced Computing Center (TACC) and a nationally recognized leader in HPC.

Together, they plan to create a national infrastructure that provides a permanent data registry and analysis records. Researchers will be able to find data and more transparently see the root of how the analysis was conducted. The infrastructure will facilitate what the NSF demands in data proposals, and what researchers want – scientific impact and replicability.

Furthermore, it means that access to data, analytical methods and computing resources will evolve towards a more equitable model, providing opportunities for many more students, educators and researchers than ever before.

“This prospect made me very excited to join the University of Texas at Austin,” said Pestilli. He moved to Austin in August 2020, amid the COVID-19 pandemic. Being at UT Austin means working with TACC – one of the main reasons he accepted a professorship in the psychology department.

“I have no doubts that we can make it happen – this vision is a crucial part of my efforts here.”


Source link

About Mariel Baker

Check Also

Computing at the frontiers of nanomedicine

01 Sep 2021 (Nanowerk News) Viruses kill millions of people around the world every year. …