NIH brain research initiative to leverage network of data resources
The $250 million effort is one of the largest neuroscience investments the agency has made for a single set of research projects.
The National Institutes of Health is putting up $250 million to create a network of integrated centers, collaborating laboratories and data resources in order to make molecular, anatomical, and functional data about brain cells available to the broader research community.
The big data effort is part of the Brain Research through Advancing Innovative Neurotechnologies (BRAIN) Initiative, a large-scale NIH program to push the boundaries of neuroscience research and equip scientists with insights necessary for treating a wide variety of brain disorders such as Alzheimer’s disease, autism, epilepsy, and schizophrenia.
The multi-year research program will make use of a variety of information technologies, including three-dimensional mapping, advanced database systems and cloud computing to advance clinical understanding of the brain and its functions.
Funding for the project will be on a par with what is being spent to support the much-heralded federal Cancer Moonshot, which is intended to rapidly advance understanding of the causes, prevention and treatment of cancer, and the Precision Medicine Initiative.
Researchers hope the brain initiative will eventually result in findings that will support clinicians who are dealing with patients’ neurological diseases.
Currently, relatively little is known about the different types of cells in the brain and what they do, according to NIH officials. To address this knowledge gap, the BRAIN Initiative Cell Census Network (BICCN) intends to discover and catalog the many cell types in human, monkey and mouse brains, and to share that data with researchers—a critical step toward understanding brain health and disease.
“Before we can fully understand how our brains work, we need to understand how the parts work,” said Francis Collins, MD, director of NIH. “Making molecular, anatomical and functional data about brain cells available to the broad research community will speed our understanding of how cells and circuits are organized, revealing the rules of communication within the world’s most complex known organ.”
BICCN comprises individual projects that will collectively define different brain cells and share the data, according to Andrea Beckel-Mitchener, lead program manager for the effort and acting director of the Office for Research on Disparities and Global Mental Health at NIH’s National Institute of Mental Health.
NIH has funded 11 grants under BICCN, totaling $50 million annually over five years. The $250 million effort is one of the largest neuroscience investments the agency has made for a single set of research projects, Beckel-Mitchener contends.
“We’ve known that there are broad cell types for many years, but what we haven’t really had a handle on are functionally what each of them are doing in a complex brain structure,” she says. “These awards are meant to help understand that better. We’re starting with the mouse because it is a great experimental system, but we also have a few awards that are in humans and one in non-human primate as well to start building our understanding going to a complex mammalian brain. It’s only when we understand what a healthy brain looks like that we can start to understand what a diseased or disordered brain looks like.”
Also See: Plan emerges to deal with deluge of data from brain research
The Allen Institute for Brain Science in Seattle has received three five-year BICCN grants totaling nearly $100 million—one of which is a $65.5 million award to create a comprehensive whole-brain atlas of cell types in the mouse.
“One of the really unique features about this project is something that the Allen Institute has created called the Common Coordinate Framework,” Beckel-Mitchener contends. “It’s basically a 3-D image of—at this point—a mouse brain, but what it allows us to do is register the data to a very specific location in the brain so we can understand a cell type. It’s almost like a kind of Google Earth or Google Maps through which you can place an individual cell in a particular region of the brain and know a lot about that cell. We don’t have a similar capability in human or non-human primate brain. So the mouse is really going to inform how we go forward and create something similar for the human brain, which is our ultimate goal.”
Under a $19.4 million BICCN grant, the Allen Institute will lead the creation of a multimodal atlas of human brain cell types, starting with a gene expression survey of single cells in the brain and spinal cord, as well as deeper analysis of samples from the cortex and hippocampus. Researchers will then analyze human neurosurgical and postmortem tissue samples, and characterize the spatial distribution and relative proportions of cells expressing different sets of genes.
In addition, they will form an international consortium of experts in studying and modeling live human cells, with the goal of classifying cells based on shape, physiology and connective properties, and then correlate these properties to gene expression data.
Each of the BICCN’s individual projects will use tissue specimens to analyze brain cell types. However, there won’t be a central biobank repository, she adds. “It really will come down to the data that are generated,” observes Beckel-Mitchener. “We have one particular grant that’s going to a data center—we’re calling it the BRAIN Cell Data Center. And, they are in charge of coordinating the intake of all of these different data types.”
As part of $14.5 million BICCN grant, the Allen Institute will be responsible for developing and maintaining the BRAIN Cell Data Center that will serve as a community resource for single cell data in the brain. The center will provide a web-portal for sharing single cell brain data, tools and knowledge generated by BICCN research partners, as well as create and implement data models, common community standards, and scientific results.
“The BRAIN Cell Data Center will be an incredibly valuable resource for the entire neuroscience community and beyond,” says Michael Hawrylycz, investigator at the Allen Institute for Brain Science. “Applying cell type semantic and spatial community standards to these data, and making them available online, provides a tremendous opportunity for improving our understanding of the diverse cell types in the brain and how they’re organized.”
According to Beckel-Mitchener, the BRAIN Cell Data Center will collect the data and run it through various quality control measures. “The actual raw data will be stored in a couple of data archives, which are also funded this year,” she adds. “So, we’ll have the raw data available as well as analyzed data from the data center.”
Raw data will be stored in two BRAIN Initiative data archives—one at the University of Maryland Baltimore and the other through a collaboration between Carnegie Mellon University, the Pittsburgh Supercomputing Center and the University of Pittsburgh.
“At the current time, there is no practical way to analyze, mine, share or interact with large (100 terabytes or larger) brain image datasets. To address this issue, this proposal establishes the BRAIN Imaging Archive (or more simply ‘the Archive’) data service in Pittsburgh,” states Carnegie Mellon University’s project summary. “The Archive encompasses the deposition of datasets, the integration of datasets into a searchable web- accessible system, the redistribution of datasets, and a computational enclave to allow researchers to process datasets in-place and share restricted and pre-release datasets. The Archive will, for the first time, provide researchers with a practical way to analyze, mine, share or interact with large image datasets by creating a unique public resource for the BRAIN research community.”
The University of Maryland Baltimore will develop the Neuroscience Multi-Omic Archive (NeMO Archive), a data repository specifically focused on the storage and dissemination of “omic” data from the BRAIN Initiative and related brain research projects.
“We will utilize a federated model for data storage such that the physical location of data can be distributed between the NeMO local file system, public repositories, and a cloud-based storage system (such as Amazon S3),” states UMB’s project abstract. “We will leverage this capability and distribute BRAIN Initiative data between our local file system and the cloud. The Nemo Archive will be a data resource consistent with the principles advanced by research community members who are launching resources in next generation NIH data ecosystem. These practices include FAIR Principles, documentation of APIs, data-indexing systems, workflow sharing, use of shareable software pipelines and storage on cloud-based systems.”
“For some investigators, it’s important to have access to the raw data—that’s how they make new discoveries,” contends Beckel-Mitchener. “They may have new software packages and algorithms to analyze the raw data, which can verify or validate some of the existing outputs, but it can also answer some other questions. For scientific purposes, it’s really important to have both kinds of data available to the research community.”
She notes that the first phase of BICCN is over five years. However, Beckel-Mitchener says NIH is hoping to also have a second phase of the program.
“The vision is to complete the mouse atlas within five years and to really launch the human studies in a much more significant way,” she concludes. “There’s a lot of challenges in doing this with human brains—as you might imagine—given the data complexity, the availability of source material and making sure that we have healthy brain tissue to analyze.”
“We don’t want identifiable (human) information” that come from patients, adds Beckel-Mitchener. “For the human data, it will likely be controlled access,” as opposed to the open access to the mouse data that will be permitted “because there are no concerns in terms of privacy,” she says.
The big data effort is part of the Brain Research through Advancing Innovative Neurotechnologies (BRAIN) Initiative, a large-scale NIH program to push the boundaries of neuroscience research and equip scientists with insights necessary for treating a wide variety of brain disorders such as Alzheimer’s disease, autism, epilepsy, and schizophrenia.
The multi-year research program will make use of a variety of information technologies, including three-dimensional mapping, advanced database systems and cloud computing to advance clinical understanding of the brain and its functions.
Funding for the project will be on a par with what is being spent to support the much-heralded federal Cancer Moonshot, which is intended to rapidly advance understanding of the causes, prevention and treatment of cancer, and the Precision Medicine Initiative.
Researchers hope the brain initiative will eventually result in findings that will support clinicians who are dealing with patients’ neurological diseases.Currently, relatively little is known about the different types of cells in the brain and what they do, according to NIH officials. To address this knowledge gap, the BRAIN Initiative Cell Census Network (BICCN) intends to discover and catalog the many cell types in human, monkey and mouse brains, and to share that data with researchers—a critical step toward understanding brain health and disease.
“Before we can fully understand how our brains work, we need to understand how the parts work,” said Francis Collins, MD, director of NIH. “Making molecular, anatomical and functional data about brain cells available to the broad research community will speed our understanding of how cells and circuits are organized, revealing the rules of communication within the world’s most complex known organ.”
BICCN comprises individual projects that will collectively define different brain cells and share the data, according to Andrea Beckel-Mitchener, lead program manager for the effort and acting director of the Office for Research on Disparities and Global Mental Health at NIH’s National Institute of Mental Health.
NIH has funded 11 grants under BICCN, totaling $50 million annually over five years. The $250 million effort is one of the largest neuroscience investments the agency has made for a single set of research projects, Beckel-Mitchener contends.
“We’ve known that there are broad cell types for many years, but what we haven’t really had a handle on are functionally what each of them are doing in a complex brain structure,” she says. “These awards are meant to help understand that better. We’re starting with the mouse because it is a great experimental system, but we also have a few awards that are in humans and one in non-human primate as well to start building our understanding going to a complex mammalian brain. It’s only when we understand what a healthy brain looks like that we can start to understand what a diseased or disordered brain looks like.”
Also See: Plan emerges to deal with deluge of data from brain research
The Allen Institute for Brain Science in Seattle has received three five-year BICCN grants totaling nearly $100 million—one of which is a $65.5 million award to create a comprehensive whole-brain atlas of cell types in the mouse.
“One of the really unique features about this project is something that the Allen Institute has created called the Common Coordinate Framework,” Beckel-Mitchener contends. “It’s basically a 3-D image of—at this point—a mouse brain, but what it allows us to do is register the data to a very specific location in the brain so we can understand a cell type. It’s almost like a kind of Google Earth or Google Maps through which you can place an individual cell in a particular region of the brain and know a lot about that cell. We don’t have a similar capability in human or non-human primate brain. So the mouse is really going to inform how we go forward and create something similar for the human brain, which is our ultimate goal.”
Under a $19.4 million BICCN grant, the Allen Institute will lead the creation of a multimodal atlas of human brain cell types, starting with a gene expression survey of single cells in the brain and spinal cord, as well as deeper analysis of samples from the cortex and hippocampus. Researchers will then analyze human neurosurgical and postmortem tissue samples, and characterize the spatial distribution and relative proportions of cells expressing different sets of genes.
In addition, they will form an international consortium of experts in studying and modeling live human cells, with the goal of classifying cells based on shape, physiology and connective properties, and then correlate these properties to gene expression data.
Each of the BICCN’s individual projects will use tissue specimens to analyze brain cell types. However, there won’t be a central biobank repository, she adds. “It really will come down to the data that are generated,” observes Beckel-Mitchener. “We have one particular grant that’s going to a data center—we’re calling it the BRAIN Cell Data Center. And, they are in charge of coordinating the intake of all of these different data types.”
As part of $14.5 million BICCN grant, the Allen Institute will be responsible for developing and maintaining the BRAIN Cell Data Center that will serve as a community resource for single cell data in the brain. The center will provide a web-portal for sharing single cell brain data, tools and knowledge generated by BICCN research partners, as well as create and implement data models, common community standards, and scientific results.
“The BRAIN Cell Data Center will be an incredibly valuable resource for the entire neuroscience community and beyond,” says Michael Hawrylycz, investigator at the Allen Institute for Brain Science. “Applying cell type semantic and spatial community standards to these data, and making them available online, provides a tremendous opportunity for improving our understanding of the diverse cell types in the brain and how they’re organized.”
According to Beckel-Mitchener, the BRAIN Cell Data Center will collect the data and run it through various quality control measures. “The actual raw data will be stored in a couple of data archives, which are also funded this year,” she adds. “So, we’ll have the raw data available as well as analyzed data from the data center.”
Raw data will be stored in two BRAIN Initiative data archives—one at the University of Maryland Baltimore and the other through a collaboration between Carnegie Mellon University, the Pittsburgh Supercomputing Center and the University of Pittsburgh.
“At the current time, there is no practical way to analyze, mine, share or interact with large (100 terabytes or larger) brain image datasets. To address this issue, this proposal establishes the BRAIN Imaging Archive (or more simply ‘the Archive’) data service in Pittsburgh,” states Carnegie Mellon University’s project summary. “The Archive encompasses the deposition of datasets, the integration of datasets into a searchable web- accessible system, the redistribution of datasets, and a computational enclave to allow researchers to process datasets in-place and share restricted and pre-release datasets. The Archive will, for the first time, provide researchers with a practical way to analyze, mine, share or interact with large image datasets by creating a unique public resource for the BRAIN research community.”
The University of Maryland Baltimore will develop the Neuroscience Multi-Omic Archive (NeMO Archive), a data repository specifically focused on the storage and dissemination of “omic” data from the BRAIN Initiative and related brain research projects.
“We will utilize a federated model for data storage such that the physical location of data can be distributed between the NeMO local file system, public repositories, and a cloud-based storage system (such as Amazon S3),” states UMB’s project abstract. “We will leverage this capability and distribute BRAIN Initiative data between our local file system and the cloud. The Nemo Archive will be a data resource consistent with the principles advanced by research community members who are launching resources in next generation NIH data ecosystem. These practices include FAIR Principles, documentation of APIs, data-indexing systems, workflow sharing, use of shareable software pipelines and storage on cloud-based systems.”
“For some investigators, it’s important to have access to the raw data—that’s how they make new discoveries,” contends Beckel-Mitchener. “They may have new software packages and algorithms to analyze the raw data, which can verify or validate some of the existing outputs, but it can also answer some other questions. For scientific purposes, it’s really important to have both kinds of data available to the research community.”
She notes that the first phase of BICCN is over five years. However, Beckel-Mitchener says NIH is hoping to also have a second phase of the program.
“The vision is to complete the mouse atlas within five years and to really launch the human studies in a much more significant way,” she concludes. “There’s a lot of challenges in doing this with human brains—as you might imagine—given the data complexity, the availability of source material and making sure that we have healthy brain tissue to analyze.”
“We don’t want identifiable (human) information” that come from patients, adds Beckel-Mitchener. “For the human data, it will likely be controlled access,” as opposed to the open access to the mouse data that will be permitted “because there are no concerns in terms of privacy,” she says.
More for you
Loading data for hdm_tax_topic #better-outcomes...