Skip to main content
Premium Trial:

Request an Annual Quote

Allen Institute Launches $173M NIH-Funded Coalition to Create Human, Primate Brain Cell Atlases


CHICAGO – A coalition led by the Allen Institute has won $173 million in a series of five grants from the US National Institutes of Health to create, among other things, the first-ever complete cell atlas for the human brain, as well as the brains of marmosets and macaques.

This total is approximately one-third of the $500 million in funding NIH announced Thursday among 11 grants for a project called the BRAIN Initiative Cell Atlas Network (BICAN). Part of the agency's Brain Research Through Advancing Innovative Neurotechnologies (BRAIN) Initiative, BICAN will attempt to map the approximately 200 billion neurons and other cells in the brain through single-cell sequencing, noninvasive medical imaging, and advanced bioinformatic analysis. The Allen Institute is likening it to a Human Genome Project for the brain.

The Seattle-based institute unveiled its part of the program on Thursday, though work actually began early this month. Besides Allen, about 20 institutions in the US, Europe, and Japan are participating , including the Broad Institute, Baylor College of Medicine, Karolinska Institutet, Vrije Universiteit Amsterdam, Université Claude-Bernard Lyon, and Riken.

Of the $173 million total, about $100 million is earmarked for the human and nonhuman primate brain atlases, according to Allen. The rest of the money will support program coordination and administration, the development of a web-based platform so researchers can access BICAN findings, and the creation of a map of a mouse brain.

BICAN is the third phase of an ongoing effort to create a cellular map of the brain, following a pilot phase and the BRAIN Initiative Cell Census Network (BICCN). In 2017, NIH awarded the Allen Institute nearly $100 million for BICCN in a series of five-year grants to establish reference brain cell atlases of the brain, building the groundwork for the new undertaking.

"These awards will enable researchers to explore the multifaceted characteristics of the more than 200 billion neurons and non-neuronal cells in the human brain at unprecedented detail and scale — a feat in advanced technologies and cross-team research collaboration that will reveal new paradigms for understanding how pathological changes in particular groups of brain cells could cause neurologic and neuropsychiatric disorders," NIH BRAIN Initiative Director John Ngai said in a statement.

The Allen Institute has a long history in brain mapping, having introduced the Allen Brain Atlas in 2006, a genome-wide image database of gene expression in the mouse brain. That atlas included viewing software called 3D Brain Explorer.

Five years later, the organization followed with the Allen Human Brain Atlas, which characterized gene expression and histological patterns at about 1,000 sites in each of two adult human brains.

Ed Lein, senior investigator at the Allen Institute for Brain Science — one of six scientific divisions within the institute — called the new effort "remarkably aspirational." The hope is for these references to serve as "the basis for starting to understand the cellular origins of disease and the linkage of genes to cell types," he told GenomeWeb.

Lein noted that the cellular composition of mouse and monkey brains is similar enough, despite the difference in size, to infer properties of the human brain. "This actually becomes a very powerful strategy for defining the cellular makeup of the brain, where you can first define the cells, then you can map them to create a spatial map," he said.

Guoping Feng, associate director of the McGovern Institute at the Massachusetts Institute of Technology and director of model systems and neurobiology at the Broad Institute's Stanley Center for Psychiatric Research, said that the project is meant to unravel the evolutionary purpose of brain cells, but, more importantly, to understand brain function and dysfunction.

"There are very limited things that you can do at a cellular level in humans," Feng said. "You cannot manipulate cellular functions."

Feng said that the brain is far more complicated than other organs like the liver, where cell types tend to be grouped together. "If we really want to understand any brain functions, we have to understand [cell types], how they connect to each other, and form functional circuits," he said.

David Van Essen, a neurobiologist at Washington University in St. Louis, who has dedicated much of his research to brain atlases in humans and nonhumans alike, called BICAN a "next-generation version of how we can compare structure, function, and molecular characteristics in humans and laboratory animals."

Van Essen was a principal investigator of the NIH-funded Human Connectome Project a decade ago that produced a network map of cortical areas of the brain. That project analyzed brain organization and circuitry in 1,200 healthy young adults.

For BICAN, the partners will conduct functional MRIs and then single-cell sequencing on four to 12 neurotypical nonhuman study subjects and single-cell sequencing on a similar number of neurotypical humans to help investigators map the functional organization of brains and create "very high quality" specimens, augmented with data from targeted brain regions in several hundred other subjects to study variation. Samples will come from adult subjects.

Lein, who called the project "very complicated" because of all the data types he and his colleagues want to integrate, said that the research will combine RNA sequencing and ATAC-seq (assay for transposase-accessible chromatin sequencing) to assess gene expression and open chromatin regions.

Most of the sequencing will be run on the 10x Genomics Chromium platform. The coalition will also be using Vizgen's Merscope spatial transcriptomics platform, according to Lein.

The Allen Institute will be conducting most of the sample collection and library preparation, though Princeton University will be the lead site for marmoset samples. Lein said that BICAN is currently soliciting contractors to perform the actual sequencing, which he expects to begin in the second quarter of 2023.

Feng said that the group has extensive planning work ahead of it before sequencing and analysis can commence. He noted that participating organizations must agree on sample collection, scanning, and imaging protocols so data will be comparable across sites.

Lein said that single-cell genomics techniques have "become the workhorses" of these atlas projects. Now, sequencing and analytics technologies have advanced "to the point where it actually became feasible to get a complete cellular understanding of the human brain, and furthermore, that that information can be used to identify homologous cell types across different species," he said.

However, there are computational hurdles associated with the scale of this project, particularly in regard to analysis of spatial transcriptomics. "Suddenly, we're not dealing with hundreds of thousands of cells," Lein said. "We're dealing with tens of millions of cells or even more, so we're having to scale up the data management component of that."

Van Essen said that the informatics challenge is to present massive amounts of data in a "coherent, spatially well-organized, and multimodal exploration of the diverse types of data within brains, across individuals, across species, to actually facilitate the ability of the scientific community to approach these challenging problems in much more sophisticated and powerful ways in the future."

Another challenge is that neuroimaging and cellular analysis have traditionally been separate, Lein noted. "This project aims to bring those together, and in doing so, to make the cellular map valuable for the imaging community, so we can link these things together and understand the cellular underpinnings of higher order of functional organization."

Feng suggested that due to the cost of assembling a complete spatial transcriptome, the NIH grants can perhaps cover the mapping of one human brain. Because a marmoset has a small brain compared to the other species, adding a map of that brain might fit within that budget, as well.

While the $100 million in NIH grants through 2027 for the brain genome maps themselves may turn out to be merely a down payment if the price tag escalates, Lein said that he would prefer to see the technology advance to bring the cost down.

By the end of the five-year funding period, Lein expects the brain mapping effort to be able to inform actual disease research. He noted that the Allen Institute has ongoing research into Alzheimer's disease that can already identify specific types of cortical cells that are lost over the course of that ailment.

The brain map, he said, should be generalizable by the time it is complete. "Anyone studying any disease will be able to map against this reference and understand the kinds of cells that are affected or the molecular pathways in particular kinds of cells in that particular disease," he said.

The atlas should also facilitate the development of genetic therapies in the future by helping investigators identify regulatory enhancer regions. "Instead of targeting a gene, [you will be able to target] a cell population and you can deliver some sort of a therapeutic gene," Lein said.