Pittsburgh scientists to build maps of body cells
Pittsburgh computer scientists will be key players in a National Institutes of Health project to map the human body at the cellular level and show how cells in the body’s tissues carry out the processes that keep people alive and healthy.
The NIH on Wednesday announced funding awards to launch the $54 million Human BioMolecular Atlas Program, with more than $2 million going to Carnegie Mellon University computer scientists and another $598,000 to the Pittsburgh Supercomputing Center and the University of Pittsburgh Department of Biomedical Informatics.
A comprehensive and accessible collection — an atlas — of every type of cell is a natural next step to the $3 billion Human Genome Project completed in 2003, according to Ziv Bar-Joseph, a professor of computational biology and machine learning in the Carnegie Mellon University School of Computer Science. He will be leading the CMU-based computational tools center for the program, developing high-resolution 3D maps of the cells in the human body.
In an announcement Wednesday, Mr. Bar-Joseph explained:
“Though the genome — the
complete set of genes — is present in all cells, it’s clearly doing different things in different organs and tissues. The job now is to create a map that shows which genes and proteins are activated in each part of the body.”
Some research groups in the program, labeled HuBMAP, will generate the new data on how cells are organized and how they vary in specific tissues or organs. Developing tools to process the massive amounts of raw data collected is the task of the Pittsburgh group, along with colleagues at the University of California Santa Cruz and the Wellcome Sanger Institute in the U.K.
NIH has labeled its group HIVE, for HuBMAP Integration, Visualization and Engagement.
The supercomputing center’s interim director, Nick Nystrom, is principal investigator of the portion of the HIVE team developing the infrastructure to store all types of large high-resolution imaging, genome sequencing and other data and make the data available to users.
“This is an incredible opportunity that will unlock access to data, allowing talented researchers nationwide to generate new discoveries,” Mr. Nystrom said in a separate announcement Wednesday.
“The overall goal is to map as much as possible the different cell types of the body,” Mr. Bar-Joseph told the Pittsburgh Post-Gazette. “We know the [genetic] sequence. Each cell of our body uses a different part. Your lungs, your heart, your brain are doing different things.”
Every type of cell has not been named, he said, but with recent technological breakthroughs, scientists are able to see what is going on inside individual cells.
“You can see the components of different cells,” Mr. Bar-Joseph said. “A lot of diseases are associated with things that go wrong.”
Just as Google Earth allows people to zoom into any detail around the world, he said, creating a user-friendly atlas could give physicians the ability to zoom into 3D images of their patients’ cells, possibly helping with diagnosing and treating the patient.
Genomic information will be linked to imaging for different parts of the body, he said.
“Even physicians, usually not comfortable with large amounts of data, should be able to interact with the map itself.”
Much more detailed information can be layered on top of images associated with each tissue sample, he said.
One of the challenges will be how to register an image’s location in the body, he said.
“What’s the address of a specific location?” he said. “How would you know where to go?”
This is where a Google expert is expected to help in the project, he said.
“This is really a discovery thing,” he said, instead of looking for genes specific to cancer in a tissue sample, the project will be identifying cells according to where they belong.
Data will be collected from healthy people, just as the Human Genome Project was conducted. The difference, Mr. Bar-Joseph said, is these will be 3D images.
The initial timeline for the project is four years, he said. That’s the “thinking part,” he explained — collecting the data, then applying the computational method to process it. Funding can be changed each year, he said, so by the third year, he expects the funding to grow. Another four years of map production will follow.
Both CMU and Pitt have invested heavily in the supercomputing center, Mr. Bar-Joseph said, giving Pittsburgh resources particularly suited to the HuBMAP project.
“It utilizes a lot of stuff going on here,” he said.