TL;DR
Get health and wellness essentials delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
The NATIVE-ID project, led by the Innovative Genomics Institute at UC Berkeley, will study early protein aggregation patterns and use the data to support AI predictions of protein dysfunction. Stowers Institute researcher Randal Halfmann’s lab is set to receive about $4.1 million over two years to measure 50,000 proteins in yeast. The project will initially focus on frontotemporal lobar degeneration, but whether its methods can predict disease onset or guide treatment remains to be shown.
A new multi-institutional research project will collect large-scale data on how proteins begin to clump together, then use those measurements to help train AI models to predict protein dysfunction linked to neurodegenerative disease. The project, called NATIVE-ID, is part of the U.S. Advanced Research Projects Agency for Health’s BIOGAMI program and will initially focus on frontotemporal lobar degeneration, or FTLD.
Stowers Institute for Medical Research investigator Randal Halfmann will lead the protein-measurement work. His laboratory is set to receive approximately $4.1 million over two years to test how changes in protein sequences affect aggregation in yeast cells. The wider research effort will receive up to $28.6 million in funding, according to the project announcement.
The lab plans to study 50,000 proteins under varied conditions intended to mimic changes that occur in aging human cells. Using Distributed Amphifluoric FRET, or DAmFRET, a technology developed by Halfmann’s team, researchers will measure protein self-assembly inside individual living cells. The team expects to analyze more than one million samples and produce more than 10 billion measurements.
NATIVE-ID is led by the Innovative Genomics Institute at UC Berkeley and includes researchers from Brown University, Emory University, Johns Hopkins University, Parallel Squared Technology Institute and Texas A&M University, alongside Stowers and UC Berkeley. The group’s work will pair the yeast-based experiments with human-neuron research so the team can examine whether model predictions hold in human cells.
Building Data for Protein Predictions
The project addresses a gap in current AI approaches to proteins. AI systems have advanced the prediction of structures for many proteins, but the project announcement says roughly one-third of proteins lack a stable structure. These intrinsically disordered proteins, or IDPs, can shift among different shapes, and some can aggregate in ways associated with diseases such as Alzheimer’s, Parkinson’s, ALS and Huntington’s.
Large-scale measurements could give researchers data to test whether sequence patterns help predict when these proteins change behavior. If the models prove useful, they might eventually help scientists identify early biological processes or select targets for further research. That is a research possibility, not an established clinical benefit: the project has not shown that it can predict an individual’s disease risk, prevent illness or improve treatment outcomes.
Halfmann said better estimates of disease probabilities and onset ages could allow more people to seek preventive or early-stage treatments or join clinical trials. That statement describes a potential future use. The immediate work is experimental data generation and model development, not a diagnostic service or treatment program.
As an affiliate, we earn on qualifying purchases.
From Yeast Studies to Human Neurons
The project draws on Halfmann’s earlier work on proteins connected with neurodegeneration. In 2023, his lab reported experimentally determining the structure of an initiating step in amyloid formation associated with Huntington’s disease. The team has also studied TDP-43, a protein associated with ALS and FTLD. Halfmann described those projects as pilot studies for the larger effort, which expands from hundreds of protein sequences to 50,000.
Yeast provides a system in which researchers can test many protein sequences at scale. The project’s researchers say earlier findings from Halfmann’s lab showed that disease-related protein behavior observed in yeast can also inform studies in human cells. NATIVE-ID plans to bring in complementary human-neuron capabilities to test how well predictions based on experimental data apply beyond yeast.
The initial focus is FTLD, which the project description says shares important genetic and biological features with ALS. The broader aim is to develop approaches that could extend to other diseases involving protein misfolding, but that wider reach remains a goal rather than a demonstrated result.
“If we can better predict the probabilities and onset ages of disease, it could allow many more people to seek preventive or early-stage treatments or enroll in clinical trials.”
— Randal Halfmann, Stowers Institute investigator
As an affiliate, we earn on qualifying purchases.
Prediction Still Needs Validation
The announcement does not report results from the planned experiments or describe a completed AI model. It is not yet clear how accurately a model will predict aggregation, how well predictions from yeast experiments will transfer to human neurons, or whether those predictions can identify disease onset in people.
The project’s stated figures are plans and expectations: the lab expects to generate more than 10 billion measurements across more than one million samples. The announcement does not specify the project’s full timeline, when model results will be available, or how any findings would be evaluated for clinical use. It also does not establish that the research will lead to preventive treatments.
neurodegenerative disease research kit
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Experiments and Neuron Testing
Halfmann’s team is expected to carry out the large-scale yeast experiments, measuring aggregation across protein sequences and conditions. Other NATIVE-ID researchers will contribute complementary work, including studies in human neurons, so the group can compare model predictions with biological behavior in a more disease-relevant setting.
The first research focus will be FTLD, with the project aiming to assess whether its methods could later apply to other disorders involving protein misfolding. The next substantive milestones are the generation and analysis of the experimental dataset, development of prediction models and testing those predictions in human-neuron systems. The announcement provides no dates for those milestones or for any clinical application.
As an affiliate, we earn on qualifying purchases.
Key Questions
What is the NATIVE-ID project?
NATIVE-ID is a multi-institutional research effort led by the Innovative Genomics Institute at UC Berkeley. It aims to collect data on protein aggregation and use that information to support AI predictions about disordered proteins.
What will the Stowers Institute team study?
Randal Halfmann’s lab plans to measure how 50,000 proteins aggregate in yeast cells under varied conditions. The team expects to produce more than 10 billion measurements using its DAmFRET technology.
Which disease will the project study first?
The project will initially focus on frontotemporal lobar degeneration (FTLD), which shares genetic and biological features with ALS. Researchers aim for methods that might apply more broadly, but that has not yet been demonstrated.
Can the research predict or prevent disease in people now?
No such clinical capability has been reported. The project is at the research and data-generation stage; whether its models can predict disease onset in people or contribute to prevention remains unknown.
How much funding is involved?
The project is described as receiving up to $28.6 million. Halfmann’s lab is set to receive approximately $4.1 million over two years for its contribution.
Source: rss
Cold & flu season Picks
humidifiers
As an affiliate, we earn on qualifying purchases.
