TL;DR
Get everyday helpers delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
A multi-institutional project called NATIVE-ID will use large-scale experiments and AI to study how disordered proteins begin to aggregate. Stowers Institute researcher Randal Halfmann’s lab is set to receive about $4.1 million over two years to measure aggregation across 50,000 proteins; the project will initially focus on frontotemporal lobar degeneration.
A new multi-institutional research project will combine large-scale protein experiments with artificial intelligence to study whether early protein dysfunction can be predicted before harmful aggregation is evident. The effort, called NATIVE-ID, is part of the Advanced Research Projects Agency for Health’s BIOGAMI program; Stowers Institute researcher Randal Halfmann’s lab is expected to receive about $4.1 million over two years to generate experimental data.
Halfmann’s group plans to test how changes in amino-acid sequences affect whether proteins stay in their normal state or begin to clump. Using yeast cells and the lab’s DAmFRET technology, the researchers will measure aggregation tendencies across 50,000 proteins under varied conditions intended to model changes associated with aging. The team expects to analyze more than one million samples and produce over 10 billion measurements.
The project is led by the Innovative Genomics Institute at the University of California, Berkeley and brings together researchers from institutions including Brown University, Emory University, Johns Hopkins University, Texas A&M University, and the Parallel Squared Technology Institute. The project’s partners are to contribute complementary expertise, including work with human neurons in which the model’s predictions will ultimately need to be tested.
The initial research focus is frontotemporal lobar degeneration (FTLD), which the source report describes as sharing genetic and biological features with ALS. The broader aim is to develop methods that may be applicable to other diseases involving protein misfolding, including Alzheimer’s, Parkinson’s, ALS, and Huntington’s. The announcement describes a research plan, not a clinical test or treatment.
Testing Predictions Before Aggregation
Many protein-prediction systems have advanced the study of proteins with stable structures. But the project addresses a different challenge: intrinsically disordered proteins can shift among many shapes rather than settling into one fixed structure. Some can aggregate in ways associated with neurodegenerative disease, making it difficult to infer their behavior from conventional structural predictions alone.
The planned measurements could give AI researchers a larger experimental foundation for studying how sequence relates to aggregation. If models can make useful predictions, researchers may gain ways to identify disease-related protein behavior earlier and guide future studies of prevention or early-stage interventions. Those are potential benefits, not results established by the funding announcement; the project must still test whether predictions hold in human cells and have practical value.
As an affiliate, we earn on qualifying purchases.
From Yeast Experiments to Human Neurons
Halfmann’s lab developed DAmFRET in 2018 to measure protein self-assembly inside individual living cells. The source report says the team will use yeast for the large-scale experiments, building on earlier work in which disease-relevant protein behavior observed in yeast also informed studies in human cells. The new project’s scale is a substantial expansion: Halfmann said earlier studies examined hundreds of protein sequences, while this effort plans to examine 50,000.
The work also builds on the lab’s research into polyglutamine proteins linked to Huntington’s disease and TDP-43, which is associated with ALS and FTLD. In 2023, the lab reported determining the structure of an initiating step in amyloid formation associated with Huntington’s. The new effort broadens that kind of inquiry, while adding AI modeling and collaboration with teams working with human neurons.
According to the report, the overall project can receive up to $28.6 million in funding. The available source gives Halfmann’s lab’s allocation and duration, but does not break down the full award among all participating institutions.
“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
Questions the Study Must Resolve
The announcement describes planned experiments and expected data output; it does not report completed results showing that AI can reliably forecast when a protein will aggregate or predict disease onset. It is also not yet clear how well findings from yeast experiments will translate to human neurons, or whether model predictions will be accurate across different cell conditions and diseases.
The source does not specify when the first datasets or trained models will be released, what performance thresholds the researchers will use, or how any findings might lead to clinical tools. A prediction of protein behavior would not by itself establish a person’s risk, provide a diagnosis, or show that an intervention can prevent disease.
Data Generation and Model Testing
Halfmann’s lab is expected to begin the large-scale experiments under its two-year funding award, measuring protein aggregation across the planned sequences and conditions. The wider NATIVE-ID collaboration is intended to use those results to train AI models and test whether predictions hold in human neurons. The project’s initial emphasis will be FTLD.
The next meaningful updates will be evidence about the dataset, the models’ predictive performance, and how reliably results transfer from yeast to human cells. The source announcement provides no specific dates for those milestones. Any proposed relevance to prevention, early treatment, or clinical-trial enrollment will depend on what the research establishes.
Key Questions
What is the NATIVE-ID project?
NATIVE-ID is a multi-institutional research effort combining protein experiments and AI to study early dysfunction and aggregation, initially focusing on FTLD.
What will Halfmann’s lab measure?
The lab plans to use DAmFRET in yeast cells to measure aggregation tendencies across 50,000 proteins under varied conditions. The team expects more than 10 billion measurements across over one million samples.
Does the announcement mean researchers can predict who will develop dementia or ALS?
No. The project aims to develop and test predictions about protein behavior. The announcement reports no validated method for predicting an individual’s disease risk or onset.
How much funding is involved?
The broader project is described as eligible for up to $28.6 million. Halfmann’s lab is expected to receive about $4.1 million over two years; the source does not provide a full institution-by-institution funding breakdown.
Source: rss
Halloween Picks
halloween
As an affiliate, we earn on qualifying purchases.
