PRESS-NEWS.org - Press Release Distribution
PRESS RELEASES DISTRIBUTION

AI learns to ‘speak’ genetic ‘dialect’ for future SARS-CoV-2 mutation prediction

Researchers develop new neural language models to predict novel mutations in protein sequences

AI learns to ‘speak’ genetic ‘dialect’ for future SARS-CoV-2 mutation prediction
2025-03-27
(Press-News.org) It’s been five years since COVID-19 was declared a global pandemic. As SARS-CoV-2 shifts to endemic status, questions about its future evolution remain. New variants of the virus will likely emerge, driven by positive selection for traits such as increased transmissibility, longer infection duration and the ability to evade immune defenses. These changes could allow the virus to spread among previously immunized populations, potentially triggering new waves of infection.

Predicting new mutations in viruses is crucial for advancing life science research, particularly when trying to understand how viruses evolve, spread and affect public health. Traditionally, researchers rely on wet-lab experiments to study mutations. However, these experiments can be costly and time-consuming.

Researchers from the College of Engineering and Computer Science at Florida Atlantic University have developed a new method to predict mutations in protein sequences called Deep Novel Mutation Search (DNMS), a type of artificial intelligence model that uses deep neural networks.

For the study, they focused on the SARS-CoV-2 spike protein – the part of the virus responsible for helping it enter human cells – and used a protein language model to predict potential new mutations in this protein never seen before.

To do this, researchers used a language model, ProtBERT, which was specifically fine-tuned to understand the “dialect” of SARS-CoV-2 spike proteins. The model works by looking at potential mutations and ranking them based on several factors. These include grammaticality, which refers to how likely or “correct” a mutation is according to the grammatical rules learned by the model, as well as how similar the mutated sequence is to the original protein, which is measured by semantic change and attention change.

Results of the study, published in the journal Communications Biology, show that the DNMS language model can separate sequences into groups based on their similarities. The model can predict which mutations are likely to occur by looking for mutations that cause only small changes in the protein’s structure and function. This is important because, in most cases, viruses like SARS-CoV-2 evolve through small changes that allow them to adapt without drastically altering their overall function.

The DNMS method uses all available information about the sequence and the mutations to create a more accurate prediction of which mutations are likely to occur. Unlike prior research, which typically looks at changes to a reference protein sequence, DNMS introduces a parent-child mutation prediction model. The parent sequence (an existing protein sequence) is used to predict mutations, and these mutations are analyzed based on how they might evolve over time.

“Our model ranks all possible mutations to find the ones that are most likely to occur in the future,” said Xingquan “Hill” Zhu, Ph.D., senior author and a professor in FAU’s Department of Electrical Engineering and Computer Science. “Our study shows that mutations following the protein’s grammars, with minimal changes compared to the original sequence and low attention differences, are considered the most likely future mutations.”

The method first takes a given SARS-CoV-2 spike protein sequence and simulates all possible single-point mutations. For each mutated version of the protein, DNMS uses the ProtBERT model to calculate how likely each mutation is to follow the “grammar” of the protein (grammaticality) and how similar the mutated sequence is to the original sequence (semantic change). Additionally, the model looks at attention, a measure that has been used to study protein structure and function, but never before applied to mutation prediction.

“The key to our method lies in using the context provided by the parent sequence. This context is crucial for evaluating whether a potential mutation aligns with the ‘grammar’ of the protein,” said Zhu. “DNMS works by selecting a parent sequence from a phylogenetic tree – basically a family tree of viral strains – and simulating all possible mutations.”  

The study also looked at the relationship between the predicted mutations and the virus’ fitness, or how well it can replicate and survive. Findings show that mutations with high grammaticality, small semantic change, and low attention change were associated with higher viral fitness. This suggests that mutations which fit well within the biological “rules” of the protein and cause minimal disruption to the protein’s structure or function are more likely to be beneficial for the virus.

“We believe that using sequence data alone can help make these predictions, as proteins follow certain biological rules,” said Zhu.

The researchers tested the effectiveness of DNMS through statistical analysis. Their results show that DNMS outperforms other methods in predicting novel mutations because it combines all the relevant factors into a single, more accurate prediction model. 

“The fine-tuned, pre-trained language model developed by our researchers can predict which SARS-CoV-2 mutations are more likely to occur in the future,” said Stella Batalama, Ph.D., dean of the College of Engineering and Computer Science. “This method can be useful for guiding experimental research, as it provides predictions about mutations before they are observed in the population, helping public health officials track and prepare for new mutations before they spread widely.”

Study co-author is Magdalyn E. Elkin, a doctoral student in FAU’s Department of Electrical Engineering and Computer Science.

The research was sponsored by the United States National Science Foundation.

- FAU -

About FAU’s College of Engineering and Computer Science:

The FAU College of Engineering and Computer Science is internationally recognized for cutting-edge research and education in the areas of computer science and artificial intelligence (AI), computer engineering, electrical engineering, biomedical engineering, civil, environmental and geomatics engineering, mechanical engineering, and ocean engineering. Research conducted by the faculty and their teams expose students to technology innovations that push the current state-of-the art of the disciplines. The College research efforts are supported by the National Science Foundation (NSF), the National Institutes of Health (NIH), the Department of Defense (DOD), the Department of Transportation (DOT), the Department of Education (DOEd), the State of Florida, and industry. The FAU College of Engineering and Computer Science offers degrees with a modern twist that bear specializations in areas of national priority such as AI, cybersecurity, internet-of-things, transportation and supply chain management, and data science. New degree programs include Master of Science in AI (first in Florida), Master of Science and Bachelor in Data Science and Analytics, and the new Professional Master of Science and Ph.D. in computer science for working professionals. For more information about the College, please visit eng.fau.edu. 

 

About Florida Atlantic University:
Florida Atlantic University, established in 1961, officially opened its doors in 1964 as the fifth public university in Florida. Today, Florida Atlantic serves more than 30,000 undergraduate and graduate students across six campuses located along the Southeast Florida coast. In recent years, the University has doubled its research expenditures and outpaced its peers in student achievement rates. Through the coexistence of access and excellence, Florida Atlantic embodies an innovative model where traditional achievement gaps vanish. Florida Atlantic is designated as a Hispanic-serving institution, ranked as a top public university by U.S. News & World Report, and holds the designation of “R1: Very High Research Spending and Doctorate Production” by the Carnegie Classification of Institutions of Higher Education. Florida Atlantic shares this status with less than 5% of the nearly 4,000 universities in the United States. For more information, visit www.fau.edu.

END


[Attachments] See images for this press release:
AI learns to ‘speak’ genetic ‘dialect’ for future SARS-CoV-2 mutation prediction AI learns to ‘speak’ genetic ‘dialect’ for future SARS-CoV-2 mutation prediction 2 AI learns to ‘speak’ genetic ‘dialect’ for future SARS-CoV-2 mutation prediction 3

ELSE PRESS RELEASES FROM THIS DATE:

$50 million gift from the Weill Family Foundation establishes the Weill Cancer Hub East

$50 million gift from the Weill Family Foundation establishes the Weill Cancer Hub East
2025-03-27
New York, N.Y., and Princeton, N.J. (March 27, 2025)—With a mission to understand how nutrition and metabolism impact the body’s ability to control cancer, four leading research institutions have united under the Weill Cancer Hub East, an innovative, collaborative partnership that aims to transform cancer treatment. The initiative connects world-class experts from Princeton University, The Rockefeller University, Weill Cornell Medicine and the Ludwig Institute for Cancer Research to enhance a ...

Physics meets art: a new twist on interference patterns

Physics meets art: a new twist on interference patterns
2025-03-27
Tokyo, Japan – One of the simplest and most beautiful naturally occurring patterns can be observed when light is shined through a pair of slightly misaligned periodic structures. This phenomenon, known as the moiré effect, is not only pretty to look at, but also has important consequences for the properties of materials. In an article published in ACS Nano, a team led by researchers from the Institute of Industrial Science, The University of Tokyo, announced the discovery of a previously unseen moiré pattern: a series of periodic one-dimensional bands in tungsten ditelluride bilayers. In nanomaterials, moiré patterns ...

Elevating global heart failure care with new certification

2025-03-27
DALLAS, March 27, 2025 — More than 56 million people globally live with heart failure (HF), which prevents the body from getting enough of the oxygen-rich blood it needs to work properly.[1] While there is no cure for HF, many people with this condition can live full, enjoyable lives and disease progression can be slowed with the right treatment. Research shows that outcomes for patients with HF improve when health care professionals and hospitals provide guideline-directed medical therapies. A new Heart Failure Center ...

The MIT Press releases 2025 Direct to Open (D2O) Impact Report

2025-03-27
The MIT Press is proud to release our 2025 Impact Report for Direct to Open (D2O), our sustainable framework for open access monographs that shifts publishing from a solely market-based purchase model where individuals and libraries buy single eBooks, to a collaborative, library-supported open access model.   The continued growth in the reach of open access publishing couldn’t be more timely. In 2025, access to truth and facts are under attack, and democratizing access to trustworthy, peer-reviewed information has never been more important. In the face of so many forces working against the spread of knowledge, Direct to Open continues to be a critical tool. To date, ...

New study reveals the curative potential of genome editing approach for genetic deafness

New study reveals the curative potential of genome editing approach for genetic deafness
2025-03-27
Congenital hearing loss refers to impaired auditory function that occurs due to genetic causes. GJB2 is the gene responsible for approximately half of all cases of hereditary hearing loss. Connexin 26 (CX26), which is encoded by GJB2, helps in the formation of intercellular gap junctions—channels that allow for the movement of ions and chemical messenger molecules between adjacent cells, where it regulates auditory function.   GJB2 mutations often lead to fragmentation of gap junctions and gap junction plaques (GJPs) which are composed ...

AAAS elects Keck School of Medicine of USC molecular biologist Yali Dou as 2025 fellow

AAAS elects Keck School of Medicine of USC molecular biologist Yali Dou as 2025 fellow
2025-03-27
Molecular biologist Yali Dou, PhD, holder of the Marion and Harry Keiper Chair in Cancer Research and professor of medicine and cancer biology at the Keck School of Medicine of USC, has been elected a fellow of the American Association for the Advancement of Science (AAAS). She is one of seven USC faculty members in the 2025 cohort of new fellows. The AAAS is the world’s oldest and largest general science organization and the publisher of Science, a top peer-reviewed academic journal. Election as a fellow is a lifetime honor — one of the AAAS’s ...

Damaging cluster of UK winter storms driven by swirling polar vortex miles above Earth

2025-03-27
University of Leeds news  Embargoed until 10:00 GMT, 27 March  Damaging cluster of UK winter storms driven by swirling polar vortex miles above Earth  Powerful winter storms which led to deaths and power outages in the UK and Ireland were made more likely by an intense swirling vortex of winds miles above the Arctic, say scientists.  A team of researchers led by the University of Leeds has pinpointed a new reason for winter storm clusters such as the trio named Dudley, Eunice and Franklin, which hit the nation within the space of a week in February 2022.  The findings which are published today in the journal ...

Losing forest carbon stocks could put climate goals out of reach

2025-03-27
In the past, intact forests absorbed 7.8 billion tonnes of CO₂ annually – about a fifth of all human emissions – but their carbon storage is increasingly at risk from climate change and human activities such as deforestation. A new study from the Potsdam Institute for Climate Impact Research (PIK) shows that failing to account for the potentially decreasing ability of forests to absorb CO₂ could make reaching the Paris agreement targets significantly harder, if not impossible, and much more costly. “Delaying action leads to disproportionately higher costs,” explains Michael Windisch, ...

From weight to wellness: New database transforms obesity research

2025-03-27
A new medical database automatically compiles the medical records of obese patients and those suffering from obesity-related diseases in a uniquely comprehensive and reliable manner. The initiative, led by Kobe University, offers valuable insights for health promotion and drug development. “Obesity is at the root of many diseases,” says OGAWA Wataru, an endocrinologist at Kobe University. Obesity has been linked to the development of diabetes, hypertension, gout, coronary heart disease, stroke and many other diseases. Monitoring, treating and preventing obesity and the diseases it can cause is therefore not only good for ...

Nature’s viny vampire: Discovering what drives parasitic Cuscuta campestris

Nature’s viny vampire: Discovering what drives parasitic Cuscuta campestris
2025-03-27
The parasitic vine Cuscuta campestris grows by latching onto the stems and leaves of plants and inserting organs called haustorium into the host plant tissues to draw nutrients. The haustorium is formed when ion channels in the cell membrane are stimulated during coiling and induce a reaction within the cell. Further, Cuscuta campestris has many types of ion channels, but which ones were linked to the development of haustorium were previously unknown. “For the first time, the genes involved in sensing ...

LAST 30 PRESS RELEASES:

New study finds peripheral artery disease often underdiagnosed and undertreated; opportunity to improve treatments, lower death rates

Use of antidepressant medication linked to substantial increase in risk of sudden cardiac death 

Atrial fibrillation diagnosed in midlife is linked to a 21% increased risk of dementia at any age and a 36% higher risk of early-onset dementia 

Mode of death in patients with heart failure with mildly reduced or preserved ejection fraction

Intravenous ferric carboxymaltose in heart failure with iron deficiency

Artificial intelligence in the prevention of sudden death

Oral semaglutide vastly reduces heart attacks, strokes in people with type 2 diabetes

Prothrombin complex concentrate vs frozen plasma for coagulopathic bleeding in cardiac surgery

Who needs a statin? New study compares prescribing recommendations based on traditional risk factors vs. coronary artery calcium scoring

Finerenone and atrial fibrillation in heart failure

Low coronary artery calcium score is associated with an excellent prognosis regardless of a person’s age, new study finds

Groundbreaking consensus statement on conduction system pacing released: a major milestone in the evolution of pacing therapy

Nuclear monitoring system suggests landslide cut off internet in west Africa

PNNL scientist elected AAAS fellow

American College of Cardiology recognizes five JACC Rocket Fuel Consultants

American College of Cardiology, Association of Black Cardiologists recognize three Merck Research Fellowship awardees

JACC to recognize 2025 Simon Dack Award recipients, Elite Reviewers

American College of Cardiology honors two recipients with the William A. Zoghbi Global Research Initiative Award

JACC recognizes five recipients of the William W. Parmley Young Author Achievement Award

Mass General Brigham researchers identify mutations that can lead to resistance to some chemotherapies

JACC journals honor 10 young researchers

Jefferson Lab Director Kimberly Sawyer named to CoVaBIZ Magazine’s 150 Most Influential People List

The world according to mosquitoes: USU ecologists lead AI-based effort to identify disease vectors

Drexel researchers develop new DNA test for personalized treatment of bacterial vaginosis

Keith T. Flaherty, MD, FAACR, elected as American Association for Cancer Research President-Elect for 2025-2026

Brownie points for ChatGPT’s food analysis skills

The Giants Foundation provide 12 schools with CPR resources to improve cardiac emergency outcomes

Why scientists are worried about weasels

American College of Cardiology recognizes 21 Distinguished Award recipients

American College of Cardiology recognizes three recipients of the Hani Najm Global Scholar Award Observership Program

[Press-News.org] AI learns to ‘speak’ genetic ‘dialect’ for future SARS-CoV-2 mutation prediction
Researchers develop new neural language models to predict novel mutations in protein sequences