Skip to main content

Past Stories

Using AI to Find New Antibiotics Still a Work in Progress

Posted on by

Protein over a computer network

Each year, more than 2.8 million people in the United States develop bacterial infections that don’t respond to treatment and sometimes turn life-threatening [1]. Their infections are antibiotic-resistant, meaning the bacteria have changed in ways that allow them to withstand our current widely used arsenal of antibiotics. It’s a serious and growing health-care problem here and around the world. To fight back, doctors desperately need new antibiotics, including novel classes of drugs that bacteria haven’t seen and developed ways to resist.

Developing new antibiotics, however, involves much time, research, and expense. It’s also fraught with false leads. That’s why some researchers have turned to harnessing the predictive power of artificial intelligence (AI) in hopes of selecting the most promising leads faster and with greater precision.

It’s a potentially paradigm-shifting development in drug discovery, and a recent NIH-funded study, published in the journal Molecular Systems Biology, demonstrates AI’s potential to streamline the process of selecting future antibiotics [2]. The results are also a bit sobering. They highlight the current limitations of one promising AI approach, showing that further refinement will still be needed to maximize its predictive capabilities.

These findings come from the lab of James Collins, Massachusetts Institute of Technology (MIT), Cambridge, and his recently launched Antibiotics-AI Project. His audacious goal is to develop seven new classes of antibiotics to treat seven of the world’s deadliest bacterial pathogens in just seven years. What makes this project so bold is that only two new classes of antibiotics have reached the market in the last 50 years!

In the latest study, Collins and his team looked to an AI program called AlphaFold2 [3]. The name might ring a bell. AlphaFold’s AI-powered ability to predict protein structures was a finalist in Science Magazine’s 2020 Breakthrough of the Year. In fact, AlphaFold has been used already to predict the structures of more than 200 million proteins, or almost every known protein on the planet [4].

AlphaFold employs a deep learning approach that can predict most protein structures from their amino acid sequences about as well as more costly and time-consuming protein-mapping techniques.
In the deep learning models used to predict protein structure, computers are “trained” on existing data. As computers “learn” to understand complex relationships within the training material, they develop a model that can then be applied for making predictions of 3D protein structures from linear amino acid sequences without relying on new experiments in the lab.

Collins and his team hoped to combine AlphaFold with computer simulations commonly used in drug discovery as a way to predict interactions between essential bacterial proteins and antibacterial compounds. If it worked, researchers could then conduct virtual rapid screens of millions of new synthetic drug compounds targeting key bacterial proteins that existing antibiotics don’t. It would also enable the rapid development of antibiotics that work in novel ways, exactly what doctors need to treat antibiotic-resistant infections.

To test the strategy, Collins and his team focused first on the predicted structures of 296 essential proteins from the Escherichia coli bacterium as well as 218 antibacterial compounds. Their computer simulations then predicted how strongly any two molecules (essential protein and antibacterial) would bind together based on their shapes and physical properties.

It turned out that screening many antibacterial compounds against many potential targets in E. coli led to inaccurate predictions. For example, when comparing their computational predictions with actual interactions for 12 essential proteins measured in the lab, they found that their simulated model had about a 50:50 chance of being right. In other words, it couldn’t identify true interactions between drugs and proteins any better than random guessing.

They suspect one reason for their model’s poor performance is that the protein structures used to train the computer are fixed, not flexible and shifting physical configurations as happens in real life. To improve their success rate, they ran their predictions through additional machine-learning models that had been trained on data to help them “learn” how proteins and other molecules reconfigure themselves and interact. While this souped-up model got somewhat better results, the researchers report that they still aren’t good enough to identify promising new drugs and their protein targets.

What now? In future studies, the Collins lab will continue to incorporate and train the computers on even more biochemical and biophysical data to help with the predictive process. That’s why this study should be interpreted as an interim progress report on an area of science that will only get better with time.

But it’s also a sobering reminder that the quest to find new classes of antibiotics won’t be easy—even when aided by powerful AI approaches. We certainly aren’t there yet, but I’m confident that we will get there to give doctors new therapeutic weapons and turn back the rise in antibiotic-resistant infections.

References:

[1] 2019 Antibiotic resistance threats report. Centers for Disease Control and Prevention.

[2] Benchmarking AlphaFold-enabled molecular docking predictions for antibiotic discovery. Wong F, Krishnan A, Zheng EJ, Stark H, Manson AL, Earl AM, Jaakkola T, Collins JJ. Molecular Systems Biology. 2022 Sept 6. 18: e11081.

[3] Highly accurate protein structure prediction with AlphaFold. Jumper J, Evans R, Pritzel A, Kavukcuoglu K, Kohli P, Hassabis D., et al. Nature. 2021 Aug;596(7873):583-589.

[4] ‘The entire protein universe’: AI predicts shape of nearly every known protein. Callaway E. Nature. 2022 Aug;608(7921):15-16.

Links:

Antimicrobial (Drug) Resistance (National Institute of Allergy and Infectious Diseases/NIH)

Collins Lab (Massachusetts Institute of Technology, Cambridge)

The Antibiotics-AI Project, The Audacious Project (TED)

AlphaFold (Deep Mind, London, United Kingdom)

NIH Support: National Institute of Allergy and Infectious Diseases; National Institute of General Medical Sciences


Suicide Prevention Research in a Rapidly Changing World

Posted on by

A person's hand being held by another person
Credit: iStock/PeopleImages

As I sit down to write this blog, the COVID-19 pandemic continues to have a widespread impact, and we’re all trying to figure out our “new normal.” For some, figuring out the new normal has been especially difficult, and that’s something for all of us to consider during September, which is National Suicide Prevention Awareness Month. It’s such an important time to share what we know about suicide prevention and consider how we can further this knowledge to those in need.

At NIH’s National Institute of Mental Health (NIMH), we’ve been asking ourselves: What have we learned about suicide risk and prevention during the pandemic? And how should our research evolve to reflect a rapidly changing world?

Addressing Disparities

Over the last few years, people have been concerned about the pandemic’s impact on suicide rates. So far, data suggest that the overall suicide rate in the U.S. has remained steady. But there is concerning evidence that the pandemic has disproportionately affected suicide risk in historically underserved communities.

For example, data suggest that people in minority racial and ethnic groups experienced greater increases in suicidal thoughts during the pandemic [1]. Additional data indicate that suicide rates may be rising among some young adult racial and ethnic minority groups [2].

Structural racism and other social and environmental factors are major drivers of mental health disparities, and NIMH continues to invest in research to understand how these social determinants of health influence suicide risk. This research includes investigations into the effects of long-term and daily discrimination.

To mitigate these effects, it is critical that we identify specific underlying mechanisms so that we can develop targeted interventions. To this end, NIMH is supporting research in underserved communities to identify suicide risk and the protective factors and effective strategies for reducing this risk (e.g., RFA-MH-22-140, RFA-MH-21-188, RFA-MH-21-187). There are important lessons to be learned that we can’t afford to miss.

Building Solid Foundations

The pandemic also underscored the urgent need to support youth mental health. Indeed, in December 2021, U.S. Surgeon General Dr. Vivek Murthy issued the Advisory on Protecting Youth Mental Health, calling attention to increasing rates of depression and suicidal behaviors among young people. Crucially, the advisory highlighted the need to “recognize that mental health is an essential part of overall health.”

At NIMH, we know that establishing a foundation for good mental health early on can support a person’s overall health and well-being over a lifetime. In light of this, we are investing in research to identify effective prevention efforts that can help set kids on positive mental health trajectories early in life.

Additionally, by re-analyzing research investments already made, we are looking to see whether these early prevention efforts have meaningful impacts on later suicide risk and mental health outcomes. These findings may help to improve a range of systems—such as schools, social services, and health care—to better support kids’ mental health needs.

Improving and Expanding Access

The pandemic has also shown us that telehealth can be an effective means of delivering and increasing access to mental health care. The NIMH has supported research examining telehealth as a tool for improving suicide prevention services, including the use of digital tools that can help extend provider reach and support individuals at risk for suicide.

At the same time, NIMH is investing in work to understand the most effective ways to help providers use evidence-based approaches to prevent suicide. This research helps inform federal partners and others about the best ways to support policies and practices that help prevent suicide deaths.

In July, the Substance Abuse and Mental Health Services Administration (SAMHSA) launched the 988 Suicide & Crisis Lifeline, a three-digit suicide prevention and mental health crisis number. This service builds on the existing National Suicide Prevention Lifeline, allowing anyone to call or text 988 to connect with trained counselors and mental health services. Research supported by NIMH helped build the case for such lifelines, and now we’re calling for research aimed at identifying the best ways to help people use this evolving crisis support system

Looking Ahead

With these and many other efforts, we are hopeful that people who are at risk for suicidal thoughts and behaviors will be able to access the evidence-based support and services they need. This National Suicide Prevention Awareness Month, I’d like to issue a call to action: Help raise awareness by sharing resources on how to recognize the warning signs for suicide and how to get help. By working together, we can prevent suicide and save lives.

References:

[1] Racial and ethnic disparities in the prevalence of stress and worry, mental health conditions, and increased substance use among adults during the COVID-19 pandemic – United States, April and May 2020. McKnight-Eily LR, Okoro CA, Strine TW, Verlenden J, Hollis ND, Njai R, Mitchell EW, Board A, Puddy R, Thomas C. MMWR Morb Mortal Wkly Rep. 2021 Feb 5;70(5):162-166.

[2] One Year In: COVID-19 and Mental Health. National Institute of Mental Health Director’s Message. April 9, 2021.

Links:

988 Suicide & Crisis Lifeline (Substance Abuse and Mental Health Services Administration, Rockville, MD)

Substance Abuse and Mental Health Services Administration Treatment Locator (SAMHSA)

Help for Mental Illnesses (National Institute of Mental Health/NIH)

Suicide Prevention (NIMH)

Digital Shareables on Suicide Prevention (NIMH)

Digital Shareables on Coping with COVID-19 (NIMH)

NIMH Director’s Messages about COVID-19 (NIMH)

NIMH Director’s Messages about Suicide (NIMH)

Note: Dr. Lawrence Tabak, who performs the duties of the NIH Director, has asked the heads of NIH’s Institutes and Centers (ICs) to contribute occasional guest posts to the blog to highlight some of the interesting science that they support and conduct. This is the 16th in the series of NIH IC guest posts that will run until a new permanent NIH director is in place.


The Amazing Brain: Where Thoughts Trigger Body Movement

Posted on by

3D column of red neurons (top) and blue neurons (middle)
Credit: Nicolas Antille, SUNY Downstate Health Sciences University, Brooklyn, NY

You’re looking at a section of a mammalian motor cortex (left), the part of the brain where thoughts trigger our body movements. Part of the section is also shown (right) in higher resolution to help you see the intricate details.

These views are incredibly detailed, and they also can’t be produced on a microscope or any current state-of-the-art imaging device. They were created on a supercomputer. Researchers input vast amounts of data covering the activity of the motor cortex to model this highly detailed and scientifically accurate digital simulation.

The vertical section (left) shows a circuit within a column of motor neurons. The neurons run from the top, where the brain meets the skull, downward to the point that the motor cortex connects with other brain areas.

The various colors represent different layers of the motor cortex, and the bright spots show where motor neurons are firing. Notice the thread-like extensions of the motor neurons, some of which double back to connect cells from one layer with others some distance away. All this back and forth makes it appear as though the surface is unraveling.

This unique imaging was part of this year’s Show Us Your Brain Photo and Video contest, supported by NIH’s Brain Research through Advancing Innovative Neurotechnologies® (BRAIN) Initiative. Nicolas Antille, an expert in turning scientific data into accurate and compelling visuals, created the images using a scientific model developed in the lab of Salvador Dura-Bernal, SUNY Downstate Health Sciences University, Brooklyn, NY. In the Dura-Bernal lab, scientists develop software and highly detailed computational models of neural circuits to better understand how they give rise to different brain functions and behavior [1].

Antille’s images make the motor neurons look densely packed, but in life the density would be five times as much. Antille has paused the computer simulation at a resolution that he found scientifically and visually interesting. But the true interconnections among neurons, or circuits, inside a real brain—even a small portion of a real brain—are more complex than the most powerful computers today can fully process.

While Antille is invested in revealing brain circuits as close to reality as possible, he also has the mind of an artist. He works with the subtle interaction of light with these cells to show how many individual neurons form this much larger circuit. Here’s more of his artistry at work. Antille wants to invite us all to ponder—even if only for a few moments—the wondrous beauty of the mammalian brain, including this remarkable place where thoughts trigger movements.

Reference:

[1] NetPyNE, a tool for data-driven multiscale modeling of brain circuits. Dura-Bernal S, Suter BA, Gleeson P, Cantarelli M, Quintana A, Rodriguez F, Kedziora DJ, Chadderdon GL, Kerr CC, Neymotin SA, McDougal RA, Hines M, Shepherd GM, Lytton WW. Elife. 2019 Apr 26;8:e44494.

Links:

Nicolas Antille

Dura-Bernal Lab (State University of New York Downstate, Brooklyn)

Brain Research through Advancing Innovative Neurotechnologies® (BRAIN) Initiative (NIH)

Show Us Your BRAINs Photo & Video Contest (BRAIN Initiative)

NIH Support: National Institute of Biomedical Imaging and Bioengineering; National Institute of Neurological Disorders and Stroke; BRAIN Initiative


The Amazing Brain: Tight-Knit Connections

Posted on by

colored tracts create a model of the entire brain
Credit: Sahar Ahmad, Ye Wu, and Pew-Thian Yap, The University of North Carolina, Chapel Hill

You’ve likely seen pictures of a human brain showing its smooth, folded outer layer, known as the cerebral cortex. Maybe you’ve also seen diagrams highlighting some of the brain’s major internal, or subcortical, structures.

These familiar representations, however, overlook the brain’s intricate internal wiring that power our thoughts and actions. This wiring consists of tightly bundled neural projections, called fiber tracts, that connect different parts of the brain into an integrated neural communications network.

The actual patterns of these fiber tracts are represented here and serve as the featured attraction in this award-winning image from the 2022 Show Us Your BRAINs Photo and Video contest. The contest is supported by NIH’s Brain Research through Advancing Innovative Neurotechnologies® (BRAIN) Initiative.

Let’s take a closer look. At the center of the brain, you see some of the major subcortical structures: hippocampus (orange), amygdala (pink), putamen (magenta), caudate nucleus (purple), and nucleus accumbens (green). The fiber tracts are presented as colorful, yarn-like projections outside of those subcortical and other brain structures. The various colors, like a wiring diagram, distinguish the different fiber tracts and their specific connections.

This award-winning atlas of brain connectivity comes from Sahar Ahmad, Ye Wu, and Pew-Thian Yap, The University of North Carolina, Chapel Hill. The UNC Chapel Hill team produced this image using a non-invasive technique called diffusion MRI tractography. It’s an emerging approach with many new possibilities for neuroscience and the clinic [1]. Ahmad’s team is putting it to work to map the brain’s many neural connections and how they change across the human lifespan.

In fact, the connectivity atlas you see here isn’t from a single human brain. It’s actually a compilation of images of the brains of multiple 30-year-olds. The researchers are using this brain imaging approach to visualize changes in the brain and its fiber tracts as people grow, develop, and mature from infancy into old age.

Sahar says their comparisons of such images show that early in life, many dynamic changes occur in the brain’s fiber tracts. Once a person reaches young adulthood, the connective wiring tends to stabilize until old age, when fiber tracts begin to break down. These and other similarly precise atlases of the human brain promise to reveal fascinating insights into brain organization and the functional dynamics of its architecture, now and in the future.

Reference:

[1] Diffusion MRI fiber tractography of the brain. Jeurissen B, Descoteaux M, Mori S, Leemans A. NMR Biomed. 2019 Apr;32(4):e3785.

Links:

Brain Basics: Know Your Brain (National Institute of Neurological Disorders and Stroke/NIH)

Sahar Ahmad (The University of North Carolina, Chapel Hill)

Ye Wu (The University of North Carolina, Chapel Hill)

Pew-Thian Yap (The University of North Carolina, Chapel Hill)

Brain Research through Advancing Innovative Neurotechnologies® (BRAIN) Initiative (NIH)

Show Us Your BRAINs Photo & Video Contest (BRAIN Initiative)

NIH Support: BRAIN Initiative; National Institute of Mental Health


The Amazing Brain: Capturing Neurons in Action

Posted on by

Credit: Andreas Tolias, Baylor College of Medicine, Houston

With today’s powerful imaging tools, neuroscientists can monitor the firing and function of many distinct neurons in our brains, even while we move freely about. They also possess another set of tools to capture remarkable, high-resolution images of the brain’s many thousands of individual neurons, tracing the form of each intricate branch of their tree-like structures.

Most brain imaging approaches don’t capture neural form and function at once. Yet that’s precisely what you’re seeing in this knockout of a movie, another winner in the Show Us Your BRAINs! Photo and Video Contest, supported by NIH’s Brain Research through Advancing Innovative Neurotechnologies® (BRAIN) Initiative.

This first-of-its kind look into the mammalian brain produced by Andreas Tolias, Baylor College of Medicine, Houston, and colleagues features about 200 neurons in the visual cortex, which receives and processes visual information. First, you see a colorful, tightly packed network of neurons. Then, those neurons, which were colorized by the researchers in vibrant pinks, reds, blues, and greens, pull apart to reveal their finely detailed patterns and shapes. Throughout the video, you can see neural activity, which appears as flashes of white that resemble lightning bolts.

Making this movie was a multi-step process. First, the Tolias group presented laboratory mice with a series of visual cues, using a functional imaging approach called two-photon calcium imaging to record the electrical activity of individual neurons. While this technique allowed the researchers to pinpoint the precise locations and activity of each individual neuron in the visual cortex, they couldn’t zoom in to see their precise structures.

So, the Baylor team sent the mice to colleagues Nuno da Costa and Clay Reid, Allen Institute for Brain Science, Seattle, who had the needed electron microscopes and technical expertise to zoom in on these structures. Their data allowed collaborator Sebastian Seung’s team, Princeton University, Princeton, NJ, to trace individual neurons in the visual cortex along their circuitous paths. Finally, they used sophisticated machine learning algorithms to carefully align the two imaging datasets and produce this amazing movie.

This research was supported by Intelligence Advanced Research Projects Activity (IARPA), part of the Office of the Director of National Intelligence. The IARPA is one of NIH’s governmental collaborators in the BRAIN Initiative.

Tolias and team already are making use of their imaging data to learn more about the precise ways in which individual neurons and groups of neurons in the mouse visual cortex integrate visual inputs to produce a coherent view of the animals’ surroundings. They’ve also collected an even-larger data set, scaling their approach up to tens of thousands of neurons. Those data are now freely available to other neuroscientists to help advance their work. As researchers make use of these and similar data, this union of neural form and function will surely yield new high-resolution discoveries about the mammalian brain.

Links:

Tolias Lab (Baylor College of Medicine, Houston)

Nuno da Costa (Allen Institute for Brain Science, Seattle)

R. Clay Reid (Allen Institute)

H. Sebastian Seung (Princeton University, Princeton, NJ)

Machine Intelligence from Cortical Networks (MICrONS) Explorer

Brain Research through Advancing Innovative Neurotechnologies® (BRAIN) Initiative (NIH)

Show Us Your BRAINs Photo & Video Contest (BRAIN Initiative)

NIH Support: BRAIN Initiative; Common Fund


Previous Page Next Page