Robot hand typing on a computer keyboard.

Open AI, Locked Minds

The scientific community needs to draw the line between use and abuse of artificial intelligence tools.

Image Credit:

©, sompong_tom

A couple of months ago, I woke up to mayhem on social media. Angry posts about how a paper featuring an image of a rodent with grotesque, biologically inaccurate genitalia had passed the peer-review process flooded my feed. In the next couple of months, more such instances arose: in one paper’s introduction, the authors erroneously left in a ChatGPT prompt, while in another instance, an AI chat response found its way into an article summary. Although the authors of the rat-gate paper had declared the usage of AI as per journal guidelines, these examples have led to a scientific discourse on the boundaries of AI usage in scientific publishing.

There’s not a straightforward answer in this case. On one hand, AI can help scientists communicate better in many ways. When Chat GPT was new, some open minded scientists started their intro slides at a conference with, “I asked Chat GPT about funding in my field” type of data. It was a fun approach to excite the audience and succinctly put forth their problem statement. Similarly, for non-native English speakers who struggle with writing papers, borrowing the right words from an AI tool could enable them to effectively convey their scientific findings. On the flip side, easy access to these technologies poses a risk for an increase in data manipulation and unreliable papers. 

While the jury is out on how AI will alter scientific publishing in the long run, in my opinion, scientists need to move past the everything-or-nothing mindset and see AI for what it is: a tool. Just like one cannot put random numbers into a calculator and expect it to file their taxes, scientists should not use AI without due diligence. Even as the scientific community navigates this wobbly path today, they can find a happy medium by simply putting human intelligence first.

What do you think about using AI in scientific publishing? 

Share Your Opinion

A baby cries

Why are Tears Important?

Human tears are a blend of elements that lubricate and protect the human eye.

Image Credit:

©, dorioconnell

Tears may evoke an association with emotions, but they are essential to human eye function. They coat the eyes to lubricate and protect them, and although their transparent appearance may trick people into thinking that they are just salty water, Jennifer Craig, an optometrist and vision sciences researcher at the University of Auckland, explained, “They’re way more complicated than that.” 

          A woman wears a black sleeveless shirt and smiles at the camera. She stands in front of a stained-glass window.
In her lab at the University of Auckland, Jennifer Craig, an optometrist and vision sciences researcher, studies ocular surface diseases, including dry eye and tear film dysfunction. 
Courtesy of New Zealand Optics

Tears consist of mucus, aqueous, and lipid layers on the eye surface. The innermost mucus layer covers the cornea and is composed primarily of mucins, large glycoproteins that are highly hydrophilic due to their multiple long carbohydrate chains.1 According to Craig, the primary function of this layer is to anchor the subsequent aqueous layer, which makes up the bulk of the tears’ volume, to the hydrophobic surface of the cornea. The aqueous layer contains many different components, including proteins, electrolytes, and antimicrobial and anti-inflammatory agents such as lysozyme and lactoferrin.2 Finally, the outermost layer is an oil layer. “This very thin lipid layer helps stop that watery layer from evaporating too quickly,” Craig said.  

As people grow older, their tears change. Aging associates with alterations in tear volume and composition, and even a reduction in the thickness of the tears’ lipid layer.3,4 These changes, combined with other biological and environmental factors, increase people’s risk of developing dry eye disease, in which decreased tear production or increased tear evaporation causes inflammation of the eyes.5  

Finding ways to help patients manage dry eye disease is an important goal for Craig. “We are still learning about the things that we can do to mitigate dry eye and prevent it from happening,” she said. 

What makes you curious? Submit a question for us to answer in future “Just Curious” columns.

Submit a Question


Tips for Optimizing Cell-Based Readouts

Explore five tips for improving detection and reducing background noise with cell-based fluorescence measurements on microplate readers.

Learn More
          Red fluorescent dyes icon
Red-shifted dyes
Cell-derived autofluorescence is mainly found in the green light range. Use red-shifted fluorescent dyes to limit and circumvent nonspecific background fluorescence.
          Focal height icon
Focal height
Modulating the microplate reader’s focal plane based on where sample signal intensity is at its highest will ensure optimal detection. For adherent cells grown on microplates, adjust the detection focus of the plate reader to the bottom of the well.
          Low autofluorescence icon
Low-autofluorescence media
Cell culture medium components such as serum or phenol red are a source of autofluorescence. Use buffers or a low-autofluorescence medium to reduce this nonspecific background signal.
          Scan options icon
Scan options
Plate readers typically measure in the center of the well, which can be problematic when working with heterogeneously distributed adherent cells. Perform well scans that spread measurements across the whole well surface, reducing data variability.
          Bottom optic icon
Bottom optics
When measuring adherent cells from above the well, excitation and emission light must pass through the potentially auto-fluorescent supernatant. Select bottom optics to avoid autofluorescence effects derived from media or buffer.

Learn more about cell-based fluorescence measurements on microplate readers.

Synthetic Systems for Studying Natural Cells

Oskar Staufer engineers synthetic systems to explore cancer biology.

          Larger green cell with two cyan spots and many smaller magenta circles in and around the cell. 
Oskar Staufer developed synthetic vesicles to study cancer cell biology. After loading the vesicles (magenta) with an enzyme that lowers the levels of reactive oxygen species, he introduced them into human breast cancer cells (green; nuclei labeled in cyan).  

Natural systems are heterogeneous and variable, complicating scientists' efforts to make sense of their observations. In contrast, synthetic systems provide a highly controllable platform for probing the natural world. To this end, scientists are creating lipid-based synthetic doppelgängers of cells and organelles to ask fundamental questions about biology. 

Oskar Staufer, a biophysicist at the Leibniz Institute for New Materials, uses synthetic biology to study cancer and the tumor immune environment. Cancer is a dysregulated system. The body loses control over proliferation and the immune environment surrounding the cell. “On a fundamental level, the idea is to regain control over the regulatory mechanisms within cancer cells,” said Staufer. This is where synthetic vesicles come into the picture. 

To generate these artificial structures, Staufer developed a microfluidic-based mechanical platform that divvies up a lipid cocktail into smaller water-in-oil droplets, or vesicles.1  At approximately two micrometers in diameter, these synthetic vesicles are small enough to enter a cell but large enough to transport sizeable cargoes, such as high concentrations of an enzyme

To probe one element of cancer cell regulation, Staufer loaded his synthetic vesicles with an enzyme that reduces the levels of reactive oxygen species (ROS), signaling molecules that are abundant in cancer cells and shape the tumor microenvironment.2,3 Then by tweaking the lipid composition and stiffness of the membrane, he created a vesicle that natural cells readily scoop up via endocytosis. Once inside the mammary gland-derived adenocarcinoma cell, shown in the above image in green, the synthetic vesicles, shown in magenta, reduced ROS levels, thus mimicking the natural functionality of peroxisomes, which are specialized organelles that carry out oxidative reactions. 

Synthetic vesicle technologies are still in their infancy, and Staufer noted, “We’re just trying to find out how the living and nonliving matter interact with each other.” 

  1. Staufer O, et al. Biomaterials. 2020;264:120203.
  2. Staufer O, et al. Small. 2020;16(27):e1906424.
  3. Perillo B, et al. Exp Mol Med. 52(2):192-203.
Cartoon of a woman, a dog, an octopus, a small spider, and a fly dreaming.

What is the Smallest Animal to Dream?

Neuroscience research suggests that many animals experience dream-like states.

Image Credit:

MODIFIED FROM © ISTOCK.COM, Rudzhan Nagiev, Arif_Vector, Nataliia Nesterenko, sceka; © STOCK.ADOBE.COM, Natalia; DESIGNED BY ERIN LEMIEUX

No two dreams look the same, but their underlying neurology is similar. “The brain activity, it looks like it’s awake,” explained Bruno van Swinderen, a neuroscientist at the University of Queensland who studies consciousness and the function of sleep. Dreaming appears to be a distinct process with neurological and physiological responses, such as brain activity and bodily movement. 

          Profile photograph of a white man in a blue, red, and yellow shirt.
Bruno van Swinderen studies the functions of sleep at the University of Queensland. 
University of Queensland

Scientists observed these types of actions across animal models using electrical brain probes and cameras to capture eye movement and other bodily changes. From flies to octopuses, they found evidence that these animals dream.1,2 If researchers can observe dream-like behaviors in so many animals, it begs the question, what is the smallest animal to dream?

To van Swinderen, this comes back to the function of dreaming, which he studies as a mechanism for anticipating and interpreting the world in which one lives. “The evolution of active sleep or the evolution of dreaming is fundamentally linked to the evolution of a capacity to pay attention,” he said. Animals that move in their environments and must react to environmental changes have to be able to make predictions. van Swinderen suspects that dreaming helps the brain practice this. 

For example, scientists used calcium imaging to detect neuron activation in sleeping mice  and found that it mirrored that of feeding behaviors.3 This neural activity appears to be tied to complicated brain structures. “The animals that don’t have [these brain structures] are the ones that basically don’t really have a brain,” van Swinderen said. These would include nematodes and jellyfish but exclude arthropods like insects and crustaceans. Considering that arthropods, such as the commonly researched fruit flies and jumping spiders, are on the scale of a few millimeters, dreaming animals can get quite small.4  

What are you curious about? Submit your question for a chance to find an answer in an upcoming “Just Curious” column.

Submit your question

Empty lecture hall with a chalkboard behind the teacher desk.

Should Research Faculty Have Teaching Experience?

Researchers bring their topical expertise to the lecture hall, but formal classroom training isn’t required for most faculty appointments. Should this change?

Image Credit:


At most universities, the research faculty’s main responsibility is to conduct research, but they are often required to teach courses as well. However, prior teaching experience isn’t required for academic faculty positions, and most researchers rarely complete formal training in instruction. Two faculty members weigh in on whether researchers should have more teaching experience or resources to fulfill these duties.

          Headshot of Mary Kay Lobo
Mary Kay Lobo is a neuroscientist at the University of Maryland at Baltimore. She teaches lectures for the medical school and has taught courses in neuroscience for the graduate program. 
Mary Kay Lobo

Mary Kay Lobo

I completed a short teaching course during graduate school and taught as a teaching assistant and gave a lecture as a postdoctoral researcher. As experts in our fields, we also present seminars, even as part of our hiring process, so we do have some experience communicating to an audience even if it isn’t formal teaching experience. 

Without a formal opportunity, though, like being a teaching assistant as a graduate student, getting teaching experience may be difficult. For instance, postdoctoral researchers can choose to teach lectures but are mostly expected to work in the lab. Having more training would have been beneficial to me, so providing an orientation to faculty with tips for teaching effectively and having resources to use for putting together courses could help us be more prepared. 

          Headshot of Tanecia Mitchell
Tanecia Mitchell is a mitochondrial and redox biologist at the University of Alabama at Birmingham. She is a course director for a graduate level ethics course and specialty course on mitochondria.
Tez Davenport

Tanecia Mitchell

I chose to complete a teaching certificate early in my career. In addition to teaching outside of the university and seeking guidance from my postdoctoral advisor, that certification helped improve my teaching abilities. 

For current research-focused faculty, a mandatory course on teaching isn’t realistic, but a short workshop or even a video series that highlights the best practices for teaching would be reasonable and helpful. Instead, a good time to provide teaching experience may be at the postdoctoral stage, since that’s when a lot of individuals consider their careers. If they want to pursue a faculty position, they should complete a teaching certificate or other training. 

These interviews have been edited for length and clarity.

What do you think? Should teaching training be required for faculty roles? 

Submit Your Answer

Cartoon of a scientist holding his hand as a glass tube explodes. A professor comes running into the room to help.

The Perils of a High-Pressure Experiment

When Carl Hanson accidentally reinvented high-performance liquid chromatography, he landed in the hospital. 

Image Credit:

Modified from ©, The img; ©, Five Stars, lemono; designed by ashleigh campsall

Back in 1967, I was a graduate student in the laboratory of biophysicist Philip Hanawalt at Stanford University. For my thesis research, I studied the properties of the weak bonds that hold nucleic acids together. When DNA winds and unwinds, hydrogen atoms in the bonds break away and exchange with those in the surrounding water. 

          Photo of Carl Hanson.
Carl Hanson recently retired from the California Department of Public Health, where he spent the last 46 years working on a number of projects, including developing virus-neutralization assays.
Peter Patiris

To study these kinetics, I routinely placed nucleic acids in radioactive water and added the slurry to a liquid chromatography column. Inside, porous beads caught tiny water molecules but allowed large nucleic acid molecules to filter through to ordinary water, where new, nonradioactive, bonds formed.

This process normally took a few minutes, but these reactions are fast, and I wanted to try and capture what was happening to the molecules after seconds. One day, I applied pressure to the column to compress the timescale. It worked! I kept cranking up the pressure to see how quickly I could filter the nucleic acids. However, I pushed things too far.

A loud bang echoed throughout the laboratory as the glass column exploded. I was in a state of shock when my professor burst into the room, quickly clamping his hand over mine to control the bleeding. In a whirlwind, he calmly but quickly leapt into action and rushed me to the emergency room, where doctors removed bits of glass from my hand. 

I eventually collected the data I wanted after I recovered from my injury.1 Later on, I realized that I had reinvented high-performance liquid chromatography (HPLC), which scientists were developing elsewhere in the world around the same time. However, I learned the hard way why real HPLC columns are made from metal! It took nearly 40 years for all the tiny shards of glass to work their way to the surface and left me with a battle scar that served as a long-lasting reminder of the importance of safety in the laboratory. 

Share your best worst mistake with us.

Submit Your Story

<style type="text/css" >p.p1 {margin: 0.0px 0.0px 0.0px 0.0px; font: 13.8px Helvetica; color: #000000}</style>The image illustrates the interaction between two bacterial species (shown in purple and green) found in the nose microbiota.

Mining Antimicrobials in the Nose

A new antimicrobial isolated from commensal bacteria may help keep their competitors in the nasal microbiota at bay.

Image Credit:

modified from ©, JakeOlimb; designed by ashleigh campsall

Nasal cavities are an underappreciated body part, but they are a gold mine for University of Tübingen microbiologist Bernhard Krismer, who explores the bacterial interactions in this ecosystem. 

     A laboratory glass container holds a white substance.
Researchers identified a novel antimicrobial compound called epifadin, which requires special storage conditions due to its unstable nature. 
Jonas Ritz, University of Tübingen

Among the nose residents, the potentially pathogenic bacteria Staphylococcus aureus (S. aureus) piqued Krismer’s attention. S. aureus does not colonize people’s nares equally, and the presence of natural bacterial competitors could help explain these differences.1 

Recently, Krismer and his colleagues discovered a new antimicrobial synthesized by a strain of the nasal commensal Staphylococcus epidermidis (S. epidermidis).2 The molecule eliminated S. aureus, revealing a novel strategy that S. epidermidis may use to outcompete S. aureus in the nares.

In their quest for nasal antimicrobials, Krismer and his team first isolated bacteria from nasal swabs from healthy volunteers. They found that one of the isolates, a strain of S. epidermidis, produced an unknown compound, epifadin, which inhibited S. aureus proliferation in vitro and in nasal colonization experiments in rats.

After confirming that epifadin lysed S. aureus cells in vitro, the researchers set out to characterize the molecule. When they attempted to purify the antimicrobial using a strategy that had worked with another antimicrobial compound, they failed.3 “[Epifadin] is highly active; it's very potent. But unfortunately, it’s very unstable,” noted Krismer. 

After two years, Krismer’s team successfully purified and solved epifadin’s chemical structure, revealing a chimeric peptide-polyene-tetramic acid structure, which is unlike any previously described antimicrobial compound. 

Identifying a short-lived antimicrobial is an unexpected discovery according to Kim Lewis, a molecular microbiologist at Northeastern University who was not involved in the research. “That is contrary to what we generally know about production of antimicrobials that tend to be stable,” he explained. 

Next, Krismer plans to keep exploring the nasal microbiota to uncover novel molecules that mediate bacterial interactions. “There is more out there,” he said. 

  1. van Belkum A, et al. J Infect Dis. 2009;199(12):1820-1826
  2. Torres Salazar BO, et al. Nat Microbiol. 2024;9(1):200-213
  3. Zipperer A, et al. Nature. 2016;535(7613):511-516.
Image of retinal organoid showing blue cones in cyan and green/red cones in green. Rod cells are marked in magenta.

Cracking the Color Cone-undrum in Human Vision

Cell color perception fates are determined by a signaling mechanism, not chance, during retinal development.

Image Credit:

Sarah Hadyniak

Humans rely on sight, which is primarily mediated by three color-sensing cone types, to perceive the world in a kaleidoscope of hues. Blue cones develop earliest, followed by the morphologically indistinguishable green and red cones.Researchers have debated if green and red cone determination is random or deliberate.

          Headshot of Sarah Hadyniak
Sarah Hadyniak is a postdoctoral researcher at Duke University, where she studies retinas.
Sarah Hadyniak

In a recent study, Sarah Hadyniak, a developmental biologist at Duke University, explored the mechanisms that drive green and red cone generation.2 Understanding cone fate signaling could help researchers with modeling retinal systems and advancing therapies for vision impairments.

Green cones detect medium (M) wavelengths, while red cones detect long (L) wavelengths of light. The only known difference between M and L cones is the expression of opsin. Hadyniak’s team used colorimetric in situ hybridization probes to distinguish ­between M-opsin and L-opsin mRNA expression. 

“[Our system] was a big milestone in being able to see what was going on with the cones,” said Hadyniak. 

The team assessed the timing of M and L cone development in human fetal retinas. The fetal retinas expressed M-opsin before L-opsin. Adult retinas expressed both opsinsindicating that green cones developed first. 

Next, the researchers examined the role of retinoic acid, which is known to influence photoreceptor development in zebrafish. The team noted high expression of RA synthesis genes early in human retinal organoid development that decreased over time, similar to what occurs in zebrafish.

To test whether RA promoted M or L cones, Hadyniak generated human retinal organoids and administered RA over various differentiation timeframes. Early administration of RA yielded organoids with almost exclusively M cones, while late RA infusion predominantly produced L cones. 

“Cones are particularly interesting as the cell type that matters most for our daily operation in color vision,” said Rui Chen, a molecular geneticist at Baylor College of Medicine who was not involved in the study. “[This study] is informative for understanding the developmental mechanism.”

Small <em >Arabidopsis</em> seedlings are grown indoors.

Bioengineering Interkingdom Communication

Genetically edited bacteria sense the environment and report their findings to “listening” plants.

Image Credit:

©, pkujiahe

          A fluorescent microscopy image of a root.
Roots of bioengineered plants sense and respond to microbial signals.
Alice Boo (MIT).

At the macroscopic level, plants do not appear terribly active, but below the ground is a maelstrom of microscopic activity, as root systems send and receive a diverse array of chemical signals from bacteria and fungi in the soil. 

While this bidirectional communication seems to be crucial for plant health, the exact identities and functions of these chemical messengers are not well understood. To circumvent this problem, researchers at the Massachusetts Institute of Technology led by synthetic biologist Christopher Voigt bioengineered a new communication channel between microbes and plants. Published recently in Nature Communications, their work demonstrates that environmental sensors can be “outsourced” to bacteria to create a plant-microbe team that can rapidly respond to threats.1

Researchers engineered bacteria to produce a signaling molecule called molecule p-coumaroyl-homoserine lactone (pC-HSL) in response to different chemical stimuli. They also engineered plants—first the model species Arabidopsis thaliana, and then the potato—to turn on a fluorescent signal when they detected pC-HSL. In this way, researchers could change the environmental stimuli that a plant responds to simply by pairing it with different types of sensor bacteria. The bacteria could even be engineered to perform simple signal processing, integrating information from two different sensor types.

          a headshot of Christopher Voigt against a white background.
Christopher Voigt engineers genetic circuits with applications in agriculture, therapeutics, and living materials.
Christopher Voigt

Plants producing a fluorescent signal is not the end goal of this project, however, said Voigt. “I want to do sense and respond for stresses and pathogens,” he said. “Instead of trying to engineer the plant to solve a problem, it sends the signal to the bacteria, and the bacteria solve it for them. Or the bacteria sense something in the soil and send the signal that this has to be dealt with to the plant.”

Eriko Takano, a synthetic biologist at the Manchester Institute of Biotechnology who was not involved in the study, believes that these technologies could reduce reliance on herbicides, fungicides, or insecticides by providing an alternative way for plants and their associated microbes to respond to threats.  

“It's very nice work,” she said. “It's really [taking steps] towards making healthier soil without using pesticides.”

Concept illustration of the placenta

Shifting Parturition Perspectives in Perinatology Research

Nardhy Gómez-López investigates the placental immunology of preterm birth.

Image Credit:

©, stereohype

As a maternal-fetal immunologist at the Washington University School of Medicine, Nardhy Gómez-López investigates the immunobiological pathways that underlie pregnancy complications. Having trained and researched across the globe alongside caring and curious physicians, Gómez-López became hooked on perinatal immunology research. Motivated to help solve the prevalent problem of preterm birth, she currently looks at labor and birth, also called parturition, through a basic research lens.

In this Science Philosophy in a Flash podcast episode, Deanna MacNeil spoke with Gómez-López to learn more about her recent work investigating cellular changes in the maternal-fetal interface during parturition and what motivates her clinically collaborative research perspective.

          Nardhy Gomez-Lopez
Nardhy Gomez-Lopez, PhD
Obstetrics and Gynecology
Pathology and Immunology
Center for Reproductive Health Sciences
Washington University School of Medicine 
June 2024 Digest Crossword

Science Crossword Puzzle

Put on your thinking cap, and take on this fun challenge.

Image Credit: