Researchers from Sciensano, partner of the DARWIN project, have published a new paper in npj Science of Food addressing one of the key scientific and regulatory challenges linked to genome-edited (GE) organisms, their reliable detection and identification in complex food mixtures.

In the European Union, GMOs in the food chain are currently regulated under Regulations (EC) No. 1829/2003 and No. 1830/2003 to ensure safety, traceability, and consumer freedom of choice. Since 2018, genome-edited organisms have also fallen under this legislation. However, detecting and unambiguously identifying these organisms remains particularly challenging, as they can differ from their wild-type counterparts by only one or a few single-nucleotide variations (SNVs).

The publication investigates for the first time the use of high-throughput sequencing combined with adaptive sampling (AS) to selectively enrich a target species of interest in food mixtures. This approach aims to reduce matrix complexity and subsequently facilitate the detection and identification of GE lines.

As a proof-of-concept, the researchers analyzed soybean mixtures containing trace levels of genome-edited or wild-type rice using three sequencing modes: standard sequencing, adaptive sampling enriching rice, and adaptive sampling depleting soybean. The resulting sequencing data were then used to evaluate the effectiveness of, on the one hand, rice enrichment in food mixtures and, on the other hand, reliable identification of each tested rice line through the detection of their respective genetic fingerprints.

The study represents a promising first step toward facilitating the detection and identification of genome-edited organisms in the food chain, supporting future traceability and regulatory compliance efforts within the European Union framework.

Researchers at Cornell University have developed a safer and more precise way to study how genes function in living tissues by refining a recently developed CRISPR-based genetic technique in fruit flies, enabling researchers to better study how genes contribute to development and disease.

Published in the Proceedings of the National Academy of Sciences, the work highlights a new method that replaces the harsh DNA cuts used in traditional CRISPR analysis with gentler cuts known as “nicks.”

According to Chun Han, associate professor in the Department of Molecular Biology and Genetics in the College of Agriculture and Life Sciences (CALS) and the Weill Institute for Cell and Molecular Biology, the approach still allows scientists to study how genes function in living tissues, but with far less unintended cellular damage and greater control over the experiment.

CRISPR is a gene-editing technology that allows scientists to precisely cut DNA to study how genes function. One method built around CRISPR, called MAGIC, creates small groups of altered cells inside an otherwise normal organism. Researchers can then observe how specific genes affect development, cell behavior, and disease.

The original MAGIC technique relies on creating double-strand breaks, which are cuts made through both strands of DNA, to trigger the genetic changes needed for the analysis.

“We mainly wanted to improve the MAGIC technique by finding a way to avoid the toxicity associated with Cas9, which acts as the ‘molecular scissors’ in the CRISPR gene-editing system,” said Han, who led the study with Cornell doctoral student Yifan Shen and undergraduate student Ann Yeung, who is now a doctoral student at Harvard.

The original MAGIC system uses the CRISPR enzyme Cas9 to cut both strands of DNA to induce recombination between homologous chromosomes, chromosome pairs inherited from each parent, in developing cells. The recombination helps create homozygous cells (cells containing two identical copies of a gene derived from a single parent) that researchers need to study gene function, but the DNA breaks can also unintentionally rearrange chromosomes.

“In our original design of MAGIC, we used Cas9 to induce double-strand breaks (DSBs) in developing cells,” Han said. “But those DNA breaks can be highly detrimental to chromosomes during cell division, harming or killing cells.”

To address the problem, the researchers turned to “nickases,” modified versions of Cas9 that cut only one strand of DNA instead of both.

“Nickases are derived from Cas9 but carry mutations that allow them to cut only one strand of DNA,” Han said. “Without producing double-strand breaks, nickases do not damage cells in the same way and can still be used with the MAGIC technique.”

One of the study’s most unexpected findings was that even a single DNA nick could trigger the recombination needed for the MAGIC technique to work, Han said.

The researchers also found that the exact pattern of DNA nicking strongly influenced how often recombination of DNA occurred, offering scientists new ways to tune experiments for different research needs.

The advance could help researchers study genes with greater confidence by reducing the risk that experimental damage itself alters cell behavior. Cleaner, more reliable genetic tools could help researchers better study how genes contribute to development and disease, Han said.

“Our ability to study biology is restrained by the limit of our tools,” he said. “Avoiding the unintended DNA damage can make researchers more confident in using this technique and interpreting their results.”

Han said the new nickase-based system, combined with a recently developed genome-wide MAGIC toolkit, could expand use of the technique across the fruit fly research community and eventually beyond it.

“Drosophila, the fruit fly, is often the birthplace of new genetic techniques,” Han said. “But many genetic techniques invented in Drosophila later found their ways into other organisms.”

Researchers from Sciensano, partner of the DARWIN project, have published a new paper in npj Science of Food addressing one of the key scientific and regulatory challenges linked to genome-edited (GE) organisms, their reliable detection and identification in complex food mixtures.

In the European Union, GMOs in the food chain are currently regulated under Regulations (EC) No. 1829/2003 and No. 1830/2003 to ensure safety, traceability, and consumer freedom of choice. Since 2018, genome-edited organisms have also fallen under this legislation. However, detecting and unambiguously identifying these organisms remains particularly challenging, as they can differ from their wild-type counterparts by only one or a few single-nucleotide variations (SNVs).

The publication investigates for the first time the use of high-throughput sequencing combined with adaptive sampling (AS) to selectively enrich a target species of interest in food mixtures. This approach aims to reduce matrix complexity and subsequently facilitate the detection and identification of GE lines.

As a proof-of-concept, the researchers analyzed soybean mixtures containing trace levels of genome-edited or wild-type rice using three sequencing modes: standard sequencing, adaptive sampling enriching rice, and adaptive sampling depleting soybean. The resulting sequencing data were then used to evaluate the effectiveness of, on the one hand, rice enrichment in food mixtures and, on the other hand, reliable identification of each tested rice line through the detection of their respective genetic fingerprints.

The study represents a promising first step toward facilitating the detection and identification of genome-edited organisms in the food chain, supporting future traceability and regulatory compliance efforts within the European Union framework.

Approximately 75% of marine organisms are bioluminescent, with specialized light-emitting organs called photophores. They use the light they produce for various purposes, like attracting mates, luring prey, or confusing predators.

How guanine structures shape the glow

Bioluminescent fish also have specialized crystalline structures called guanine platelets that play a key role in how their light shines. While all bioluminescent fish have photophores and platelets, the number, location, and shape of these biological structures vary in different fish.

In a paper published in Biointerphases, a researcher from Hiroshima University closely examined the light-emitting organ in a deep-sea fish called the slender fangjaw (Sigmops gracilis) to reveal layers of localized guanine platelets that do more than just reflect the light—they scatter the light in complex ways.

“While examining deep-sea fish on board a research vessel, I realized important insights could not be obtained using only laboratory-based materials,” author Masakazu Iwasaka said. “This experience led me to explore a new direction—biomimetics inspired by unknown phenomena observed in the field.”

Iwasaka has been researching guanine crystals in fish for 20 years, and he hypothesized they may play an important role in bioluminescence.

“Both my own observations and previous studies have shown that guanine crystals can form layers on the surfaces of photophores in some fish species,” Iwasaka said.

“In this study, I confirmed strong anisotropic reflection—meaning the reflected light changes significantly depending on the direction the light comes from. This suggests a previously unrecognized role guanine crystals play in controlling light direction.”

From tiny mirrors to living prisms

The guanine platelets Iwasaka examined on the slender fangjaw are needle-shaped structures clustered locally around its light organs. When light hits the guanine crystals, their shape causes light scattering.

“In earlier work, I showed that guanine crystals from goldfish act like tiny mirrors, producing anisotropic reflection due to their slightly tilted orientation,” Iwasaka said. “In contrast, the higher-aspect-ratio crystals studied here behave more like prisms, redirecting light rather than simply reflecting it. Their layered arrangement exhibits properties similar to photonic crystals.”

The layered crystalline guanine platelets provide insights into highly efficient biomimetic designs that maximize and recycle leaked light, rather than just reflecting emitted light.

Potential for future biomimetic technologies

Iwasaka used electromagnets to test different orientations of the guanine crystals and exposed them to an external light source to record the scattering results of different light angles. Since these tiny structures perform in water, insights from the study could be useful in implanted biomedical device design.

“While deep-sea fish are difficult to obtain, the research is highly worthwhile,” Iwasaka said. “Investigating guanine in various fish species will lead to a treasure trove of biomimetics knowledge.”

Many drug and antibody discovery pathways focus on intricately folded cell membrane proteins. When molecules of a drug candidate bind to these proteins, like a key going into a lock, they trigger chemical cascades that alter cellular behavior. Understanding how proteins fold and move is therefore essential for developing drugs that interact well with their targets.

Artificial intelligence (AI) is a very useful tool to generate novel protein structures, but most systems—including Google DeepMind’s AlphaFold—focus on producing static “snapshots” of proteins. Subtle rearrangements of atoms in structures called side chains, which influence a protein’s interactions with other molecules, are not captured.

Now, scientists in EPFL’s School of Life Sciences have teamed up with data processing experts in the School of Engineering to solve this problem.

Researchers led by Patrick Barth of the Laboratory of Protein and Cell Engineering (LPCE) and Pierre Vandergheynst of the Signal Processing Laboratory (LTS2) have developed an AI-based generative framework called Latent Diffusion for Full Protein Generation (LD-FPG), which produces complete, all-atom structural ensembles of proteins and their movements.

“Proteins are like tiny machines that dance and switch on and off to work, but generating this ‘movie’ in full detail has been an unsolved challenge,” says LPCE researcher Aditya Sengar.

“Our LD-FPG framework is the first to do this. Instead of trying to predict the exact coordinates of atoms in space, our model learns a low-dimensional map of the protein’s shape changes. This conceptual shift is what makes generating all-atom dynamics possible.”

The new framework can notably generate the full range of motion for complex drug targets like G-protein coupled receptors (GPCRs), a focus of the global drug development industry.

“LD-FPG opens the door to designing new medicines that target a protein’s dynamic behavior, not just its shape. Our work represents a new paradigm for computational biology, and a meaningful step forward at the interface of AI and structural biology,” says Barth.

The work has been published in the Proceedings of NeurIPS 2025.

Capturing a protein’s dance

Because systems like AlphaFold use AI to predict the spatial position of every atom in a protein, they require vast amounts of computing power and biology and computer science expertise.

LD-FPG simplifies this problem using something called a graph neural network (GNN). The GNN treats each protein like a mathematical graph, where atoms represent “nodes” and the bonds between them represent “edges.” Using this low-level representation, it essentially compresses protein structure data into a simplified, or latent, map.

Next, an AI model studies this map and “learns” the representations of the protein’s structure and movements. Once trained, the model generates latent data for entirely new structures. Finally, these simplified data are converted back into high-resolution proteins—complete with side chains and dynamic movements.

In one experiment, the team generated high-fidelity, dynamic representations of the dopamine D2 receptor in both its active and inactive states. This protein detects the neurotransmitter dopamine and controls key cellular responses, making it one of the most-studied GPCRs. The researchers have published this dataset with open access to facilitate further research.

“In addition to enhancing biological understanding, we believe our work will help improve virtual screening processes for proteins, which currently involve a lot of trial and error, thereby accelerating drug discovery,” Sengar says.

Going forward, the team aims to streamline the AI framework for even greater accuracy and realism, and to enable it to model larger proteins. But Vandergheynst emphasizes that high-quality data will remain the bedrock of success. “Many assume that feeding massive datasets to AI models will automatically solve scientific problems or replace researchers.

“However, much of that data is noisy or poorly evaluated. We need human scientists to produce the clean data and rigorous benchmarks AI requires, much like we need journalists to safeguard against disinformation.”