Chinese Academy of Sciences

A new sunflower family tree used skimmed genomes to increase the number of species sampled, revealing that flower symmetry evolved multiple times independently, a process called convergent evolution, among the members of this large plant family. The sunflower family tree revealed that flower symmetry evolved multiple times independently, a process called convergent evolution, among the members of this large plant family, according to a new analysis. The research team, led by a Penn State biologist, resolved more of the finer branches of the family tree, providing insight into how the sunflower family -- which includes asters, daisies and food crops like lettuce and artichoke -- evolved. A paper describing the analysis and findings, which researchers said may help identify useful traits to selectively breed plants with more desirable characteristics, appeared online in the journal Plant Communication. "Convergent evolution describes the independent evolution of what appears to be the same trait in different species, like wings in birds and bats," said Hong Ma, Huck Chair in Plant Reproductive Development and Evolution, professor of biology in the Eberly College of Science at Penn State and the leader of the research team. "This can make it difficult to determine how closely related two species are by comparing their traits, so having a detailed family tree based on DNA sequence is crucial to understanding how and when these traits evolved." The sunflower head, for example, is actually a composite composed of multiple much smaller flowers. While the head is generally radially symmetric -- it can be divided into two equal halves in multiple directions like a starfish or a pie -- the individual flowers can have different forms of symmetry. According to the new study, bilateral symmetry -- where there is only one line that divides the flower into two equal halves -- has evolved and been lost multiple times independently in sunflowers over evolutionary history. The researchers found that this convergent evolution is likely related to changes in the number of copies and the expression patterns of the floral regulatory gene, CYC2. In recent years, many family trees for a group of related species have been built by extensively using transcriptomes, which are the genetic sequences of essentially all of the genes expressed by a species, the researchers explained. Transcriptomes are easier to acquire than high-quality whole-genome sequences for a species but are still difficult and costly to prepare and require fresh plant samples. To increase the number of species available for comparison the team turned to low-coverage genome sequences, which are produced through a process called genome skimming and are relatively inexpensive and easy to prepare, even from dried plant samples. "To get an accurate whole-genome sequence for a species, each letter of its DNA alphabet must be read -- or covered -- multiple times to minimize errors," Ma said. "For the purposes of building a family tree, we show in this paper that we can get away with lower coverage genome sequences. This allowed us to increase the number of species in our analysis, which, in turn, allowed us to resolve more of the finer branches on the sunflower family tree." The team used a combination of publicly available and newly generated transcriptomes, along with a large number of newly obtained skimmed genomes, for a total of 706 species with representatives from 16 subfamilies, 41 tribes and 144 subtribe-level groups in the sunflower family. The subfamilies are major subdivisions of the family, while the tribes and subtribe can contain one or more of genera, which is the classification level just above the species. "Previous versions of the sunflower family tree had established the relationships among most of the subfamilies and many tribes, which are equivalent to the main branches of a tree," Ma said. "With our increased sample size, we were able to resolve more of the smaller branches and twigs at the subtribe and genus level. This higher-resolution tree allowed us to reconstruct where and when traits like flower symmetry evolved, demonstrating that bilateral symmetry must have evolved many times independently." The team also studied the molecular evolution of genes involved in flower development among sunflowers. They found that one of these genes, CYC2, which is found in multiple copies in the genomes of each species, was activated in species with bilaterally symmetric flowers, suggesting that it might be part of the molecular basis for the convergent evolution of this trait. To further test this, the team performed experiments to quantify CYC2 gene expression in the flowers of species with different types of symmetry. "Our analysis showed a clear relationship between CYC2 expression and flower symmetry, suggesting that changes in how these genes are used in various sunflower species is likely involved in the convergent evolution observed in the family," Ma said. "The sunflower family is one of the two largest families of flowering plants containing over 28,000 species, including many economically important agricultural and horticultural species. Understanding how these species are related to one another allows us to determine how and when their traits evolved. This knowledge could also be used to identify useful traits that could be bred into domesticated species from closely related wild ones." In addition to Ma, the research team includes Guojin Zhang at Penn State; Junbo Yang, Jie Cai, Zhi-Rong Zhang and Lian-Ming Gao at the Kunming Institute of Botany in Kunming, China; Caifei Zhang at the Wuhan Botanical Garden and Sino-Africa Joint Research Centre in Wuhan, China; Bohan Jiao and Tiangang Gao at the State Key Laboratory of Plant Diversity and Specialty Crops in Beijing, China; and Jose L. Panero at the University of Texas, Austin. Funding from the Eberly College of Science and the Huck Institutes of the Life Sciences at Penn State, the Strategic Priority Research Program of the Chinese Academy of Sciences, the Large-scale Scientific Facilities of the Chinese Academy of Sciences, and the National Natural Science Foundation of China supported this research.

A recent study reveals an unexpected link between manganese -- a naturally occurring mineral -- and increased carbon emissions from high-latitude boreal forests. The study uncovers how manganese increases carbon emissions from boreal forest soil, over time, challenging conventional understanding of the role of manganese in the carbon cycle. Manganese in the soil of boreal forests has been found to work against the carbon storage capacity of these crucial northern habitats. Located predominantly in cold regions at high latitude, boreal forests are estimated to store nearly 30 percent of the world's soil carbon, making them the world's largest reservoir of land-based carbon. This stored carbon is found mostly in the forests' humus layer, which contains decomposed leaves and other organic matter. A global, long-term study led by Duke University researchers has found that higher levels of manganese in this layer stimulated decomposition of soil organic matter, and released more carbon dioxide than did those forest plots with less or no manganese. The work appeared March 19 in the Proceedings of the National Academy of Sciences (PNAS). "Conventional wisdom is these forests are like a global vault of carbon, where carbon is put into the vault versus taken out," said William H. Schlesinger, professor emeritus at Duke University's Nicholas School of the Environment and study co-author. "These findings reveal a crack in the vault, where enough manganese over time stimulates the release of carbon dioxide into the atmosphere, which has implications for climate mitigation efforts and the global carbon cycle," said Schlesinger. Certain industrial processes, such as metal smelting or combustion of manganese-containing fuels, can release airborne manganese which is later deposited in soils downwind. This is one of many human activities, such as the burning of fossil fuels, deforestation, and land-use changes, that have disturbed the natural carbon cycle, leading to an increase in atmospheric carbon dioxide concentrations that contribute to global warming and climate change. "Carbon inventorying is still an evolving science," said Yunyu Zhang, lead author and graduate student from the Chinese Academy of Sciences. "It is essential to figure out which factors regulate this huge carbon pool [in boreal forests' soil], especially given continuous… industrialization." Researchers analyzed data from boreal forests worldwide, and fertilized soil with manganese over 14 years (2009-2023) in China's Daxing'an Mountains. Results showed the level of exchangeable manganese -- the part of manganese that plants can use as nutrients -- determined how much carbon was stored in boreal forest soil. After four years, carbon storage on plots fertilized with manganese fell by nearly 13 percent, meaning more carbon was released into the atmosphere. "To develop effective and sustainable strategies, it is critical to understand complex interactions between trace nutrients and carbon storage," Zhang said. "It is even more important to predict how those interactions work in the long term, considering the impact of human activities." Schlesinger emphasized the need for further research and action, noting how the study's findings highlight the importance of soil nutrient dynamics, such as the level of exchangeable manganese, in climate change mitigation efforts. He urged further study of the role of manganese not only in soil, but also in the air, on land-based carbon emissions, the boreal forest ecosystem, and climate mitigation. "There's no proverbial foolproof vault or absolute forest sink," Schlesinger said. "We need integrated approaches to land management and climate mitigation. Climate has traditionally been considered a major factor in carbon storage, but we now see how manganese is also a key indicator, something that has long been overlooked and underexamined." The National Natural Science Foundation of China and Chinese Academy of Sciences co-funded the study.

Hong Kong-based Centre for Artificial Intelligence and Robotics (CAIR) under the Chinese Academy of Sciences (CAS), one of China's national research institutes, has developed a large language model-based chatbot to assist doctors in making medical diagnoses and treatments. HOW IT WORKS The chatbot, CARES Copilot 1.0, is built based on Meta's Llama 2 LLM. Trained using 100 GPUs from Huawei and Nvidia, it can process various data, including images, text, voice, video, MRI, CT scans, and ultrasound. Fed with millions of records, including teaching materials, expert guidelines, and medical literature, CARES can retrieve information "within a few seconds," with up to 95% accuracy, CAIR claims. CARES, equipped with a 100K context window, can also pull out information from over 3,000 pages of surgical materials. Based on internal testing, it can support functions such as surgical phase identification, segmentation of instruments and anatomical structures, instrument detection and counting, and generation of MRI high-resolution images. The assistant AI tool, which can be integrated with medical devices, is now undergoing continuous optimisation in seven hospitals in Beijing. THE LARGER TREND China is playing catch up with global innovators in developing AI technologies by funding local initiatives. CAIR's latest AI development is one of those projects co-funded by the Hong Kong government's InnoHK research programme. In August last year, the Chinese government approved the rollout of the first homegrown genAI-enabled service, the TaiChu model, also by CAS. ON THE RECORD "[LLMs], tailored for neuroscience, can surpass traditional teaching needs, enabling applications in clinical settings, operating rooms, and research institutes in conjunction with surgery, image guidance, and robotics. These models directly support frontline medical staff in emergencies by supervising, issuing warnings, preventing risky procedures, and pushing the boundaries of neuroscience. The combination of LLMs and surgical navigation can offer real-time anatomical positioning information to surgeons, enhancing surgical safety," Ming Feng, professor and chief physician at the Department of Neurosurgery, Peking Union Medical College Hospital, commented on the release of CARES.