该文旨在比较DeepSeek-V3.1与ChatGPT-5在结直肠癌肝转移(CRLM)多学科团队(MDT)决策中的应用表现,评估其与MDT专家意见的一致性,为大语言模型(LLMs)的临床实践提供循证依据与优化方向。该文基于真实世界数据与最新指南,设计了6例涵盖不...该文旨在比较DeepSeek-V3.1与ChatGPT-5在结直肠癌肝转移(CRLM)多学科团队(MDT)决策中的应用表现,评估其与MDT专家意见的一致性,为大语言模型(LLMs)的临床实践提供循证依据与优化方向。该文基于真实世界数据与最新指南,设计了6例涵盖不同肿瘤负荷、基因突变谱和体能状态的虚拟CRLM病例,通过结构化的提示策略,在DeepSeek-V3.1与ChatGPT-5模型中分别生成MDT治疗建议。由4名MDT专家采用7维度5级李克特量表对模型输出进行独立评审。并通过统计学分析对两款模型在各个病例、各项维度和各个学科的表现分别进行比较。2款大语言模型在所有病例中的综合得分均≥4.0分(满分5分),表明其在复杂的MDT决策场景下具备可接受的临床效能。在跨维度分析中,两者在清晰度、个体化程度、抗幻觉能力和伦理安全4项上得分较高,而在准确性、全面性和前沿性方面仍有一定提升空间。DeepSeek-V3.1在整体表现(4.27±0.77 vs 4.08±0.86)、前沿性(3.90±0.65 vs 3.24±0.72)与伦理安全(4.87±0.34 vs 4.58±0.65)方面显著优于ChatGPT-5(P<0.05);在放疗领域亦明显领先(4.55±0.67 vs 3.38±0.91,P<0.01)。ChatGPT-5则在胃肠外科领域表现优于DeepSeek-V3.1(4.48±0.67 vs 4.17±0.85,P=0.02)。DeepSeek-V3.1与ChatGPT-5均表现出为CRLM-MDT决策提供可靠建议的良好能力。其中,Deep Seek-V3.1在前沿知识整合、伦理安全性及放射肿瘤学领域展现出显著优势,而Chat GPT-5则在胃肠外科方面表现更优,二者形成优势互补。该文证实了大型语言模型作为“MDT协作者”的可行性,为缩小地域间诊疗水平差距、提升临床决策效率提供了一项便捷可靠的技术方案。展开更多
冰裂隙是冰川动态变化的重要特征,通过研究冰裂隙,可以有效监测冰川的变化情况。然而,传统的冰裂隙遥感监测方法往往存在人工干预多、效率低下或对复杂冰川环境适应性不足等问题。为解决以上问题,以GeoScene Pro 3.1软件为平台,利用高...冰裂隙是冰川动态变化的重要特征,通过研究冰裂隙,可以有效监测冰川的变化情况。然而,传统的冰裂隙遥感监测方法往往存在人工干预多、效率低下或对复杂冰川环境适应性不足等问题。为解决以上问题,以GeoScene Pro 3.1软件为平台,利用高分辨率遥感影像,选择长江源各拉丹冬冰川区为试验区,结合深度学习技术,探讨了U-Net像素分类模型、HED边缘检测器和BDCN边缘检测器在冰裂隙检测与提取中的应用。首先,通过调整不同的训练样本集组合与训练参数,对以上3种模型分别进行了5次模型训练和冰裂隙检测分析。检测结果表明,U-Net像素分类模型和HED边缘检测器都存在检测效果不佳的问题,而BDCN边缘检测器能够有效抑制检测噪声干扰,并能准确地检测并识别出冰裂隙的分布位置和宽度。其次,利用BDCN边缘检测器对各拉丹冬4条冰川进行冰裂隙检测,冰裂隙检测率分别达到90%、92%、89%、93%,模型检测宽度与人工采样宽度的最小差值仅为0.3 m,充分证明了该模型在不同冰川区域应用时具备良好的鲁棒性和泛化能力。研究表明BDCN边缘检测器更适合冰裂隙的检测识别,可为各拉丹冬冰川及类似区域的冰裂隙周期性监测提供快速有效的技术支持。展开更多
Mechanosensitive ion channels are essential for sensing and converting mechanical forces into electrical or chemical signals.These channels are widely distributed across bacteria,animals,and plants.In Arabidopsis thal...Mechanosensitive ion channels are essential for sensing and converting mechanical forces into electrical or chemical signals.These channels are widely distributed across bacteria,animals,and plants.In Arabidopsis thaliana,the OSCA family has been identified as mechanically activated ion channels that respond to osmotic stress by allowing calcium ions to enter the cell.This influx increases the cytoplasmic calcium concentration,triggering osmotic stress-induced signal transduction cascades in plants.In this study,we determined the structures of OSCA2.2 and OSCA3.1 via cryoelectron microscopy(cryo-EM).Both proteins form homodimers consisting of 11 transmembrane helices(TM1–11).The ion conduction pathway is formed by TM4–8.Despite belonging to the same family,OSCA2.2 and OSCA3.1 exhibit notable structural variations.Structural analysis revealed that both OSCA2.2 and OSCA3.1 exhibit a closed conformation.We also conducted functional studies on OSCA proteins via electrophysiological experiments and confirmed the role of key amino acids in the process of ion permeation.展开更多
Rice,a critical global staple crop,relies heavily on heading date,a key agronomic trait marking the transition from vegetative to reproductive growth.Understanding the genetic regulation of heading date is vital for e...Rice,a critical global staple crop,relies heavily on heading date,a key agronomic trait marking the transition from vegetative to reproductive growth.Understanding the genetic regulation of heading date is vital for enhancing the adaptability of high-quality rice varieties across diverse geographical regions and for bolstering local food security.In this study,we uncovered a novel role for OsCATA,a catalase gene,in the regulation of photoperiodic flowering in rice.We identified a novel allele of OsELF3.1,whose mutation resulted in delayed heading.Further analyses revealed that OsELF3.1 physically interacted with OsCATA.Notably,OsCATA exhibited rhythmic expression patterns similar to OsELF3.1 and,when mutated,also delayed flowering.Expression analyses showed that the delayed heading phenotype could be attributed to elevated Ghd7 expression under both long-day and short-day conditions,with OsCATA expression positively regulated by OsELF3.1.Double mutants of OsELF3.1 and OsCATA displayed a heading delay similar to that of oself3.1 single mutants.Additionally,OsELF3.1 could interact with Ghd7 in vivo,alleviating its suppression of Ehd1.Luciferase assays confirmed that Ghd7 repressed Ehd1 expression,while OsELF3.1 mitigated this repression.Collectively,our findings reveal that OsCATA is critical in suppressing Ghd7 expression through the OsELF3.1-OsCATA-Ghd7 transcriptional pathway,thereby regulating rice heading.展开更多
文摘该文旨在比较DeepSeek-V3.1与ChatGPT-5在结直肠癌肝转移(CRLM)多学科团队(MDT)决策中的应用表现,评估其与MDT专家意见的一致性,为大语言模型(LLMs)的临床实践提供循证依据与优化方向。该文基于真实世界数据与最新指南,设计了6例涵盖不同肿瘤负荷、基因突变谱和体能状态的虚拟CRLM病例,通过结构化的提示策略,在DeepSeek-V3.1与ChatGPT-5模型中分别生成MDT治疗建议。由4名MDT专家采用7维度5级李克特量表对模型输出进行独立评审。并通过统计学分析对两款模型在各个病例、各项维度和各个学科的表现分别进行比较。2款大语言模型在所有病例中的综合得分均≥4.0分(满分5分),表明其在复杂的MDT决策场景下具备可接受的临床效能。在跨维度分析中,两者在清晰度、个体化程度、抗幻觉能力和伦理安全4项上得分较高,而在准确性、全面性和前沿性方面仍有一定提升空间。DeepSeek-V3.1在整体表现(4.27±0.77 vs 4.08±0.86)、前沿性(3.90±0.65 vs 3.24±0.72)与伦理安全(4.87±0.34 vs 4.58±0.65)方面显著优于ChatGPT-5(P<0.05);在放疗领域亦明显领先(4.55±0.67 vs 3.38±0.91,P<0.01)。ChatGPT-5则在胃肠外科领域表现优于DeepSeek-V3.1(4.48±0.67 vs 4.17±0.85,P=0.02)。DeepSeek-V3.1与ChatGPT-5均表现出为CRLM-MDT决策提供可靠建议的良好能力。其中,Deep Seek-V3.1在前沿知识整合、伦理安全性及放射肿瘤学领域展现出显著优势,而Chat GPT-5则在胃肠外科方面表现更优,二者形成优势互补。该文证实了大型语言模型作为“MDT协作者”的可行性,为缩小地域间诊疗水平差距、提升临床决策效率提供了一项便捷可靠的技术方案。
文摘冰裂隙是冰川动态变化的重要特征,通过研究冰裂隙,可以有效监测冰川的变化情况。然而,传统的冰裂隙遥感监测方法往往存在人工干预多、效率低下或对复杂冰川环境适应性不足等问题。为解决以上问题,以GeoScene Pro 3.1软件为平台,利用高分辨率遥感影像,选择长江源各拉丹冬冰川区为试验区,结合深度学习技术,探讨了U-Net像素分类模型、HED边缘检测器和BDCN边缘检测器在冰裂隙检测与提取中的应用。首先,通过调整不同的训练样本集组合与训练参数,对以上3种模型分别进行了5次模型训练和冰裂隙检测分析。检测结果表明,U-Net像素分类模型和HED边缘检测器都存在检测效果不佳的问题,而BDCN边缘检测器能够有效抑制检测噪声干扰,并能准确地检测并识别出冰裂隙的分布位置和宽度。其次,利用BDCN边缘检测器对各拉丹冬4条冰川进行冰裂隙检测,冰裂隙检测率分别达到90%、92%、89%、93%,模型检测宽度与人工采样宽度的最小差值仅为0.3 m,充分证明了该模型在不同冰川区域应用时具备良好的鲁棒性和泛化能力。研究表明BDCN边缘检测器更适合冰裂隙的检测识别,可为各拉丹冬冰川及类似区域的冰裂隙周期性监测提供快速有效的技术支持。
基金supported by the National Natural Science Foundation of China(32322041,W2412029,32321001,32471279)USTC Research Funds of the Double First-Class Initiative(YD9100002004,YD9100002020)+2 种基金Fundamental Research Funds for the Central Universities(WK9100000031)Research Funds of Center for Advanced Interdisciplinary Science and Biomedicine of IHM(QYPY20230034)the Natural Science Foundation of Anhui Province(2408085JX005).
文摘Mechanosensitive ion channels are essential for sensing and converting mechanical forces into electrical or chemical signals.These channels are widely distributed across bacteria,animals,and plants.In Arabidopsis thaliana,the OSCA family has been identified as mechanically activated ion channels that respond to osmotic stress by allowing calcium ions to enter the cell.This influx increases the cytoplasmic calcium concentration,triggering osmotic stress-induced signal transduction cascades in plants.In this study,we determined the structures of OSCA2.2 and OSCA3.1 via cryoelectron microscopy(cryo-EM).Both proteins form homodimers consisting of 11 transmembrane helices(TM1–11).The ion conduction pathway is formed by TM4–8.Despite belonging to the same family,OSCA2.2 and OSCA3.1 exhibit notable structural variations.Structural analysis revealed that both OSCA2.2 and OSCA3.1 exhibit a closed conformation.We also conducted functional studies on OSCA proteins via electrophysiological experiments and confirmed the role of key amino acids in the process of ion permeation.
基金funded by the Biological Breeding-National Science and Technology Major Projects,China(Grant No.2023ZD04066)the Zhejiang Provincial Natural Science Foundation of China(Grant Nos.LZ24C130006,LTGN24C130007)+5 种基金the Open Project Program of the State Key Laboratory of Rice Biology and Breeding,China(Grant No.20240107)the Xi’nan League Science and Technology Project,China(Grant No.2023DXZD0001)the Joint Research and Development Program on Rice Breeding in Inner Mongolia Autonomous Region,China(Grant No.YZ2023004)the China Agriculture Research System(Grant No.CARS-01)the Central Public-interest Scientific Institution Basal Research Fund,China(Grant No.CPSIBRF-CNRRI-202301)the Agricultural Science and Technology Innovation Program(ASTIP).
文摘Rice,a critical global staple crop,relies heavily on heading date,a key agronomic trait marking the transition from vegetative to reproductive growth.Understanding the genetic regulation of heading date is vital for enhancing the adaptability of high-quality rice varieties across diverse geographical regions and for bolstering local food security.In this study,we uncovered a novel role for OsCATA,a catalase gene,in the regulation of photoperiodic flowering in rice.We identified a novel allele of OsELF3.1,whose mutation resulted in delayed heading.Further analyses revealed that OsELF3.1 physically interacted with OsCATA.Notably,OsCATA exhibited rhythmic expression patterns similar to OsELF3.1 and,when mutated,also delayed flowering.Expression analyses showed that the delayed heading phenotype could be attributed to elevated Ghd7 expression under both long-day and short-day conditions,with OsCATA expression positively regulated by OsELF3.1.Double mutants of OsELF3.1 and OsCATA displayed a heading delay similar to that of oself3.1 single mutants.Additionally,OsELF3.1 could interact with Ghd7 in vivo,alleviating its suppression of Ehd1.Luciferase assays confirmed that Ghd7 repressed Ehd1 expression,while OsELF3.1 mitigated this repression.Collectively,our findings reveal that OsCATA is critical in suppressing Ghd7 expression through the OsELF3.1-OsCATA-Ghd7 transcriptional pathway,thereby regulating rice heading.
