SynGFN: learning across chemical space with generative flow-based molecular discovery
SynGFN is a novel generative flow-based model that enhances molecular discovery by simulating chemical reactions. It significantly accelerates the design-make-test-analyze cycle, enabling the identification of diverse and synthesizable molecules. The model has shown potential in designing inhibitors for neuropsychiatric disorders, indicating its relevance in drug discovery. Developed by a team from Zhejiang University and Lepu Medical Technology, SynGFN could transform approaches to molecular design and optimization.
Gene therapy CM-YAPon protects the mouse heart from myocardial infarction
CM-YAPon是一种新型基因治疗方法，利用腺病毒载体在心肌细胞中激活YAP，显示出在小鼠心肌梗死模型中显著改善心脏功能的潜力。研究表明，该疗法不仅能在心肌梗死后提供保护，还能促进心肌细胞的再生，具有重要的临床应用前景。
Pfizer finalizes Metsera buy after contentious bidding war with Novo Nordisk
Pfizer完成对Metsera的收购，交易总值约100亿美元，Metsera的MET-097i即将进入III期临床试验，显示出其在肥胖治疗领域的潜力。此次收购不仅是财务交易，更是对未来医学的战略投资，反映出Pfizer在高增长治疗领域的布局。
Kyowas bet on Kura pays off as partners oral med for AML subset wins FDA approval
Kura Oncology的menin抑制剂Komzifti获得FDA批准，专门用于治疗急性髓性白血病(NPM1突变患者)。在KOMET-001临床试验中，该药物显示出21.4%的完全缓解率，且具有良好的安全性。Kyowa Kirin与Kura的合作为该药物的开发和商业化提供了强有力的支持，预计将对市场产生重要影响。
Psychedelics produce enduring behavioral effects and functional plasticity through mechanisms independent of structural plasticity
研究表明，迷幻药如迷幻蘑菇（psilocybin）通过激活5-HT2A受体，能够在行为和细胞层面产生持久的抗抑郁效果。实验结果显示，单次给药后，迷幻药在大鼠模型中持续改善抑郁症状，并通过功能性可塑性而非结构性可塑性实现这些效果。这项研究为迷幻药在抑郁症治疗中的应用提供了新的科学依据，具有重要的临床和科研价值。
Subtractive proteomic analysis to identify highly antigenic proteins and reverse vaccinology based novel vaccine development against Haemophillius parainfluenzae
本研究通过减法蛋白质组学分析，开发了一种针对H. parainfluenzae的多表位疫苗，展示了高免疫原性和稳定性。研究利用免疫信息学和逆向疫苗学方法，筛选出高免疫原性的蛋白质，构建了疫苗模型，并进行了分子对接和动力学模拟，显示出良好的结合能力和稳定性。这一创新性疫苗设计为应对H. parainfluenzae感染提供了新的策略，具有重要的生物技术应用潜力。
Quantum and molecular modelling of abrocitinib as a potential disruptor of cell polarity in colorectal cancer
本研究探讨了abrocitinib作为结直肠癌的潜在药物重定位候选者。通过量子化学计算和分子对接分析，abrocitinib展示了良好的电子特性和强结合能力，优于现有的结直肠癌药物如氟尿嘧啶和卡培他滨。研究结果为abrocitinib作为调节细胞极性信号的治疗潜力提供了初步证据，强调了其在结直肠癌治疗中的应用前景。
The consumption of onion extract tablet modulates expression of human plasma microRNAs: a randomized placebo-controlled clinical trial
本研究通过随机双盲安慰剂对照试验，探讨了洋葱提取物对人类血浆miRNA表达的影响。结果显示，miR-106b-5p、miR-339-3p和miR-181b-5p在洋葱提取物组显著升高，表明洋葱提取物可能作为饮食摄入的生物标志物。尽管样本量较小，结果为未来更大规模的研究提供了基础。
Barium titanate piezoelectric nanoparticles induce M1 polarization in mouse macrophages via ultrasound in vitro
研究表明，钛酸钡压电纳米颗粒能够通过超声刺激诱导小鼠巨噬细胞的M1极化，具有潜在的生物医学应用价值，尤其在炎症和癌症治疗方面。该研究展示了压电纳米颗粒在调节免疫细胞功能中的创新应用，可能为未来的治疗策略提供新的思路。
Colistin resistance plasmids dually enhance bacterial virulence and antibiotic resistance via surface polysaccharide biosynthesis
研究表明，携带mcr-1质粒的细菌不仅在抗生素耐药性上表现出显著增强，同时也提高了其致病性。这一机制通过激活表面多糖的合成实现，表明mcr-1质粒在细菌生理和抗药性中的重要作用。研究还发现，mcr-1和ecaR在全球范围内的广泛共存，提示其在临床环境中的重要性。这一发现为抗生素耐药性和细菌致病性之间的关系提供了新的视角，具有重要的生物技术和公共卫生意义。
HPCAL1 as a novel prognostic biomarker in acute myeloid leukemia and its correlation with immune infiltrates
HPCAL1被发现是急性髓性白血病(AML)的新型预后生物标志物，其表达水平与患者生存率显著相关。研究表明，HPCAL1的高表达与免疫微环境的变化密切相关，可能通过影响单核细胞的功能和肿瘤免疫逃逸机制来促进AML的进展。这些发现为HPCAL1作为潜在治疗靶点提供了基础，强调了其在AML管理中的重要性。
Mechanism of the traditional Chinese medicine SMBJ alleviates diabetes mellitus-induced Leydig cell dysfunction in rats testes
Simiao Biejia Decoction (SMBJ), a traditional Chinese herbal formula, shows promise in alleviating diabetes-induced Leydig cell dysfunction in rats. The study reveals that SMBJ enhances testosterone secretion and reduces oxidative stress by upregulating METTL3 and activating the PI3K/AKT signaling pathway. This research highlights SMBJ's potential as a therapeutic agent for metabolic disorders, suggesting significant implications for early-stage investment in biopharmaceuticals targeting diabetes-related complications.
The expression signature, prognostic significance and immune cell infiltration of the OAS gene family in gastric cancer
本研究发现OAS基因家族在胃癌中显著上调，并与患者预后不良相关。通过整合生物信息学分析与实验验证，研究揭示了OAS基因家族在肿瘤微环境中的作用，特别是与免疫细胞浸润的关系。OASL的上调促进了胃癌细胞的增殖和迁移，表明其作为潜在治疗靶点的价值。
Four-class classification of tumor-induced colorectal obstruction histopathology: A ResNet–mamba-mased study on cellular interaction pattern recognition
This study introduces a deep learning model for the classification of tumor-induced colorectal obstruction (TICO) lesions, achieving an 85% validation accuracy. Utilizing a ResNet-Mamba architecture, the model analyzes cellular interaction patterns in histopathological slides, demonstrating strong diagnostic potential. The research focuses on four histological categories: normal mucosa, serrated lesions, adenomas, and adenocarcinomas, providing a foundation for early diagnosis and personalized treatment strategies in oncology.
A hybrid attention network for accurate breast tumor segmentation in ultrasound images
HA-Net是一种新提出的混合注意力网络，专为乳腺超声图像中的肿瘤分割而设计。该网络结合了DenseNet-121编码器和多种注意力机制，能够有效捕捉全局上下文关系，同时保持细粒度的空间细节。通过在多个公开数据集上的广泛测试，HA-Net展现出优越的分割性能，超越了现有的主流方法，显示出其在临床应用中的潜力，能够帮助放射科医生进行早期乳腺癌诊断。