Search

Jianxia Ma,* Ke Wang,* Jun Wang, Qinlian Zeng, Kangwei Liu, Songbai Zheng, Yuanwen Chen, Jianfeng Yao Department of Gastroenterology, Hua Dong Hospital of Fu Dan University, Shanghai, 200040, People’s Republic of China*These authors contributed equally to this workCorrespondence: Jianfeng Yao, Department of Gastroenterology, HuaDong Hospital of Fu Dan University, No. 221. Road West Yanan, Shanghai, 200040, People’s Republic of China, Tel +8602162483180, Email yjf1966@126.comObjective: The fecal microbiota was studied in patients with inflammatory bowel disease (IBD), and the characteristics of gut microbiota were compared among patients with different subtypes and stages of IBD, aiming to identify the gut microbiota associated with IBD.Methods: Fecal samples were collected from 41 IBD patients (18 patients with ulcerative colitis [UC] and 23 patients with Crohn’s disease [CD]) in the Department of Gastroenterology of East China Hospital, Fudan University between January 2021 and January 2022. In addition, fecal samples were collected from 20 healthy volunteers. The fecal microbiota was subjected to 16S rRNA gene sequencing, followed by bioinformatics analysis.Results: There was significant difference in the fecal microbiota between IBD patients and controls. The abundance and diversity of fecal microbiota in the IBD patients were significantly lower than in controls. The relative abundance of Subdoligranulum, Ruminococcus, Anaerostipes and Lachnospira was reduced markedly in the IBD patients. As compared to controls, the relative abundance of Streptococcus increased dramatically in the UC patients. The relative abundance of Lachnoclostridium, Fusobacterium, Cloacibacillus and Erysipelatoclostridium significantly increased in the CD patients. As compared to CD patients, the relative abundance of Alistipes was reduced markedly in the UC patients; the relative abundance of Faecalibacterium, Roseburia and Haemophilus was reduced dramaticall

Robotic agents must master common sense and long-term sequential decisions to solve daily tasks through natural language instruction. The developments in Large Language Models (LLMs) in natural language processing have inspired efforts to use LLMs in complex robot planning. Despite LLMs' great generalization and comprehension of instruction tasks, LLMs-generated task plans sometimes lack feasibility and correctness. To address the problem, we propose a RoboGPT agent\footnote{our code and dataset will be released soon} for making embodied long-term decisions for daily tasks, with two modules: 1) LLMs-based planning with re-plan to break the task into multiple sub-goals; 2) RoboSkill individually designed for sub-goals to learn better navigation and manipulation skills. The LLMs-based planning is enhanced with a new robotic dataset and re-plan, called RoboGPT. The new robotic dataset of 67k daily instruction tasks is gathered for fine-tuning the Llama model and obtaining RoboGPT. RoboGPT planner with strong generalization can plan hundreds of daily instruction tasks. Additionally, a low-computational Re-Plan module is designed to allow plans to flexibly adapt to the environment, thereby addressing the nomenclature diversity challenge. The proposed RoboGPT agent outperforms SOTA methods on the ALFRED daily tasks. Moreover, RoboGPT planner exceeds SOTA LLM-based planners like ChatGPT in task-planning rationality for hundreds of unseen daily tasks, and even other domain tasks, while keeping the large model's original broad application and generality.

Hongyi Zhu,* Yuanwen Chen,* Jie Zhang, Changlin Qian, Weiqing Qiu, Huojian Shen, Zhiyong Shen Department of General Surgery, South Campus, Renji Hospital, School of Medicine, Shanghai Jiao Tong University, Shanghai 201112, People’s Republic of China*These authors contributed equally to this workCorrespondence: Huojian Shen; Zhiyong ShenDepartment of General Surgery, South Campus, Renji Hospital, School of Medicine, Shanghai Jiao Tong University, 2000 JiangYue Road, Minhang District, Shanghai 201112, People’s Republic of ChinaTel +8621-58752345Email 2643500550@qq.com; doctor_szy7089@163.comObjective: Gastric cancer (GC), a malignant tumor of the gastric mucosa, is the second leading cause of cancer deaths worldwide. Although the incidence and mortality of gastric cancer have been reduced in the US and elsewhere, it is still a major public health concern. In this study, we attempted to investigate the function of tripartite motif-containing protein 37 (TRIM37) in GC cell lines in order to propose a new therapy for GC.Methods: The expression of TRIM37 in GC patients and cell lines was detected by immunohistochemistry, real-time PCR and Western blotting analysis. After TRIM37 knockdown or overexpression, the cell cycle, proliferation and apoptosis, as well as the expression of related proteins, were detected. In addition, in vivo experiments on nude mice were performed.Results: We found that TRIM37 expression was significantly elevated in tumor tissues of GC patients and GC cell lines, and patients with high expression of TRIM37 had a poor prognosis. Knockdown of TRIM37 in GC cells significantly inhibited cell proliferation and cell cycle progression, promoted apoptosis, increased cleaved caspase 3 and decreased c-myc and phosphorylation of protein kinase 1/2 (p-ERK1/2). Effects of TRIM37 overexpression were opposite to that of TRIM37 knockdown and were potently attenuated by an ERK1/2 inhibitor. In addition, an ERK1/2 agonist increased TRIM37 and p