Taiyi: a bilingual fine-tuned large language model for diverse biomedical tasks.

OBJECTIVE: Most existing fine-tuned biomedical large language models (LLMs) focus on enhancing performance in monolingual biomedical question answering and conversation tasks. To investigate the effectiveness of the fine-tuned LLMs on diverse biomedical natural language processing (NLP) tasks in different languages, we present Taiyi, a bilingual fine-tuned LLM for diverse biomedical NLP tasks. MATERIALS AND METHODS: We first curated a comprehensive collection of 140 existing biomedical text mining datasets (102 English and 38 Chinese datasets) across over 10 task types. Subsequently, these corpora were converted to the instruction data used to fine-tune the general LLM. During the supervised fine-tuning phase, a 2-stage strategy is proposed to optimize the model performance across various tasks. RESULTS: Experimental results on 13 test sets, which include named entity recognition, relation extraction, text classification, and question answering tasks, demonstrate that Taiyi achieves superior performance compared to general LLMs. The case study involving additional biomedical NLP tasks further shows Taiyi's considerable potential for bilingual biomedical multitasking. CONCLUSION: Leveraging rich high-quality biomedical corpora and developing effective fine-tuning strategies can significantly improve the performance of LLMs within the biomedical domain. Taiyi shows the bilingual multitasking capability through supervised fine-tuning. However, those tasks such as information extraction that are not generation tasks in nature remain challenging for LLM-based generative approaches, and they still underperform the conventional discriminative approaches using smaller language models.

A Newly Synbiotic Combination Alleviates Obesity by Modulating the Gut Microbiota-Fat Axis and Inhibiting the Hepatic TLR4/NF-κB Signaling Pathway.

SCOPE: Obesity has been recognized as a worldwide public health crisis, this is accompanied by dysregulation of the intestinal microbiota and upregulation of liver steatosis and adipose inflammation. Synbiotic as a novel alternative therapy for obesity have recently gained much attention. METHODS: This study innovatively research the anti-obesity properties of a newly synbiotic composed of Lactobacillus acidophilus, Bifidobacterium infantis and konjac glucomannan oligosaccharides. RESULTS: The synbiotic treatment can reduce body weight, fat mass, blood sugar, liver steatosis and adipose inflammation in obesity mice fed by high-fat diet (HFD). Meanwhile, synbiotic treatment activated brown adipose tissue and improve energy, glucose and lipid metabolism. In addition, synbiotic treatment not solely enhanced the protection of intestinal barrier, but also ameliorated gut microbiota dysbiosis directly by enhancing beneficial microbes and reducing potentially harmful bacteria. Furthermore, the microbiome phenotype and functional prediction showed that synbiotic treatment can improve the gut microbiota functions involving inflammatory state, immune response, metabolism and pathopoiesia. CONCLUSION: The synbiotic may be an effective candidate treatment strategy for the clinical prevention and treatment of obesity and other associated metabolic diseases such as hyperlipidemia, nonalcoholic fatty liver diseases by alleviating inflammatory response, regulating energy metabolism and maintaining the balance of intestinal microecology.

[Effects of bisphenol A on voltage-dependent T-type calcium channels in mouse testis and epididymis, and the role of estrogen receptors].

OBJECTIVE: To explore the possible effects of bisphenol A (BPA) on voltage-dependent T-type calcium channels in mouse testis and epididymis, the role of the two ERs was also investigated. METHODS: Adult male wild type (WT), ERs knock out mice (αERKO and ßERKO) were randomly divided into control and BPA [100 µg/(kg. d) for 30 days via intragastric administration] treatment groups. Quantitative real time PCR was used to examine Cav3.1, Cav3.2 and Cav3.3 genes of the testis and epididymis. RESULTS: In the testis, the mRNA expression of Cav3.1, Cav3.2 and Cav3.3 were increased in BPA-treated animals. In the epididymis, the mRNA expression of Cav3. 1 and Cav3.2 were increased, however, Cav3.3 were decreased in BPA-treated animals. BPA up-regulation of Cav3.1 and Cav3.2 mRNA expression of the epididymis were dependent on both ERα and ERß, whereas BPA regulation of Cav 3.1 - Cav3.3 of the testis and Cav3.3 of the epididymis appeared to be dependent on ERß only. CONCLUSION: It suggested that the ERs signalling was necessary for the regulation of the Cav3 channels mRNA expression by BPA, which could affect the male reproductive function.