目的: 探索肠道微生物在首发未用药精神分裂症患者与健康对照者间的差异，多维度纵向分析探索抗精神病药物(antipsychotic drugs, APDs)治疗后肠道微生物与临床表型及血清代谢组学的关联。方法: 选择2017年6—12月于北京大学第六医院门诊就诊的28例首发未用药的精神分裂症患者(病例组)及同期性别、年龄、受教育程度相匹配的29例健康志愿者(对照组)为研究对象，其中病例组接受了APDs治疗。研究采集病例组入组基线和治疗6周后的粪便和血清样本，并对粪便样本微生物种类和临床症状及血清代谢物进行了检测和分析。结果: 菌群多样性分析发现，病例组alpha多样性指数(chao1、ACE、goods_coverage)低于对照组，并且组间差异有统计学意义，病例组与对照组在beta多样性方面有明显的区分。菌群组成分析结果显示Bacteroides、Streptococcus、Romboutsia、Eubacterium ruminantium group在病例组患者接受治疗前后发生变化，且差异具有统计学意义，这些菌群可能反映了APDs治疗的影响。病例组与对照组相比，有更多的微生物种属发生变化。LEfSe种群丰度差异分析显示Prevotel-la_9和Bacteroides在病例组富集，而Blautia、Dialister、Roseburia在对照组富集。肠道微生物与临床症状的关联分析显示，Bifidobacterium在病例组患者中与PANSS (positive and negative syndrome scale)量表中一般精神病理学症状分量表的减分率呈正相关。体重指数的增加与药物治疗前后Clostridium_sensu_stricto_1的变化呈正相关，与基线Bacteroides也呈正相关。进一步的代谢组学分析显示，某些菌属的差异与特定代谢物，如L-甲硫氨酸、L-脯氨酸、高香草酸、N-乙酰血清素和维生素B6的相关性具有统计学意义。结论: 在精神分裂症患者和正常对照之间存在一些微生物群体特征差异且组间差异具有统计学意义，一些菌种与抗精神病药物疗效相关。结合代谢组学结果分析，提示微生物种群可能在抗精神病药物疗效和血清代谢物之间发挥中介作用，为解析抗精神病药物治疗前后肠道微生物对疾病的影响提供了新的证据。

Objective: To explore the role of the microbiota in drug naïve first-onset schizophrenia patients and to seek evidence from multidimensional longitudinal analyses of the intestinal microbiome and clinical phenotype with antipsychotic drugs (APDs) therapy. Methods: In this study, 28 drug naïve first onset schizophrenia patients and age-, gender- and education-matched 29 healthy controls were included, and the patients were treated with APDs. We collected fecal and serum samples at baseline and after 6 weeks of treatment to identify the different microbiota strains and analyse their correlation with clinical symptoms and serum metabolites. The 16S rRNA genes of the gut microbiota were sequenced, and the diversity and relative abundance at the phylum and genus levels were analyzsed in detail. The PANSS score, BMI changed value, and serum metabolome were included in the data analyses. Results: A multiomics study found a potential connection among the clinical phenotype, microbiota and metabolome. The species diversity analyses revealed that the alpha diversity index (chao1, ACE, and goods_coverage) in the schizophrenia APDs group was significantly lower than that in the control group, and the schizophrenia group had clear demarcation from the control group. The microbiota composition analysis results showed that the relative abundance of the genera of Bacteroides, Streptococcus, Romboutsia, and Eubacterium ruminantium group significantly changed after APDs treatment in the schizophrenia patients. These strains could reflect the APDs treatment effect. More genera had differences between the patient and control groups. The LEfSe analysis showed that Prevotella_9 and Bacteroides were enriched in schizophrenia, while Blautia, Dialister, and Roseburia were enriched in the control group. The correlation analysis between microbiota and clinical symptoms showed that Bifidobacterium in schizophrenia was positively correlated with the PANSS reduction rate of the general psychopathology scale. The BMI changed value was positively correlated with the alteration of Clostridium_sensu_stricto_1 during treatment and the baseline abundance of Bacteroides. Moreover, metabolomic data analysis revealed a significant correlation between specific genera and metabolites, such as L-methionine, L-proline, homovanillic acid, N-acetylserotonin, and vitamin B6. Conclusion: Our study found some microbiota features in schizophrenia patients and healthy controls, and several strains were correlated with APDs effects. Furthermore, the multiomics analysis implies the intermediate role of microbiota between antipsychotic effects and serum metabolites and provides new evidence to interpret the difference from multiple levels in the pathogenesis and pharmacological mechanism of schizophrenia.