Background:Sjogren’s syndrome(pSS) is a chronic, progressive, and systematic autoimmune disease characterized by lymphocytic infiltration of exocrine glands 1 2. Sicca symptoms and abnormal fatigue are the main clinical presentation, but those symptoms are non-specific to patients, which lead to delayed diagnosis 1 3. The heterogeneous of clinical manifestation raise challenges regarding diagnosis and therapy in pSS, thus it’s necessary for us to sub-classify pSS.Objectives:To explore new biomarkers for diagnosis and subtypes of pSS based on Machine Learning Primary.Methods:All microarray raw datas (CEL files) were screened and downloaded from Gene Expression Omnibus (GEO). Meta-analysis to identify the consistent DEGs by MetaOmics. Weighted gene co-expression network analysis (WGCNA) was used to the modules related to SS for further analysis. Subclasses were computed using a consensus Non-negative Matrix Factorization (NMF) clustering method. Immune cell infiltration was used to evaluate the expression of immune cells and obtain various immune cell proportions from samples. P value < 0.05 were considered statistically significant. All the analyses were conducted under R environment (version 4.03).Results:A total of 3715 consistent DEGs were identified from the four datasets, including 1748 up-regulated and 1967 down-regulated genes. Tour meaningful modules, including yellow, turquoise, grey60 and bule, were identified (Figure 1A,1B). And 183 overlapping gene were screened from the DEGs and the Hub genes in the four modles for further analysis. We final divided pSS patients into three subtypes, of which yellow and turquoise in Sub1, grey60 in Sub2 and blue in Sub3. Sub1 and Sub3 were related to cell metabolism, while Sub2 had connection with virus infection (Figure 1C,1D). Infiltrated immune cells were also different among these three types (Figure 1E,1F).Conclusion:Patients with pSS could be classified into 3 subtypes, this classification might help for assessing prognosis and guiding precise treatment.References:[1]Ramos-Casals M, Brito-Zerón P, Sisó-Almirall A, et al. Primary Sjogren syndrome. BMJ (Clinical research ed) 2012;344:e3821. doi: 10.1136/bmj.e3821 [published Online First: 2012/06/16].[2]Brito-Zeron P, Baldini C, Bootsma H, et al. Sjogren syndrome. Nat Rev Dis Primers 2016;2:16047. doi: 10.1038/nrdp.2016.47 [published Online First: 2016/07/08].[3]Segal B, Bowman SJ, Fox PC, et al. Primary Sjogren’s Syndrome: health experiences and predictors of health quality among patients in the United States. Health Qual Life Outcomes 2009;7:46. doi: 10.1186/1477-7525-7-46 [published Online First: 2009/05/29].Acknowledgements:This project was supported by National Science Foundation of China (82001740), Open Fund from the Key Laboratory of Cellular Physiology (Shanxi Medical University) (KLCP2019) and Innovation Plan for Postgraduate Education in Shanxi Province (2020BY078).Disclosure of Interests:None declared
Predicting Lymphoma Development by Exploiting Genetic Variants and Clinical Findings in a Machine Learning-Based Methodology With Ensemble Classifiers in a Cohort of Sjögren's Syndrome Patients
