Development of a multiple convolutional neural network-facilitated diagnostic screening program for immunofluorescence images of IgA nephropathy and idiopathic membranous nephropathy.

Background: Immunoglobulin A nephropathy (IgAN) and idiopathic membranous nephropathy (IMN) are the most common glomerular diseases. Immunofluorescence (IF) tests of renal tissues are crucial for the diagnosis. We developed a multiple convolutional neural network (CNN)-facilitated diagnostic program to assist the IF diagnosis of IgAN and IMN. Methods: The diagnostic program consisted of four parts: a CNN trained as a glomeruli detection module, an IF intensity comparator, dual-CNN (D-CNN) trained as a deposition appearance and location classifier and a post-processing module. A total of 1573 glomerular IF images from 1009 patients with glomerular diseases were used for the training and validation of the diagnostic program. A total of 1610 images of 426 patients from different hospitals were used as test datasets. The performance of the diagnostic program was compared with nephropathologists. Results: In >90% of the tested images, the glomerulus location module achieved an intersection over union >0.8. The accuracy of the D-CNN in recognizing irregular granular mesangial deposition and fine granular deposition along the glomerular basement membrane was 96.1% and 93.3%, respectively. As for the diagnostic program, the accuracy, sensitivity and specificity of diagnosing suspected IgAN were 97.6%, 94.4% and 96.0%, respectively. The accuracy, sensitivity and specificity of diagnosing suspected IMN were 91.7%, 88.9% and 95.8%, respectively. The corresponding areas under the curve (AUCs) were 0.983 and 0.935. When tested with images from the outside hospital, the diagnostic program showed stable performance. The AUCs for diagnosing suspected IgAN and IMN were 0.972 and 0.948, respectively. Compared with inexperienced nephropathologists, the program showed better performance. Conclusion: The proposed diagnostic program could assist the IF diagnosis of IgAN and IMN.

Analysis of Glomerular IgG Subclasses Switch in Idiopathic Membranous Nephropathy Classified by Glomerular Phospholipase A2 Receptor Antigen and Serum Antibody.

BACKGROUND: The role of IgG subclass in idiopathic membranous nephropathy (IMN) was unclarified. Recent study found IgG subtype switches from IgG1 to IgG4 in the early pathological stage in IMN. The profile of IgG subclass in phospholipase A2 receptor- (PLA2R-) related and PLA2R-unrelated IMN was unrevealed. Our study is aimed at testifying whether IgG subclass switch existed in PLA2R-related and PLA2R-unrelated IMN, respectively. METHODS: Our study retrospectively enrolled 157 Chinese patients with biopsy-confirmed IMN between September 2017 and November 2019. We measured glomerular PLA2R antigen and serum anti-PLA2R antibody to classify the patients into PLA2R-related (n = 132) and PLA2R-unrelated (n = 25) subgroup. We evaluated glomerular IgG subclass by immunofluorescence (IF) predominance. Our study defined IgG subclass deposition as predominant if the IF score was higher than the other three and ≥1 +, or as codominant if the IF intensity was equal to any other and ≥1 +. We explored the relationship between IF predominance of glomerular IgG subtype and electron microscopic (EM) stages of IMN. RESULTS: We did not find statistical difference of predominant or codominant rate (pre/co-rate) among EM stages in any subclass (P > 0.05). Pre/co-rate of IgG3 linearly associated with EM stage in total and PLA2R-related subgroup (P = 0.044, P = 0.013). PLA2R-related subgroup showed higher IgG4 intensity (2.1 ± 0.6 vs. 1.6 ± 0.7, P = 0.001) and pre/co-rate of IgG4 in stage 1 (97% vs. 57%, P = 0.015) than PLA2R-unrelated group. We found no difference of IgG subclass pre/co-rate in different EM stages or linear association between pre/co-rate of IgG1, IgG2, IgG4, and EM stages (P > 0.05). CONCLUSIONS: Pre/co-rate of IgG3 declined with EM stage in total and PLA2R-related subgroup. We did not find IgG subclass switches from IgG1 to IgG4 in either IMN patients or subgroups.

Genetic variations in microRNA processing genes are associated with susceptibility in depression.

MicroRNAs (miRNAs) are new prominent gene expression regulators that have critical roles in neural development by regulating synaptic functions, and miRNA biogenesis may play an important role in psychiatric disorders. Despite emerging evidences demonstrating that single-nucleotide polymorphisms in the miRNA processing genes were associated with cancer and cardiovascular disorders, evidences about association between variants of the genes and depression are lacking. This study aims to find the association between miRNA processing gene variants and depression. We genotyped three polymorphisms from three miRNA processing genes in a case-control study including 314 patients and 252 matched healthy controls. The high-resolution melting method was used to genotype the three loci. Frequencies of genotypes and alleles showed significant difference between patients with depression and healthy controls in DGCR8 rs3757 and AGO1 rs636832. An allele frequency was significantly higher in rs3757 and lower in rs636832, respectively. Variant allele of DGCR8 rs3757 was associated with increased risk of suicidal tendency and improvement response to antidepressant treatment, whereas the variant of AGO1 rs636832 showed decreased risk of suicidal tendency, suicidal behavior, and recurrence. Besides allele frequency showed significant difference when compared patients with remission to controls, no significant differences were found in GEMIN4 rs7813 between patients and healthy controls. DGCR8 rs3757 and AGO1 rs636832 were found to have significant association with depression, and GEMIN4 rs7813 did not affect susceptibility to depression. These observations suggested that miRNA processing polymorphisms may affect depression risk and treatment.