Validation of a microRNA target site polymorphism in H3F3B that is potentially associated with a broad schizophrenia phenotype

MicroRNAs in Genetic Etiology of Human Diseases

Since their first discovery more than 20 years ago, miRNAs have been subject to deliberate research and analysis for revealing their physiological or pathological involvement. Regulatory roles of miRNAs in signal transduction, gene expression, and cellular processes in development, differentiation, proliferation, apoptosis, and homeostasis also imply their critical role in disease pathogenesis. Their roles in cancer, neurodegenerative diseases, and other systemic diseases have been studied broadly. In these regulatory pathways, their mutations and target sequence variations play critical roles to determine their functional repertoire. In this chapter, we summarize studies that investigated the role of mutations, polymorphisms, and other variations of miRNAs in respect to pathological processes.

Effects of HTR1B 3' region polymorphisms and functional regions on gene expression regulation

Background

The HTR1B gene encodes the 5-hydroxytryptamine (5-HT1B) receptor, which is involved in a variety of brain activities and mental disorders. The regulatory effects of non-coding regions on genomic DNA are one of many reasons for the cause of genetic-related diseases. Post-transcriptional regulation that depends on the function of 3′ regulatory regions plays a particularly important role. This study investigated the effects, on reporter gene expression, of several haplotypes of the HTR1B gene (rs6297, rs3827804, rs140792648, rs9361234, rs76194807, rs58138557, and rs13212041) and truncated fragments in order to analyze the function of the 3′ region of HTR1B.

Results

We found that the haplotype, A-G-Del-C-T-Ins-A, enhanced the expression level compared to the main haplotype; A-G-Del-C-G-Ins-A; G-G-Del-C-G-Ins-G decreased the expression level. Two alleles, rs76194807T and rs6297G, exhibited different relative luciferase intensities compared to their counterparts at each locus. We also found that + 2440 ~ + 2769 bp and + 1953 ~ + 2311 bp regions both had negative effects on gene expression.

Conclusions

The 3′ region of HTR1B has a regulatory effect on gene expression, which is likely closely associated with the interpretation of HTR1B-related disorders. In addition, the HTR1B 3′ region includes several effector binding sites that induce an inhibitory effect on gene expression.

The interaction of histone modification related H3F3B and NSD2 genes increases the susceptibility to schizophrenia in a Chinese population

The role of histone modifications in the pathogenesis of schizophrenia has been proposed previously. H3F3B is a member of the histone 3. NSD2 is a histone methyltransferase that mediates dimethylation of Histone 3 lysine 36 (H3K36me2). The aim of the current study was to explore the associations between SNPs within H3F3B gene (rs60700976, rs3214028) and NSD2 gene (rs13148597, rs75820801) and the susceptibility to schizophrenia in a Chinese population. A total of 810 patients and 490 healthy controls were recruited and genetic association analyses were performed. The H3F3B gene polymorphisms rs3214028 and rs60700976 were significantly associated with schizophrenia. Rs60700976 was also associated with psychotic symptoms in schizophrenia patients. Furthermore, we found the interaction between NSD2 gene and H3F3B gene was related to the susceptibility to schizophrenia. The corresponding best three-locus model was H3F3B (rs60700976) - NSD2 (rs75820801, rs13148597), and the high-risk genotype combination was rs13148597(CC)- rs60700976(GG)-rs75820801(TT) (OR = 1.388[1.091-1.766], P = .007). The low-risk genotype combination was rs13148597(CC)-rs60700976(GG)-rs75820801(CT) (OR = 0.57 [0.330-0.985], P = .042). Our findings provided the preliminary evidence that the histone modification related H3F3B and NSD2 genes may confer the susceptibility to schizophrenia in a Chinese population.

Deep Feature Selection and Causal Analysis of Alzheimer's Disease

Deep convolutional neural networks (DCNNs) have achieved great success for image classification in medical research. Deep learning with brain imaging is the imaging method of choice for the diagnosis and prediction of Alzheimer’s disease (AD). However, it is also well known that DCNNs are “black boxes” owing to their low interpretability to humans. The lack of transparency of deep learning compromises its application to the prediction and mechanism investigation in AD. To overcome this limitation, we develop a novel general framework that integrates deep leaning, feature selection, causal inference, and genetic-imaging data analysis for predicting and understanding AD. The proposed algorithm not only improves the prediction accuracy but also identifies the brain regions underlying the development of AD and causal paths from genetic variants to AD via image mediation. The proposed algorithm is applied to the Alzheimer’s Disease Neuroimaging Initiative (ADNI) dataset with diffusion tensor imaging (DTI) in 151 subjects (51 AD and 100 non-AD) who were measured at four time points of baseline, 6 months, 12 months, and 24 months. The algorithm identified brain regions underlying AD consisting of the temporal lobes (including the hippocampus) and the ventricular system.