Multiple Complex Diseases Research Articles

Computational Chemistry in Structure-Based Solute Carrier Transporter Drug Design: Recent Advances and Future Perspectives.

Solute carrier transporters (SLCs) are a class of important transmembrane proteins that are involved in the transportation of diverse solute ions and small molecules into cells. There are approximately 450 SLCs within the human body, and more than a quarter of them are emerging as attractive therapeutic targets for multiple complex diseases, e.g., depression, cancer, and diabetes. However, only 44 unique transporters (∼9.8% of the SLC superfamily) with 3D structures and specific binding sites have been reported. To design innovative and effective drugs targeting diverse SLCs, there are a number of obstacles that need to be overcome. However, computational chemistry, including physics-based molecular modeling and machine learning- and deep learning-based artificial intelligence (AI), provides an alternative and complementary way to the classical drug discovery approach. Here, we present a comprehensive overview on recent advances and existing challenges of the computational techniques in structure-based drug design of SLCs from three main aspects: (i) characterizing multiple conformations of the proteins during the functional process of transportation, (ii) identifying druggability sites especially the cryptic allosteric ones on the transporters for substrates and drugs binding, and (iii) discovering diverse small molecules or synthetic protein binders targeting the binding sites. This work is expected to provide guidelines for a deep understanding of the structure and function of the SLC superfamily to facilitate rational design of novel modulators of the transporters with the aid of state-of-the-art computational chemistry technologies including artificial intelligence.

Ultrahigh Enzyme Loading Metal-Organic Frameworks for Deep Tissue Pancreatic Cancer Photoimmunotherapy.

Protein drugs hold promise in treating multiple complex diseases, including cancer. The priority of protein drug application is precise delivery of substantial bioactive protein into tumor site. Metal-organic-framework (MOF) is widely considered as a promising carrier to encapsulate protein drug owing to the noncovalent interaction between carrier and protein. However, limited loading efficiency and potential toxicity of metal ion in MOF restrict its application in clinical research. Herein, a tumor targeted collagenase-encapsulating MOF via protein-metal ion-organic ligand coordination (PMOCol ) for refining deep tissue pancreatic cancer photoimmunotherapy is developed. By an expedient method in which the ratio of metal ion, histidine residues of protein and ligand is precisely controlled, PMOCol is constructed with ultrahigh encapsulation efficiency (80.3 wt%) and can release collagenase with high enzymatic activity for tumor extracellular matrix (ECM) regulation after reaching tumor microenvironment (TME). Moreover, PMOcol exhibits intensively poorer toxicity than the zeolitic imidazolate framework-8 biomineralized protein. After treatment, the pancreatic tumor with abundant ECM shows enhanced immunocyte infiltration owing to extracellular matrix degradation that improves suppressive TME. By integrating hyperthermia agent with strong near-infrared absorption (1064 nm), PMOCol can induce acute immunogenicity to host immunity activation and systemic immune memory production to prevent tumor development and recurrence.

Hypoxic regulation of hypoxia inducible factor 1 alpha via antisense transcription

Impaired oxygen homeostasis is a frequently encountered pathophysiological factor in multiple complex diseases, including cardiovascular disease and cancer. While the canonical hypoxia response pathway is well characterized, less is known about the role of noncoding RNAs in this process. Here, we investigated the nascent and steady-state noncoding transcriptional responses in endothelial cells and their potential roles in regulating the hypoxic response. Notably, we identify a novel antisense long noncoding RNA that convergently overlaps the majority of the hypoxia inducible factor 1 alpha (HIF1A) locus, which is expressed across several cell types and elevated in atherosclerotic lesions. The antisense (HIF1A-AS) is produced as a stable, unspliced, and polyadenylated nuclear retained transcript. HIF1A-AS is highly induced in hypoxia by both HIF1A and HIF2A and exhibits anticorrelation with the coding HIF1A transcript and protein expression. We further characterized this functional relationship by CRISPR-mediated bimodal perturbation of the HIF1A-AS promoter. We provide evidence that HIF1A-AS represses the expression of HIF1a in cis by repressing transcriptional elongation and deposition of H3K4me3, and that this mechanism is dependent on the act of antisense transcription itself. Overall, our results indicate a critical regulatory role of antisense mediated transcription in regulation of HIF1A expression and cellular response to hypoxia.

Methods for estimating insulin resistance from untargeted metabolomics data

ContextInsulin resistance is associated with multiple complex diseases; however, precise measures of insulin resistance are invasive, expensive, and time-consuming.ObjectiveDevelop estimation models for measures of insulin resistance, including insulin sensitivity index (SI) and homeostatic model assessment of insulin resistance (HOMA-IR) from metabolomics data.DesignInsulin Resistance Atherosclerosis Family Study (IRASFS).SettingCommunity based.ParticipantsMexican Americans (MA) and African Americans (AA).Main outcomeEstimation models for measures of insulin resistance, i.e. SI and HOMA-IR.ResultsLeast Absolute Shrinkage and Selection Operator (LASSO) and Elastic Net regression were used to build insulin resistance estimation models from 1274 metabolites combined with clinical data, e.g. age, sex, body mass index (BMI). Metabolite data were transformed using three approaches, i.e. inverse normal transformation, standardization, and Box Cox transformation. The analysis was performed in one MA recruitment site (San Luis Valley, Colorado (SLV); N = 450) and tested in another MA recruitment site (San Antonio, Texas (SA); N = 473). In addition, the two MA recruitment sites were combined and estimation models tested in the AA recruitment sample (Los Angeles, California; N = 495). Estimated and empiric SI were correlated in the SA (r2 = 0.77) and AA (r2 = 0.74) testing datasets. Further, estimated and empiric SI were consistently associated with BMI, low-density lipoprotein cholesterol (LDL), and triglycerides. We applied similar approaches to estimate HOMA-IR with similar results.ConclusionsWe have developed a method for estimating insulin resistance with metabolomics data that has the potential for application to a wide range of biomedical studies and conditions.

Tissue-specific DNA methylation variability and its potential clinical value.

Complex diseases have multifactorial etiologies making actionable diagnostic biomarkers difficult to identify. Diagnostic research must expand beyond single or a handful of genetic or epigenetic targets for complex disease and explore a broader system of biological pathways. With the objective to develop a diagnostic tool designed to analyze a comprehensive network of epigenetic profiles in complex diseases, we used publicly available DNA methylation data from over 2,400 samples representing 20 cell types and various diseases. This tool, rather than detecting differentially methylated regions at specific genes, measures the intra-individual methylation variability within gene promoters to identify global shifts away from healthy regulatory states. To assess this new approach, we explored three distinct questions: 1) Are profiles of epigenetic variability tissue-specific? 2) Do diseased tissues exhibit altered epigenetic variability compared to normal tissue? 3) Can epigenetic variability be detected in complex disease? Unsupervised clustering established that global epigenetic variability in promoter regions is tissue-specific and promoter regions that are the most epigenetically stable in a specific tissue are associated with genes known to be essential for its function. Furthermore, analysis of epigenetic variability in these most stable regions distinguishes between diseased and normal tissue in multiple complex diseases. Finally, we demonstrate the clinical utility of this new tool in the assessment of a multifactorial condition, male infertility. We show that epigenetic variability in purified sperm is correlated with live birth outcomes in couples undergoing intrauterine insemination (IUI), a common fertility procedure. Men with the least epigenetically variable promoters were almost twice as likely to father a child than men with the greatest number of epigenetically variable promoters. Interestingly, no such difference was identified in men undergoing in vitro fertilization (IVF), another common fertility procedure, suggesting this as a treatment to overcome higher levels of epigenetic variability when trying to conceive.

A hybrid deep forest-based method for predicting synergistic drug combinations

Combination therapy is a promising approach in treating multiple complex diseases. However, the large search space of available drug combinations exacerbates challenge for experimental screening. To predict synergistic drug combinations in different cancer cell lines, we propose an improved deep forest-based method, ForSyn, and design two forest types embedded in ForSyn. ForSyn handles imbalanced and high-dimensional data in medium-/small-scale datasets, which are inherent characteristics of drug combination datasets. Compared with 12 state-of-the-art methods, ForSyn ranks first on four metrics for eight datasets with different feature combinations. We conduct a systematic analysis to identify the most appropriate configuration parameters. We validate the predictive value of ForSyn with cell-based experiments on several previously unexplored drug combinations. Finally, a systematic analysis of feature importance is performed on the top contributing features extracted by ForSyn. The resulting key genes may play key roles on corresponding cancers.

Denoising Autoencoder based Long non-coding RNA-Disease Association Prediction

Long non-coding RNAs (lncRNAs) are recent listing in RNA Bioinformatics, which is getting more popular due to their important functional roles. According to the available research, lncRNAs play an essential role in multiple complex diseases. Determining the function of lncRNAs in diseases will help to comprehend many missing links in the disease mechanism. Predicting lncRNAdisease association (LDA) is a crucial stage in this process which is getting at most research interest nowadays. The developments in machine learning and deep learning technologies influenced recent research on LDA models. Most of the methods analyse the interactions of lncRNA with other molecules such as microRNA (miRNA), messenger RNA(mRNA), and proteins. Deep learning models, specifically from autoencoder classes, used extensively in unsupervised learning of features from these associations. This research paper proposes a denoising autoencoder (DAE) based LDA prediction approach. The proposed model uses DAE to learn lncRNA-disease representations from multiple biological networks such as lncRNA-miRNA, miRNA-disease, and disease-lncRNA interactions. The experiments show that the model outperforms other state-of-the-art LDA models concerning the area under the ROC curve (AUC-ROC, 0.94) and the area under precision-recall (AUPR, 0.9592).

A machine learning framework based on multi-source feature fusion for circRNA-disease association prediction.

Circular RNAs (circRNAs) are involved in the regulatory mechanisms of multiple complex diseases, and the identification of their associations is critical to the diagnosis and treatment of diseases. In recent years, many computational methods have been designed to predict circRNA-disease associations. However, most of the existing methods rely on single correlation data. Here, we propose a machine learning framework for circRNA-disease association prediction, called MLCDA, which effectively fuses multiple sources of heterogeneous information including circRNA sequences and disease ontology. Comprehensive evaluation in the gold standard dataset showed that MLCDA can successfully capture the complex relationships between circRNAs and diseases and accurately predict their potential associations. In addition, the results of case studies on real data show that MLCDA significantly outperforms other existing methods. MLCDA can serve as a useful tool for circRNA-disease association prediction, providing mechanistic insights for disease research and thus facilitating the progress of disease treatment.

Integrated co-expression network analysis uncovers novel tissue-specific genes in major depressive disorder and bipolar disorder.

Tissue-specific gene expression has been found to be associated with multiple complex diseases including cancer, metabolic disease, aging, etc. However, few studies of brain-tissue-specific gene expression patterns have been reported, especially in psychiatric disorders. In this study, we performed joint analysis on large-scale transcriptome multi-tissue data to investigate tissue-specific expression patterns in major depressive disorder (MDD) and bipolar disorder (BP). We established the strategies of identifying tissues-specific modules, annotated pathways for elucidating biological functions of tissues, and tissue-specific genes based on weighted gene co-expression network analysis (WGCNA) and robust rank aggregation (RRA) with transcriptional profiling data from different human tissues and genome wide association study (GWAS) data, which have been expanded into overlapping tissue-specific modules and genes sharing with MDD and BP. Nine tissue-specific modules were identified and distributed across the four tissues in the MDD and six modules in the BP. In general, the annotated biological functions of differentially expressed genes (DEGs) in blood were mainly involved in MDD and BP progression through immune response, while those in the brain were in neuron and neuroendocrine response. Tissue-specific genes of the prefrontal cortex (PFC) in MDD-, such as IGFBP2 and HTR1A, were involved in disease-related functions, such as response to glucocorticoid, taste transduction, and tissue-specific genes of PFC in BP-, such as CHRM5 and LTB4R2, were involved in neuroactive ligand-receptor interaction. We also found PFC tissue-specific genes including SST and CRHBP were shared in MDD-BP, SST was enriched in neuroactive ligand-receptor interaction, and CRHBP shown was related to the regulation of hormone secretion and hormone transport.

Finding Lung-Cancer-Related lncRNAs Based on Laplacian Regularized Least Squares With Unbalanced Bi-Random Walk.

Lung cancer is one of the leading causes of cancer-related deaths. Thus, it is important to find its biomarkers. Furthermore, there is an increasing number of studies reporting that long noncoding RNAs (lncRNAs) demonstrate dense linkages with multiple human complex diseases. Inferring new lncRNA-disease associations help to identify potential biomarkers for lung cancer and further understand its pathogenesis, design new drugs, and formulate individualized therapeutic options for lung cancer patients. This study developed a computational method (LDA-RLSURW) by integrating Laplacian regularized least squares and unbalanced bi-random walk to discover possible lncRNA biomarkers for lung cancer. First, the lncRNA and disease similarities were computed. Second, unbalanced bi-random walk was, respectively, applied to the lncRNA and disease networks to score associations between diseases and lncRNAs. Third, Laplacian regularized least squares were further used to compute the association probability between each lncRNA-disease pair based on the computed random walk scores. LDA-RLSURW was compared using 10 classical LDA prediction methods, and the best AUC value of 0.9027 on the lncRNADisease database was obtained. We found the top 30 lncRNAs associated with lung cancers and inferred that lncRNAs TUG1, PTENP1, and UCA1 may be biomarkers of lung neoplasms, non-small–cell lung cancer, and LUAD, respectively.

LRP6 Receptor Plays Essential Functions in Development and Human Diseases.

LRP6 is a member of the low-density lipoprotein receptor superfamily of cell-surface receptors. It is required for the activation of the Wnt/β-catenin signalling pathway. LRP6 is detected in different tissue types and is involved in numerous biological activities such as cell proliferation, specification, metastatic cancer, and embryonic development. LRP6 is essential for the proper development of different organs in vertebrates, such as Xenopus laevis, chickens, and mice. In human, LRP6 overexpression and mutations have been reported in multiple complex diseases including hypertension, atherosclerosis, and cancers. Clinical studies have shown that LRP6 is involved in various kinds of cancer, such as bladder and breast cancer. Therefore, in this review, we focus on the structure of LRP6 and its interactions with Wnt inhibitors (DKK1, SOST). We also discuss the expression of LRP6 in different model systems, with emphasis on its function in development and human diseases.

Leveraging pleiotropic association using sparse group variable selection in genomics data

BackgroundGenome-wide association studies (GWAS) have identified genetic variants associated with multiple complex diseases. We can leverage this phenomenon, known as pleiotropy, to integrate multiple data sources in a joint analysis. Often integrating additional information such as gene pathway knowledge can improve statistical efficiency and biological interpretation. In this article, we propose statistical methods which incorporate both gene pathway and pleiotropy knowledge to increase statistical power and identify important risk variants affecting multiple traits.MethodsWe propose novel feature selection methods for the group variable selection in multi-task regression problem. We develop penalised likelihood methods exploiting different penalties to induce structured sparsity at a gene (or pathway) and SNP level across all studies. We implement an alternating direction method of multipliers (ADMM) algorithm for our penalised regression methods. The performance of our approaches are compared to a subset based meta analysis approach on simulated data sets. A bootstrap sampling strategy is provided to explore the stability of the penalised methods.ResultsOur methods are applied to identify potential pleiotropy in an application considering the joint analysis of thyroid and breast cancers. The methods were able to detect eleven potential pleiotropic SNPs and six pathways. A simulation study found that our method was able to detect more true signals than a popular competing method while retaining a similar false discovery rate.ConclusionWe developed feature selection methods for jointly analysing multiple logistic regression tasks where prior grouping knowledge is available. Our method performed well on both simulation studies and when applied to a real data analysis of multiple cancers.

Network Integration Analysis and Immune Infiltration Analysis Reveal Potential Biomarkers for Primary Open-Angle Glaucoma.

Primary open-angle glaucoma (POAG) is a progressive optic neuropathy and its damage to vision is irreversible. Therefore, early diagnosis assisted by biomarkers is essential. Although there were multiple researches on the identification of POAG biomarkers, few studies systematically revealed the transcriptome dysregulation mechanism of POAG from the perspective of pre- and post-transcription of genes. Here, we have collected multiple sets of POAG’s aqueous humor (AH) tissue transcription profiles covering long non-coding RNA (lncRNA), mRNA and mircoRNA (miRNA). Through differential expression analysis, we identified thousands of significant differentially expressed genes (DEGs) between the AH tissue of POAG and non-glaucoma. Further, the DEGs were used to construct a competing endogenous RNA (ceRNA) regulatory network and 1,653 qualified lncRNA-miRNA-mRNA regulatory units were identified. Two ceRNA regulatory subnets were identified based on the random walk algorithm and revealed to be involved in the regulation of multiple complex diseases. At the pre-transcriptional regulation level, a transcriptional regulatory network was constructed and three transcription factors (FOS, ATF4, and RELB) were identified to regulate the expression of multiple genes and participate in the regulation of T cells. Moreover, we revealed the immune desert status of AH tissue for POAG patients based on immune infiltration analysis and identified a specific AL590666.2-hsa−miR−339−5p-UROD axis can be used as a biomarker of POAG. Taken together, the identification of regulatory mechanisms and biomarkers will contribute to the individualized diagnosis and treatment for POAG.

In Silico Prediction of Potential Drug Combinations for Type 2 Diabetes Mellitus by an Integrated Network and Transcriptome Analysis.

Type 2 diabetes mellitus (T2DM) is a heterogeneous disorder, so achieving the desired therapeutic efficacy through monotherapy is tricky. Drug combinations play a vital role in treating multiple complex diseases by providing increased efficacy and reduced toxicity. Here, we adopted a computational framework to discover potential drugs and drug pairs for T2DM. Firstly, we collected T2DM-associated genes and constructed the disease module for T2DM. Then, by quantifying the proximity between drugs and the disease module, we found out potential drugs. Based on the drug-induced gene expression profiles, we further performed Gene Set Enrichment Analysis (GSEA) on these drugs and identified several potential candidates. In addition, through network-based separation, potential drug combinations for T2DM were predicted. Results from this study could provide insights for anti-T2DM drug discovery and rational drug use of existing agents. As a useful computational framework, our approach could also be applied in drug research for other complex diseases.

METTL3-m6A-Rubicon axis inhibits autophagy in nonalcoholic fatty liver disease

N6-methyladenosine (m6A) mRNA modification plays critical roles in various biological events and is involved in multiple complex diseases. However, the role of m6A modification in autophagy in nonalcoholic fatty liver disease (NAFLD) remains largely unknown. Here, we report that m6A modification was increased in livers of NAFLD mouse models and in free fatty acid (FFA)-treated hepatocytes, and the abnormal m6A modification was attributed to the upregulation of methyltransferase like 3 (METTL3) induced by lipotoxicity. Knockdown of METTL3 promoted hepatic autophagic flux and clearance of lipid droplets (LDs), while overexpression of METTL3 inhibited these processes. Mechanistically, METTL3 directly bound to Rubicon mRNA and mediated the m6A modification, while YTH N6-methyladenosine RNA binding protein 1 (YTHDF1), as a partner of METTL3, interacted with the m6A-marked Rubicon mRNA and promoted its stability. Subsequently, RUBICON inhibited autophagosome-lysosome fusion and further blocked clearance of LDs. Taken together, our results showed a critical role of METTL3 and YTHDF1 in regulating lipid metabolism via the autophagy pathway and provided a novel insight into m6A mRNA methylation in NAFLD.

Fine-Scale Genetic Structure and Natural Selection Signatures of Southwestern Hans Inferred From Patterns of Genome-Wide Allele, Haplotype, and Haplogroup Lineages.

The evolutionary and admixture history of Han Chinese have been widely discussed via traditional autosomal and uniparental genetic markers [e.g., short tandem repeats, low-density single nucleotide polymorphisms). However, their fine-scale genetic landscapes (admixture scenarios and natural selection signatures) based on the high-density allele/haplotype sharing patterns have not been deeply characterized. Here, we collected and generated genome-wide data of 50 Han Chinese individuals from four populations in Guizhou Province, one of the most ethnolinguistically diverse regions, and merged it with over 3,000 publicly available modern and ancient Eurasians to describe the genetic origin and population admixture history of Guizhou Hans and their neighbors. PCA and ADMIXTURE results showed that the studied four populations were homogeneous and grouped closely to central East Asians. Genetic homogeneity within Guizhou populations was further confirmed via the observed strong genetic affinity with inland Hmong-Mien people through the observed genetic clade in Fst and outgroup f3/f4-statistics. qpGraph-based phylogenies and f4-based demographic models illuminated that Guizhou Hans were well fitted via the admixture of ancient Yellow River Millet farmers related to Lajia people and southern Yangtze River farmers related to Hanben people. Further ChromoPainter-based chromosome painting profiles and GLOBETROTTER-based admixture signatures confirmed the two best source matches for southwestern Hans, respectively, from northern Shaanxi Hans and southern indigenes with variable mixture proportions in the historical period. Further three-way admixture models revealed larger genetic contributions from coastal southern East Asians into Guizhou Hans compared with the proposed inland ancient source from mainland Southeast Asia. We also identified candidate loci (e.g., MTUS2, NOTCH4, EDAR, ADH1B, and ABCG2) with strong natural selection signatures in Guizhou Hans via iHS, nSL, and ihh, which were associated with the susceptibility of the multiple complex diseases, morphology formation, alcohol and lipid metabolism. Generally, we provided a case and ideal strategy to reconstruct the detailed demographic evolutionary history of Guizhou Hans, which provided new insights into the fine-scale genomic formation of one ethnolinguistically specific targeted population from the comprehensive perspectives of the shared unlinked alleles, linked haplotypes, and paternal and maternal lineages.

Efficient detection and classification of epigenomic changes under multiple conditions.

Epigenomics, the study of the human genome and its interactions with proteins and other cellular elements, has become of significant interest in recent years. Such interactions have been shown to regulate essential cellular functions and are associated with multiple complex diseases. Therefore, understanding how these interactions may change across conditions is central in biomedical research. Chromatin immunoprecipitation followed by massively parallel sequencing (ChIP-seq) is one of several techniques to detect local changes in epigenomic activity (peaks). However, existing methods for differential peak calling are not optimized for the diversity in ChIP-seq signal profiles, are limited to the analysis of two conditions, or cannot classify specific patterns of differential change when multiple patterns exist. To address these limitations, we present a flexible and efficient method for the detection of differential epigenomic activity across multiple conditions. We utilize data from the ENCODE Consortium and show that the presented method, epigraHMM, exhibits superior performance to current tools and it is among the fastest algorithms available, while allowing the classification of combinatorial patterns of differential epigenomic activity and the characterization of chromatin regulatorystates.

The environmental toxicant ziram enhances neurotransmitter release and increases neuronal excitability via the EAG family of potassium channels

Environmental toxicants have the potential to contribute to the pathophysiology of multiple complex diseases, but the underlying mechanisms remain obscure. One such toxicant is the widely used fungicide ziram, a dithiocarbamate known to have neurotoxic effects and to increase the risk of Parkinson's disease. We have used Drosophila melanogaster as an unbiased discovery tool to identify novel molecular pathways by which ziram may disrupt neuronal function. Consistent with previous results in mammalian cells, we find that ziram increases the probability of synaptic vesicle release by dysregulation of the ubiquitin signaling system. In addition, we find that ziram increases neuronal excitability. Using a combination of live imaging and electrophysiology, we find that ziram increases excitability in both aminergic and glutamatergic neurons. This increased excitability is phenocopied and occluded by null mutant animals of the ether a-go-go (eag) potassium channel. A pharmacological inhibitor of the temperature sensitive hERG (human ether-a-go-go related gene) phenocopies the excitability effects of ziram but only at elevated temperatures. seizure (sei), a fly ortholog of hERG, is thus another candidate target of ziram. Taken together, the eag family of potassium channels emerges as a candidate for mediating some of the toxic effects of ziram. We propose that ziram may contribute to the risk of complex human diseases by blockade of human eag and sei orthologs, such as hERG.

Vitamin D deficiency in Mexicans have a high prevalence: a cross-sectional analysis of the patients from the Centro Médico Nacional 20 de Noviembre.

To know the prevalence of vitamin D insufficiency (21-29ng/mL) and deficiency (< 20ng/mL) in Mexican patients; although it is a risk factor for developing multiple complex diseases, its prevalence in the population is still unknown. Cross-sectional study carried out at the endocrinology service of the highly specialized national center November 20. Data on cardiovascular risk factors were obtained and 25-hydroxy vitamin D was measured by chemiluminescence. Prevalence was calculated, and the results were analyzed to categorize the patients according to 25-hydroxy vitamin D deficient or insufficient levels. The mean value of the serum vitamin D concentration was 18.37ng/mL. Of the 117 patients, 93.2% (n = 109) have decreased vitamin D values; 62.4% (n = 73) of the patients had vitamin D deficiency and 30.8% (n = 36) vitamin D insufficiency. The prevalence of vitamin D deficiency was 62.4% and 30.8% for vitamin D insufficiency. The total prevalence of alterations in vitamin D levels in this population was 93.2%. This study reports a prevalence of vitamin D deficiency and insufficiency much higher than those described by previous studies, which is of utmost importance for the population due to the morbidities associated with these alterations.

FCGCNMDA: predicting miRNA-disease associations by applying fully connected graph convolutional networks.

Growing evidence indicates that the development and progression of multiple complex diseases are influenced by microRNA (miRNA). Identifying more miRNAs as biomarkers for clinical diagnosis, treatment and prognosis is vital to promote the development of bioinformatics and medicine. Considering that the traditional biological experimental methods are generally time-consuming and expensive, high-efficient computational methods are encouraged to uncover potential disease-related miRNAs. In this paper, FCGCNMDA is presented to predict latent miRNA-disease associations by utilizing fully connected graph convolutional networks. Specially, our method first constructs a fully connected graph in which edge weights represent correlation coefficient between any two pairs of miRNA-disease pair, and then feeds this fully connected graph along with miRNA-disease pairs feature matrix into a two-layer graph convolutional networks (GCN) for training. At last, we utilize the trained network to predict the scores for unknown miRNA-disease pairs. As a result, FCGCNMDA achieves AUC value of [Formula: see text] and AUPRC value of [Formula: see text] in HMDD v2.0 based on five-fold cross validation. Moreover, case studies on Lymphoma, Breast Neoplasms and Prostate Neoplasms shown that 98%, 98%, 98% of the top 50 selected miRNAs were validated by recent experimental evidence. From above results, we can deduce that FCGCNMDA can be regarded as reliable method for potential miRNA-disease associations prediction.