Decoding the genetic landscape of juvenile dermatomyositis: insights from phosphorylation-associated single nucleotide polymorphisms.

Genome-wide association studies (GWASs) have identified genetic susceptibility loci associated with juvenile dermatomyositis (JDM). Single nucleotide polymorphisms related to phosphorylation (phosSNPs) are critical nonsynonymous mutations exerting substantial influence on gene expression regulation. The aim of this study was to identify JDM susceptibility genes in the GWAS loci by the use of phosSNPs. We explored quantitative trait loci (QTLs) among the phosSNPs associated with JDM using data from eQTL (bulk tissues and single-cell) and pQTL studies. For gene expression and protein levels significantly influenced by JDM-associated phosSNPs, we assessed their associations with JDM through MR analyses. Additionally, we conducted differential expression gene analyses, incorporating single-cell transcriptomic profiling of 6 JDM cases and 11 juvenile controls (99,396 cells). We identified 31 phosSNPs situated in the 6p21 locus that were associated with JDM. Half of these phosSNPs showed effects on gene expression in various cells and circulating protein levels. In MR analyses, we established associations between the expression levels of pivotal JDM-associated genes, including MICB, C4A, HLA-DRB1, HLA-DRB5, and PSMB9, in skin, muscle, or blood cells and circulating levels of C4A, with JDM. Utilizing single-cell eQTL data, we identified a total of 276 association signals across 14 distinct immune cell types for 28 phosSNPs. Further insights were gained through single-cell differential expression analysis, revealing differential expression of PSMB9, HLA-A, HLA-B, HLA-C, HLA-DPB1, HLA-DQA1, HLA-DQB1, and HLA-DRB1 in immune cells. The present study pinpointed phosSNPs within susceptibility genes for JDM and unraveled the intricate relationships among these SNPs, gene expression levels, and JDM.

Sex-specific association between maternal mild anemia and children's behavioral development: a birth cohort study.

There has been limited research on maternal anemia affecting children's behavioral development, with a lack of studies focusing on sex differences in this association. Based on the Ma'anshan Birth Cohort, 2132 mother-child pairs were included. Maternal anemia was evaluated based on the hemoglobin concentration and children's behavioral development was assessed by Achenbach Child Behavior Checklist 1.5-5. Binary logistic regression models indicated that compared with children born of mothers without anemia throughout pregnancy, maternal mild anemia during pregnancy or only anemia in the 3rd trimester was associated with increased risks of aggressive behaviors in boys. Maternal mild anemia only in the 2nd trimester was associated with increased risks of attention problems in boys. In girls, maternal mild anemia during pregnancy was associated with increased risks of withdrawn, internalizing problems and total problems. Girls born of mothers with mild anemia only in the 2nd trimester had higher risks of total problems. Maternal mild anemia in both 2nd and 3rd trimesters was associated with increased risks of internalizing problems in girls. Our study identified sex-specific effects of maternal mild anemia during pregnancy on children's behavioral development problems.

A Heterozygous Variant of FGF13 Caused X-Linked Developmental and Epileptic Encephalopathy 90 in a Chinese Family.

Developmental and epileptic encephalopathy (DEE) refers to a group of severe epilepsy encephalopathy and development disorders, and its typical clinical features include seizures, drug resistance, and developmental delay or regression. To date, limited studies have reported DEEs driven by FGF13. Here, we reported a girl with developmental and epileptic encephalopathy 90 caused by variant of FGF13. Her electroencephalogram (EEG) showed discontinuous hypsarrhythmia, and a heterozygous nonsynonymous variant in FGF13 [NM_004114.4: c.5C>G, p.(Ala2Gly)] was identified from the proband. The variant was not reported in public databases such as gnomAD and Exome Aggregation Consortium (ExAC), and was predicted to be damaging to proteins and classified as likely pathogenic according to the ACMG guidelines. The seizure was finally controlled by a combination of ACTH + zonisamide (10 mg/kg.d) + levetiracetam (52 mg/kg.d) + clonazepam (0.7 mg/kg.d).

Phosphorylation of transcription factor bZIP21 by MAP kinase MPK6-3 enhances banana fruit ripening.

Ripening of fleshy fruits involves both diverse post-translational modifications (PTMs) and dynamic transcriptional reprogramming, but the interconnection between PTMs, such as protein phosphorylation and transcriptional regulation, in fruit ripening remains to be deciphered. Here, we conducted a phosphoproteomic analysis during banana (Musa acuminata) ripening and identified 63 unique phosphopeptides corresponding to 49 proteins. Among them, a Musa acuminata basic leucine zipper transcription factor21 (MabZIP21) displayed elevated phosphorylation level in the ripening stage. MabZIP21 transcript and phosphorylation abundance increased during banana ripening. Genome-wide MabZIP21 DNA binding assays revealed MabZIP21-regulated functional genes contributing to banana ripening, and electrophoretic mobility shift assay, chromatin immunoprecipitation coupled with quantitative polymerase chain reaction, and dual-luciferase reporter analyses demonstrated that MabZIP21 stimulates the transcription of a subset of ripening-related genes via directly binding to their promoters. Moreover, MabZIP21 can be phosphorylated by MaMPK6-3, which plays a role in banana ripening, and T318 and S436 are important phosphorylation sites. Protein phosphorylation enhanced MabZIP21-mediated transcriptional activation ability, and transient overexpression of the phosphomimetic form of MabZIP21 accelerated banana fruit ripening. Additionally, MabZIP21 enlarges its role in transcriptional regulation by activating the transcription of both MaMPK6-3 and itself. Taken together, this study reveals an important machinery of protein phosphorylation in banana fruit ripening in which MabZIP21 is a component of the complex phosphorylation pathway linking the upstream signal mediated by MaMPK6-3 with transcriptional controlling of a subset of ripening-associated genes.

Thyroid function test abnormalities-isolated TPOAb+, SCH and hypothyroxinemia and preschool children's neurodevelopment.

OBJECTIVE: Thyroid function test abnormalities are frequent and associated with the offspring's adverse neurodevelopment. This study aimed to examine the relationship between maternal thyroid function test abnormalities before 20 gestational weeks and children's cognitive, emotional and behavioural development at 3-6 years of age. PATIENTS AND MEASUREMENTS: A total of 2243 mother-child pairs were included in the final analysis. Maternal thyroid function was evaluated retrospectively during the children's preschool period. The serum thyrotrophin, free thyroxine and thyroid peroxidase antibodies (TPOAb) were detected by chemiluminescence immunoassay during the follow-up period. The neurodevelopmental status of preschoolers aged 3-6 years was evaluated by parental versions of The Behavior Rating Inventory of Executive Function-Preschool and The Strengths and Difficulties Questionnaires. The associations between maternal thyroid function test abnormalities and preschoolers' neurodevelopment were examined using Poisson regression models. RESULTS: After adjusting for potential confounders in Poisson regression analyses, it showed that maternal isolated TPOAb positivity before 20 gestational weeks may be associated with the increased risk of abnormalities in peer problems (odds ratio [OR] = 1.90, 95% confidence interval [CI]: 1.26, 287). Maternal isolated SCH before 20 gestational weeks was observed to be related with increased risk of abnormalities in inhibition (OR = 2.73, 95% CI: 1.37, 5.41), working memory (OR = 1.67, 95% CI: 1.04, 2.70), conduct problems (OR = 1.80, 95% CI: 1.05, 3.09), hyperactivity (OR = 1.94, 95% CI:1.08, 3.49) and total difficulties (OR = 1.94, 95% CI: 1.13, 3.34). Maternal isolated hypothyroxinemia before 20 gestational was observed to be related with increased risk of abnormalities in peer problems (OR = 2.71, 95% CI: 1.17, 6.27). CONCLUSIONS: Thyroid function test abnormalities before 20 gestational weeks may be associated with children's neurodevelopment at 3-6 years of age.

Maternal pregnancy-related anxiety and children's physical growth: the Ma'anshan birth cohort study.

BACKGROUND: Epidemiological studies have identified maternal antenatal anxiety and several adverse birth outcomes, but limited studies have focused on the relationship with the long-term physical growth of children. The study aimed to assess the influence of maternal pregnancy-related anxiety on physical growth in children at different exposure periods during pregnancy. METHODS: 3,154 mother-child pairs were included based on the Ma'anshan birth cohort study. Maternal prenatal anxiety was obtained by administering a questionnaire using the pregnancy-related anxiety questionnaire (PRAQ) scale during the 1st, 2nd and 3rd trimesters of pregnancy. Body fat (BF) (48 to 72 months) and Body Mass Index (BMI) (birth to 72 months) were collected repeatedly for children. Group-based trajectory models were applied to fit the different trajectories of BMI and BF. RESULTS: Maternal anxiety in the 2nd (OR = 0.81; 95% CI: 0.68 to 0.98; P < 0.025) and 3rd (OR = 0.80; 95% CI: 0.67 to 0.97; P = 0.020) trimesters was associated with a decreased risk of rapid weight gain (RWG) in the first year of life. Children aged 48 to 72 months of mothers with anxiety in the 3rd trimester had lower BMI (ß = -0.161; 95% CI, -0.293 to -0.029; P = 0.017) and BF (ß = -0.190; 95% CI, -0.334 to -0.046; P = 0.010), and these children were less likely to develop a very high BMI trajectory (OR = 0.54; 95% CI: 0.34 to 0.84; P = 0.006), and a high BF trajectory (OR = 0.72; 95% CI: 0.53 to 0.99; P = 0.043). Similar associations were found between maternal anxiety in both 2nd and 3rd trimesters and children's physical growth. CONCLUSIONS: Offspring of mothers with prenatal anxiety in the 2nd and 3rd trimesters predicts poorer growth in infancy and preschool age. Early improvement and treatment of prenatal anxiety could benefit physical health and development in early childhood.

Giant axonal neuropathy (GAN) in an 8-year-old girl caused by a homozygous pathogenic splicing variant in GAN gene.

Giant axonal neuropathy (GAN) is a progressive disease that involves the peripheral and central nervous systems. This neurodegenerative disease is caused by variants in the GAN gene encoding gigaxonin, and is inherited in an autosomal recessive manner. Herein, we performed whole-exome sequencing on a 8-year-old child with dense, curly hair, weakness in both lower limbs, and abnormal MRI. The child was born to consanguineous parents. Our results revealed that the child carried the c.1373+1G>A homozygous pathogenic variant of the GAN gene, while both parents were heterozygous carriers. According to the validation at the cDNA levels, the splicing variant led to the skipping of exon 8 and affected the Kelch domain's formation. Unlike the previously reported cases of GAN, the child's clinical manifestations revealed peripheral nervous system involvement, no vertebral signs, cerebellar signs, and spasticity, but only MRI abnormalities. These results suggested that the patient's central nervous system was mildly involved, which may be related to the genotype. In order to further clarify the correlation between GAN genotype and phenotype, combined with this patient, 54 cases of reported homozygous variants of the GAN gene were merged for the analysis of genotype and phenotype. The results revealed a certain correlation between the GAN gene variant domain and the patient's clinical phenotype, such as central nervous system involvement and age of onset.

PhosSNPs-Regulated Gene Network and Pathway Significant for Rheumatoid Arthritis.

OBJECTIVES: Peripheral blood mononuclear cells (PBMCs) are critical for immunity and participate in multiple human diseases, including rheumatoid arthritis (RA). PhosSNPs are nonsynonymous SNPs influencing protein phosphorylation, thus probably modulate cell signaling and gene expression. We aimed to identify phosSNPs-regulated gene network/pathway potentially significant for RA. METHODS: We collected genome-wide phosSNP genotyping data and transcriptome-wide mRNA expression data from PBMCs of a Chinese sample. We discovered and verified with public datasets differentially expressed genes (DEGs) associated with RA, and replicated RA-associated SNPs in our study sample. We performed a targeted expression quantitative trait locus (eQTL) study on significant phosSNPs and DEGs. RESULTS: We identified 29 nominally significant eQTL phosSNPs and 83 target genes, and constructed comprehensive regulatory/interaction networks, highlighting the vital effects of two eQTL phosSNPs (rs371513 and rs4824675, FDR <0.05) and four critical node genes (HSPA4, NDUFA2, MRPL15, and ATP5O). Besides, two node/key genes NDUFA2 and ATP5O, regulated by rs371513, were significantly enriched in mitochondrial oxidative phosphorylation pathway. Besides, four pairs of eQTL effects were replicated independently in whole blood and/or transformed fibroblasts. CONCLUSIONS: The findings delineated a potential role of protein phosphorylation and genetic variations in RA and warranted the significant roles of phosSNPs in regulating RA-associated genes expression in PBMCs. The results pointed out the relevance and significance of oxidative phosphorylation pathway to RA.

Deciphering transcriptional regulators of banana fruit ripening by regulatory network analysis.

Fruit ripening is a critical phase in the production and marketing of fruits. Previous studies have indicated that fruit ripening is a highly coordinated process, mainly regulated at the transcriptional level, in which transcription factors play essential roles. Thus, identifying key transcription factors regulating fruit ripening as well as their associated regulatory networks promises to contribute to a better understanding of fruit ripening. In this study, temporal gene expression analyses were performed to investigate banana fruit ripening with the aim to discern the global architecture of gene regulatory networks underlying fruit ripening. Eight time points were profiled covering dynamic changes of phenotypes, the associated physiology and levels of known ripening marker genes. Combining results from a weighted gene co-expression network analysis (WGCNA) as well as cis-motif analysis and supported by EMSA, Y1H, tobacco-, banana-transactivation experimental results, the regulatory network of banana fruit ripening was constructed, from which 25 transcription factors were identified as prime candidates to regulate the ripening process by modulating different ripening-related pathways. Our study presents the first global view of the gene regulatory network involved in banana fruit ripening, which may provide the basis for a targeted manipulation of fruit ripening to attain higher banana and loss-reduced banana commercialization.

Mendelian randomization analysis revealed potential causal factors for systemic lupus erythematosus.

Genome-wide association studies (GWAS) have identified many loci for systemic lupus erythematosus (SLE). However, identification of functionally relevant genes remains a challenge. The aim of this study was to highlight potential causal genes for SLE in the GWAS loci. By applying Mendelian randomization (MR) methods, such as summary data-based MR (SMR), generalized SMR and MR pleiotropy residual sum and outlier, we identified DNA methylations in 15 loci and mRNA expression of 21 genes that were causally associated with SLE. The identified genes enriched in 14 specific KEGG pathways (e.g. SLE, viral carcinogenesis) and two GO terms (interferon-γ-mediated signaling pathway and innate immune response). Among the identified genes, UBE2L3 and BLK variants were significantly associated with UBE2L3 and BLK methylations and gene expressions, respectively. UBE2L3 was up-regulated in SLE patients in several types of immune cells. Methylations (e.g. cg06850285) and mRNA expression of UBE2L3 were causally associated with SLE. Methylation site cg09528494 and mRNA expression of BLK were causally associated with SLE. BLK single nucleotide polymorphisms that were significantly associated with SLE were strongly associated with plasma cathepsin B level. Deep analysis identified that plasma cathepsin B level was causally associated with SLE. In summary, this study identified hundreds of DNA methylations and genes as potential risk factors for SLE. Genetic variants in UBE2L3 gene might affect SLE by influencing gene expression. Genetic variants in BLK gene might affect SLE by influencing BLK gene expression and plasma cathepsin B protein level.

Autosomal dominant hereditary spastic paraplegia caused by mutation of UBAP1.

Hereditary spastic paraplegias (HSP) are a group of rare neurodegenerative diseases characterized by progressive spastic paraparesis. UBAP1 was recently found to induce a rare type of HSP (SPG80). We identified a family with eight inherited spastic paraplegic patients carrying a novel heterozygous mutation c.279delG (p.S94Vfs*9) of UBAP1. We demonstrated a lack of functional UBAP1 in these patients, resulting in the neurological disorder caused by interceptions of the ESCRT pathway. Extending from the older onset-age identified from this family, we found that comparing with the European and other populations, Asian patients displayed less proportion of severe patients and an older average age at onset. The origins of SPG80 patients associated with both their onset age and their disease severity, while the age at onset was not correlated with the disease severity.

LION LBD: a literature-based discovery system for cancer biology.

MOTIVATION: The overwhelming size and rapid growth of the biomedical literature make it impossible for scientists to read all studies related to their work, potentially leading to missed connections and wasted time and resources. Literature-based discovery (LBD) aims to alleviate these issues by identifying implicit links between disjoint parts of the literature. While LBD has been studied in depth since its introduction three decades ago, there has been limited work making use of recent advances in biomedical text processing methods in LBD. RESULTS: We present LION LBD, a literature-based discovery system that enables researchers to navigate published information and supports hypothesis generation and testing. The system is built with a particular focus on the molecular biology of cancer using state-of-the-art machine learning and natural language processing methods, including named entity recognition and grounding to domain ontologies covering a wide range of entity types and a novel approach to detecting references to the hallmarks of cancer in text. LION LBD implements a broad selection of co-occurrence based metrics for analyzing the strength of entity associations, and its design allows real-time search to discover indirect associations between entities in a database of tens of millions of publications while preserving the ability of users to explore each mention in its original context in the literature. Evaluations of the system demonstrate its ability to identify undiscovered links and rank relevant concepts highly among potential connections. AVAILABILITY AND IMPLEMENTATION: The LION LBD system is available via a web-based user interface and a programmable API, and all components of the system are made available under open licenses from the project home page http://lbd.lionproject.net. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Phosphorylation-related SNPs influence lipid levels and rheumatoid arthritis risk by altering gene expression and plasma protein levels.

OBJECTIVES: Phosphorylation-related single-nucleotide polymorphisms (phosSNPs) are missense SNPs that may influence protein phosphorylation. The aim of this study was to evaluate the effect of phosSNPs on lipid levels and RA. METHODS: We examined the association of phosSNPs with lipid levels and RA in large-scale genome-wide association studies (GWAS) and performed random sampling and fgwas analyses to determine whether the phosSNPs associated with lipid levels and RA were significantly enriched. Furthermore, we performed QTL analysis and Mendelian randomization analysis to obtain additional evidence to be associated with the identified phosSNPs and genes. RESULTS: We found 483 phosSNPs for lipid levels and 243 phosSNPs for RA in the GWAS loci (P < 1.0 × 10-5). SNPs associated with high-density lipoprotein cholesterol, low-density lipoprotein cholesterol, Total cholesterol (TC) and RA were significantly enriched with phosSNPs. Almost all of the identified phosSNPs showed expression quantitative trait loci (eQTL) effects. A total of 48 protein QTLs and 9 metabolite QTLs were found. The phosSNP rs3184504 (p.Trp262Arg) at SH2B3 was significantly associated with RA, SH2B3 expression level, and plasma levels of high-density lipoprotein cholesterol, low-density lipoprotein cholesterol, TC, hypoxanthine and 80 proteins, including beta-2-microglobulin. SH2B3 was differentially expressed between RA cases and controls in peripheral blood mononuclear cells and synovial tissues. Mendelian randomization analysis showed that SH2B3 expression level was significantly associated with TC level and RA. Plasma beta-2-microglobulin level was causally associated with high-density lipoprotein cholesterol, low-density lipoprotein cholesterol, TC levels and RA. CONCLUSION: The findings suggested that phosSNPs may play important roles in lipid metabolism and the pathological mechanisms of RA. PhosSNPs may influence lipid levels and RA risk by altering gene expression and plasma protein levels.

Banana MaBZR1/2 associate with MaMPK14 to modulate cell wall modifying genes during fruit ripening.

KEY MESSAGE: Banana MaBZR1/2 interact with MaMPK14 to enhance the transcriptional inhibition of cell wall modifying genes including MaEXP2, MaPL2 and MaXET5. Fruit ripening and softening, the major attributes to perishability in fleshy fruits, are modulated by various plant hormones and gene expression. Banana MaBZR1/2, the central transcription factors of brassinosteroid (BR) signaling, mediate fruit ripening through regulation of ethylene biosynthesis, but their possible roles in fruit softening as well as the underlying mechanisms remain to be determined. In this work, we found that MaBZR1/2 directly bound to and repressed the promoters of several cell wall modifying genes such as MaEXP2, MaPL2 and MaXET5, whose transcripts were elevated concomitant with fruit ripening. Moreover, yeast two-hybrid (Y2H) and bimolecular fluorescence complementation (BiFC) assays indicated that MaBZR1/2 physically interacted with a mitogen-activated protein kinase MaMPK14, and this interaction strengthened the MaBZR1/2-mediated transcriptional inhibitory abilities. Collectively, our study provides insight into the mechanism of MaBZR1/2 contributing to fruit ripening and softening, which may have potential for banana molecular improvement.

MaMPK2 enhances MabZIP93-mediated transcriptional activation of cell wall modifying genes during banana fruit ripening.

Transcriptional regulation is an essential molecular machinery in controlling gene expression in diverse plant developmental processes including fruit ripening. This involves the interaction of transcription factors (TFs) and promoters of target genes. In banana, although a number of fruit ripening-associated TFs have been characterized, their number is relatively small. Here we identified a nuclear-localized basic leucine zipper (bZIP) TF, MabZIP93, associated with banana ripening. MabZIP93 activated cell wall modifying genes MaPL2, MaPE1, MaXTH23 and MaXGT1 by directly binding to their promoters. Transient over-expression of MabZIP93 in banana fruit resulted in the increased expression of MaPL2, MaPE1, MaXTH23 and MaXGT1. Moreover, a mitogen-activated protein kinase MaMPK2 and MabZIP93 were found to interact with MabZIP93. The interaction of MabZIP93 with MaMPK2 enhanced MabZIP93 activation of cell wall modifying genes, which was likely due to the phosphorylation of MabZIP93 mediated by MaMPK2. Overall, this study shows that MaMPK2 interacts with and phosphorylates MabZIP93 to promote MabZIP93-mediated transcriptional activation of cell wall modifying genes, thereby expanding our understanding of gene networks associated with banana fruit ripening.

Rheumatoid arthritis-associated DNA methylation sites in peripheral blood mononuclear cells.

OBJECTIVES: To identify novel DNA methylation sites significant for rheumatoid arthritis (RA) and comprehensively understand their underlying pathological mechanism. METHODS: We performed (1) genome-wide DNA methylation and mRNA expression profiling in peripheral blood mononuclear cells from RA patients and health controls; (2) correlation analysis and causal inference tests for DNA methylation and mRNA expression data; (3) differential methylation genes regulatory network construction; (4) validation tests of 10 differential methylation positions (DMPs) of interest and corresponding gene expressions; (5) correlation between PARP9 methylation and its mRNA expression level in Jurkat cells and T cells from patients with RA; (6) testing the pathological functions of PARP9 in Jurkat cells. RESULTS: A total of 1046 DNA methylation positions were associated with RA. The identified DMPs have regulatory effects on mRNA expressions. Causal inference tests identified six DNA methylation-mRNA-RA regulatory chains (eg, cg00959259-PARP9-RA). The identified DMPs and genes formed an interferon-inducible gene interaction network (eg, MX1, IFI44L, DTX3L and PARP9). Key DMPs and corresponding genes were validated their differences in additional samples. Methylation of PARP9 was correlated with mRNA level in Jurkat cells and T lymphocytes isolated from patients with RA. The PARP9 gene exerted significant effects on Jurkat cells (eg, cell cycle, cell proliferation, cell activation and expression of inflammatory factor IL-2). CONCLUSIONS: This multistage study identified an interferon-inducible gene interaction network associated with RA and highlighted the importance of PARP9 gene in RA pathogenesis. The results enhanced our understanding of the important role of DNA methylation in pathology of RA.

MaBZR1/2 act as transcriptional repressors of ethylene biosynthetic genes in banana fruit.

Banana fruit (Musa acuminate L.) ripening is a complex genetical process affected by multiple phytohormones and expression of various genes. However, whether plant hormone brassinosteroid (BR) is involved in this process remains obscure. In this work, three genes that encode BR core signaling components brassinazole resistant (BZR) proteins, namely MaBZR1 to MaBZR3, were characterized from banana fruit. MaBZR1-MaBZR3 exhibited both nuclear and cytoplasmic localization and behaved as transcription inhibitors. Expression analysis showed that MaBZR1/2/3 were continuously decreased as fruit ripening proceeded, indicating their negative roles in banana ripening. Moreover, gel shift and transient expression assays demonstrated that MaBZR1/2 could suppress the transcription of ethylene biosynthetic genes, including MaACS1, MaACO13 and MaACO14, which increased gradually during the banana ripening, via specifically binding to CGTGT/CG sequence in their promoters. Importantly, exogenous application of BRs promotes banana ripening, which is presumably due to the accelerated expression of MaACS1 and MaACO13/14, and consequently the ethylene production. Our study indicates that MaBZR1/2 act as transcriptional repressors of ethylene biosynthetic genes during banana fruit ripening.

Neural networks for link prediction in realistic biomedical graphs: a multi-dimensional evaluation of graph embedding-based approaches.

BACKGROUND: Link prediction in biomedical graphs has several important applications including predicting Drug-Target Interactions (DTI), Protein-Protein Interaction (PPI) prediction and Literature-Based Discovery (LBD). It can be done using a classifier to output the probability of link formation between nodes. Recently several works have used neural networks to create node representations which allow rich inputs to neural classifiers. Preliminary works were done on this and report promising results. However they did not use realistic settings like time-slicing, evaluate performances with comprehensive metrics or explain when or why neural network methods outperform. We investigated how inputs from four node representation algorithms affect performance of a neural link predictor on random- and time-sliced biomedical graphs of real-world sizes (∼ 6 million edges) containing information relevant to DTI, PPI and LBD. We compared the performance of the neural link predictor to those of established baselines and report performance across five metrics. RESULTS: In random- and time-sliced experiments when the neural network methods were able to learn good node representations and there was a negligible amount of disconnected nodes, those approaches outperformed the baselines. In the smallest graph (∼ 15,000 edges) and in larger graphs with approximately 14% disconnected nodes, baselines such as Common Neighbours proved a justifiable choice for link prediction. At low recall levels (∼ 0.3) the approaches were mostly equal, but at higher recall levels across all nodes and average performance at individual nodes, neural network approaches were superior. Analysis showed that neural network methods performed well on links between nodes with no previous common neighbours; potentially the most interesting links. Additionally, while neural network methods benefit from large amounts of data, they require considerable amounts of computational resources to utilise them. CONCLUSIONS: Our results indicate that when there is enough data for the neural network methods to use and there are a negligible amount of disconnected nodes, those approaches outperform the baselines. At low recall levels the approaches are mostly equal but at higher recall levels and average performance at individual nodes, neural network approaches are superior. Performance at nodes without common neighbours which indicate more unexpected and perhaps more useful links account for this.

Identification of expression quantitative trait loci (eQTLs) in human peripheral blood mononuclear cells (PBMCs) and shared with liver and brain.

PBMCs are essential for immunity and involved in various diseases. To identify genetic variations contributing to PBMCs transcriptome-wide gene expression, we performed a genome-wide eQTL analysis by using genome-wide SNPs data and transcriptome-wide mRNA expression data. To assess whether there are common regulation patterns shared among different tissues/organs, public datasets were utilized to identify common eQTLs shared with PBMCs in lymphoblastoid, monocytes, liver, and brain. Allelic expression imbalance (AEI) assay was employed to validate representative eQTLs identified. We identified 443 cis- and 2386 trans-eSNPs (FDR <0.05), which regulated 128 and 635 target genes, respectively. A transcriptome-wide expression regulation network was constructed, highlighting the importance of 28 pleiotropic eSNPs and 18 dually (cis- and trans-) regulated genes. Three genes, that is, TIPRL, HSPB8, and EGLN3, were commonly regulated by hundreds of eSNPs and constituted a very complex interaction network. Strikingly, the missense SNP rs371513 trans- regulated 25 target genes, which were functionally related to poly(A) RNA binding. Among 8904 eQTLs (P < 0.001) identified herein in PBMCs, a minority (163) was overlapped with lymphoblastoid, monocytes, liver, and/or brain. Besides, two cis-eSNPs in PBMC were confirmed by AEI. The present results demonstrated a comprehensive expression regulation network for human PBMCs and may provide novel insights into the pathogenesis of immunological diseases related to PBMCs.

Multiple correlation analyses revealed complex relationship between DNA methylation and mRNA expression in human peripheral blood mononuclear cells.

DNA methylation is an important regulator on the mRNA expression. However, a genome-wide correlation pattern between DNA methylation and mRNA expression in human peripheral blood mononuclear cells (PBMCs) is largely unknown. The comprehensive relationship between mRNA and DNA methylation was explored by using four types of correlation analyses and a genome-wide methylation-mRNA expression quantitative trait locus (eQTL) analysis in PBMCs in 46 unrelated female subjects. An enrichment analysis was performed to detect biological function for the detected genes. Single pair correlation coefficient (r T1) between methylation level and mRNA is moderate (-0.63-0.62) in intensity, and the negative and positive correlations are nearly equal in quantity. Correlation analysis on each gene (T4) found 60.1% genes showed correlations between mRNA and gene-based methylation at P < 0.05 and more than 5.96% genes presented very strong correlation (R T4 > 0.8). Methylation sites have regulation effects on mRNA expression in eQTL analysis, with more often observations in region of transcription start site (TSS). The genes under significant methylation regulation both in correlation analysis and eQTL analysis tend to cluster to the categories (e.g., transcription, translation, regulation of transcription) that are essential for maintaining the basic life activities of cells. Our findings indicated that DNA methylation has predictive regulation effect on mRNA with a very complex pattern in PBMCs. The results increased our understanding on correlation of methylation and mRNA and also provided useful clues for future epigenetic studies in exploring biological and disease-related regulatory mechanisms in PBMC.