Identifying glioblastoma gene networks based on hypergeometric test analysis

Meta-Analysis of RNA-Seq Datasets Identifies Novel Players in Glioblastoma

Glioblastoma is a devastating grade IV glioma with poor prognosis. Identification of predictive molecular biomarkers of disease progression would substantially contribute to better disease management. In the current study, we performed a meta-analysis of different RNA-seq datasets to identify differentially expressed protein-coding genes (PCGs) and long non-coding RNAs (lncRNAs). This meta-analysis aimed to improve power and reproducibility of the individual studies while identifying overlapping disease-relevant pathways. We supplemented the meta-analysis with small RNA-seq on glioblastoma tissue samples to provide an overall transcriptomic view of glioblastoma. Co-expression correlation of filtered differentially expressed PCGs and lncRNAs identified a functionally relevant sub-cluster containing DANCR and SNHG6, with two novel lncRNAs and two novel PCGs. Small RNA-seq of glioblastoma tissues identified five differentially expressed microRNAs of which three interacted with the functionally relevant sub-cluster. Pathway analysis of this sub-cluster identified several glioblastoma-linked pathways, which were also previously associated with the novel cell death pathway, ferroptosis. In conclusion, the current meta-analysis strengthens evidence of an overarching involvement of ferroptosis in glioblastoma pathogenesis and also suggests some candidates for further analyses.

Revealing the Molecular Mechanisms of Alzheimer's Disease Based on Network Analysis

The complex pathology of Alzheimer's disease (AD) emphasises the need for comprehensive modelling of the disease, which may lead to the development of efficient treatment strategies. To address this challenge, we analysed transcriptome data of post-mortem human brain samples of healthy elders and individuals with late-onset AD from the Religious Orders Study and Rush Memory and Aging Project (ROSMAP) and Mayo Clinic (MayoRNAseq) studies in the AMP-AD consortium. In this context, we conducted several bioinformatics and systems medicine analyses including the construction of AD-specific co-expression networks and genome-scale metabolic modelling of the brain in AD patients to identify key genes, metabolites and pathways involved in the progression of AD. We identified AMIGO1 and GRPRASP2 as examples of commonly altered marker genes in AD patients. Moreover, we found alterations in energy metabolism, represented by reduced oxidative phosphorylation and ATPase activity, as well as the depletion of hexanoyl-CoA, pentanoyl-CoA, (2E)-hexenoyl-CoA and numerous other unsaturated fatty acids in the brain. We also observed that neuroprotective metabolites (e.g., vitamins, retinoids and unsaturated fatty acids) tend to be depleted in the AD brain, while neurotoxic metabolites (e.g., β-alanine, bilirubin) were more abundant. In summary, we systematically revealed the key genes and pathways related to the progression of AD, gained insight into the crucial mechanisms of AD and identified some possible targets that could be used in the treatment of AD.

Separating the Wheat from the Chaff: The Use of Upstream Regulator Analysis to Identify True Differential Expression of Single Genes within Transcriptomic Datasets

The development of DNA microarray and RNA-sequencing technology has led to an explosion in the generation of transcriptomic differential expression data under a wide range of biologic systems including those recapitulating the monogenic muscular dystrophies. Data generation has increased exponentially due in large part to new platforms, improved cost-effectiveness, and processing speed. However, reproducibility and thus reliability of data remain a central issue, particularly when resource constraints limit experiments to single replicates. This was observed firsthand in a recent rare disease drug repurposing project involving RNA-seq-based transcriptomic profiling of primary cerebrocortical cultures incubated with clinic-ready blood-brain penetrant drugs. Given the low validation rates obtained for single differential expression genes, alternative approaches to identify with greater confidence genes that were truly differentially expressed in our dataset were explored. Here we outline a method for differential expression data analysis in the context of drug repurposing for rare diseases that incorporates the statistical rigour of the multigene analysis to bring greater predictive power in assessing individual gene modulation. Ingenuity Pathway Analysis upstream regulator analysis was applied to the differentially expressed genes from the Care4Rare Neuron Drug Screen transcriptomic database to identify three distinct signaling networks each perturbed by a different drug and involving a central upstream modulating protein: levothyroxine (DIO3), hydroxyurea (FOXM1), dexamethasone (PPARD). Differential expression of upstream regulator network related genes was next assessed in in vitro and in vivo systems by qPCR, revealing 5× and 10× increases in validation rates, respectively, when compared with our previous experience with individual genes in the dataset not associated with a network. The Ingenuity Pathway Analysis based gene prioritization may increase the predictive value of drug-gene interactions, especially in the context of assessing single-gene modulation in single-replicate experiments.

An integrated systematic approach for investigating microcurrent electrical nerve stimulation (MENS) efficacy in STZ-induced diabetes mellitus

Diabetes mellitus (DM) is a major metabolic disorder and an increasing health problem worldwide. Effective non-invasive therapies for DM are still lacking. Here, we have developed Microcurrent electrical nerve stimulation (MENS), a non-invasive therapy, and tested on 46 mice clustered into five groups, such as control, STZ-induced DM, and MENS treatment groups. Experimental results show that MENS treatment is able to improve seven biochemical indexes (e.g., hemoglobin A1c and glucose level). To investigate the mechanisms of MENS treatment on STZ-induced DM, we selected six representative samples to perform microarray experiments for several groups and developed an integrated Hierarchical System Biology Model (HiSBiM) to analyze these omics data. The results indicate that MENS can affect fatty acid metabolism pathways, peroxisome proliferator-activated receptor (PPAR) signaling pathway and cell cycle. Additionally, the DM biochemical indexes and omics data profiles of MENS treatment were found to be consistent. We then compared the therapeutic effects of MENS with anti-diabetic compounds (e.g., quercetin, metformin, and rosiglitazone), using the HiSBiM four-level biological functions and processes of multiple omics data. The results show MENS and these anti-diabetic compounds have similar effect pathways highly correlated to the diabetes processes, such as the PPAR signaling pathway, bile secretion, and insulin signaling pathways. We believe that MENS is an effective and non-invasive therapy for DM and our HiSBiM is an useful method for investigating multiple omics data.

Disentangling the therapeutic tactics in GBM: From bench to bedside and beyond

Glioblastoma multiforme (GBM) is one of the most common and malignant form of adult brain tumor with a high mortality rate and dismal prognosis. The present standard treatment comprising surgical resection followed by radiation and chemotherapy using temozolomide can broaden patient's survival to some extent. However, the advantages are not palliative due to the development of resistance to the drug and tumor recurrence following the multimodal treatment approaches due to both intra- and intertumoral heterogeneity of GBM. One of the major contributors to temozolomide resistance is O⁶ -methylguanine-DNA methyltransferase. Furthermore, deficiency of mismatch repair, base excision repair, and cytoprotective autophagy adds to temozolomide obstruction. Rising proof additionally showed that a small population of cells displaying certain stem cell markers, known as glioma stem cells, adds on to the resistance and tumor progression. Collectively, these findings necessitate the discovery of novel therapeutic avenues for treating glioblastoma. As of late, after understanding the pathophysiology and biology of GBM, some novel therapeutic discoveries, such as drug repurposing, targeted molecules, immunotherapies, antimitotic therapies, and microRNAs, have been developed as new potential treatments for glioblastoma. To help illustrate, "what are the mechanisms of resistance to temozolomide" and "what kind of alternative therapeutics can be suggested" with this fatal disease, a detailed history of these has been discussed in this review article, all with a hope to develop an effective treatment strategy for GBM.

Characterization of the hepatic transcriptome following phenobarbital induction in mice with AIP

Acute Intermittent Porphyria (AIP), an autosomal dominant hepatic disorder, results from hydroxymethylbilane synthase (HMBS) mutations that decrease the encoded enzymatic activity, thereby predisposing patients to life-threatening acute neurovisceral attacks. The ~1% penetrance of AIP suggests that other genetic factors modulate the onset and severity of the acute attacks. Here, we characterized the hepatic transcriptomic response to phenobarbital (PB) administration in AIP mice, which mimics the biochemical attacks of AIP. At baseline, the mRNA profiles of 14,138 hepatic genes prior to treatment were remarkably similar between AIP and the congenic wild-type (WT) mice. After PB treatment (~120 mg/kg x 3d), 1347 and 1120 genes in AIP mice and 422 and 404 genes in WT mice were uniquely up- and down-regulated, respectively, at a False Discovery Rate < 0.05. As expected, the ALAS1 expression increased 4.5-fold and 15.9-fold in the WT and AIP mice, respectively. ALA-dehydrogenase also was induced ~1.7-fold in PB-induced AIP mice, but was unchanged in PB-induced WT mice. There was no statistically significant difference in the overall expression of 155 hepatic cytochrome P450 enzymes, although Cyp2c40, Cyp2c68, Cyp2c69, Mgst3 were upregulated only in PB-induced AIP mice (>1.9-fold) and Cyp21a1 was upregulated only in PB-induced WT mice (>9-fold). Notably, the genes differentially expressed in induced AIP mice were enriched in circadian rhythm, mitochondria biogenesis and electron transport, suggesting these pathways were involved in AIP mice responding to PB treatment. These results advance our understanding of the hepatic metabolic changes in PB-induced AIP mice and have implications in the pathogenesis of AIP acute attacks.

A survey of transcriptome complexity using PacBio single-molecule real-time analysis combined with Illumina RNA sequencing for a better understanding of ricinoleic acid biosynthesis in Ricinus communis

Background

Ricinus communis is a highly economically valuable oil crop plant from the spurge family, Euphorbiaceae. However, the available reference genomes are incomplete and to date studies on ricinoleic acid biosynthesis at the transcriptional level are limited.

Results

In this study, we combined PacBio single-molecule long read isoform and Illumina RNA sequencing to identify the alternative splicing (AS) events, novel isoforms, fusion genes, long non-coding RNAs (lncRNAs) and alternative polyadenylation (APA) sites to unveil the transcriptomic complexity of castor beans and identify critical genes related to ricinoleic acid biosynthesis. Here, we identified 11,285 AS-variants distributed in 21,448 novel genes and detected 520 fusion genes, 320 lncRNAs and 9511 (APA-sites). Furthermore, a total of 6067, 5983 and 4058 differentially expressed genes between developing beans of the R. communis lines 349 and 1115 with extremely different oil content were identified at 7, 14 and 21 days after flowering, respectively. Specifically, 14, 18 and 11 DEGs were annotated encoding key enzymes related to ricinoleic acid biosynthesis reflecting the higher castor oil content of 1115 compared than 349. Quantitative real-time RT-PCR further validated fifteen of these DEGs at three-time points.

Conclusion

Our results significantly improved the existed gene models of R. communis, and a putative model of key genes was built to show the differences between strains 349 and 1115, illustrating the molecular mechanism of castor oil biosynthesis. A multi-transcriptome database and candidate genes were provided to further improve the level of ricinoleic acid in transgenic crops.

Electronic supplementary material

The online version of this article (10.1186/s12864-019-5832-9) contains supplementary material, which is available to authorized users.

Drug Repositioning in Glioblastoma: A Pathway Perspective

Glioblastoma multiforme (GBM) is the most malignant primary adult brain tumor. The current standard of care is surgical resection, radiation, and chemotherapy treatment, which extends life in most cases. Unfortunately, tumor recurrence is nearly universal and patients with recurrent glioblastoma typically survive <1 year. Therefore, new therapies and therapeutic combinations need to be developed that can be quickly approved for use in patients. However, in order to gain approval, therapies need to be safe as well as effective. One possible means of attaining rapid approval is repurposing FDA approved compounds for GBM therapy. However, candidate compounds must be able to penetrate the blood-brain barrier (BBB) and therefore a selection process has to be implemented to identify such compounds that can eliminate GBM tumor expansion. We review here psychiatric and non-psychiatric compounds that may be effective in GBM, as well as potential drugs targeting cell death pathways. We also discuss the potential of data-driven computational approaches to identify compounds that induce cell death in GBM cells, enabled by large reference databases such as the Library of Integrated Network Cell Signatures (LINCS). Finally, we argue that identifying pathways dysregulated in GBM in a patient specific manner is essential for effective repurposing in GBM and other gliomas.

Clinical aggressiveness of malignant gliomas is linked to augmented metabolism of amino acids

Glutamine, glutamate, asparagine, and aspartate are involved in an enzyme-network that controls nitrogen metabolism. Branched-chain-amino-acid aminotransferase-1 (BCAT1) promotes proliferation of gliomas with wild-type IDH1 and is closely connected to the network. We hypothesized that metabolism of asparagine, glutamine, and branched-chain-amino-acids is associated with progression of malignant gliomas. Gene expression for asparagine synthetase (ASNS), glutaminase (GLS), and BCAT1 were analyzed in 164 gliomas from 156 patients [33-anaplastic gliomas (AG) and 131-glioblastomas (GBM), 64 of which were recurrent GBMs]. ASNS and GLS were twofold higher in GBMs versus AGs. BCAT1 was also higher in GBMs. ASNS expression was twofold higher in recurrent versus new GBMs. Five patients had serial samples: 4-showed higher ASNS and 3-higher GLS at recurrence. We analyzed grade and treatment in 4 groups: (1) low ASNS, GLS, and BCAT1 (n = 96); (2) low ASNS and GLS, but high BCAT1 (n = 26); (3) high ASNS or GLS, but low BCAT1 (n = 25); and (4) high ASNS or GLS and high BCAT1 (n = 17). Ninety-one  % of patients (29/32) with grade-III lesions were in group 1. In contrast, 95 % of patients (62/65) in groups 2-4 had GBMs. Treatment was similar in 4 groups (radiotherapy-80 %; temozolomide-30 %; other chemotherapy-50 %). High expression of ASNS, GLS, and BCAT1 were each associated with poor survival in the entire group. The combination of lower ASNS, GLS, and BCAT1 levels correlated with better survival for newly diagnosed GBMs (66 patients; P = 0.0039). Only tumors with lower enzymes showed improved outcome with temozolomide. IDH1(WT) gliomas had higher expression of these genes. Manipulation of amino acid metabolism in malignant gliomas may be further studied for therapeutics development.

Molecular classification and prediction in gastric cancer

Gastric cancer, a highly heterogeneous disease, is the second leading cause of cancer death and the fourth most common cancer globally, with East Asia accounting for more than half of cases annually. Alongside TNM staging, gastric cancer clinic has two well-recognized classification systems, the Lauren classification that subdivides gastric adenocarcinoma into intestinal and diffuse types and the alternative World Health Organization system that divides gastric cancer into papillary, tubular, mucinous (colloid), and poorly cohesive carcinomas. Both classification systems enable a better understanding of the histogenesis and the biology of gastric cancer yet have a limited clinical utility in guiding patient therapy due to the molecular heterogeneity of gastric cancer. Unprecedented whole-genome-scale data have been catalyzing and advancing the molecular subtyping approach. Here we cataloged and compared those published gene expression profiling signatures in gastric cancer. We summarized recent integrated genomic characterization of gastric cancer based on additional data of somatic mutation, chromosomal instability, EBV virus infection, and DNA methylation. We identified the consensus patterns across these signatures and identified the underlying molecular pathways and biological functions. The identification of molecular subtyping of gastric adenocarcinoma and the development of integrated genomics approaches for clinical applications such as prediction of clinical intervening emerge as an essential phase toward personalized medicine in treating gastric cancer.

A null model for Pearson coexpression networks

Gene coexpression networks inferred by correlation from high-throughput profiling such as microarray data represent simple but effective structures for discovering and interpreting linear gene relationships. In recent years, several approaches have been proposed to tackle the problem of deciding when the resulting correlation values are statistically significant. This is most crucial when the number of samples is small, yielding a non-negligible chance that even high correlation values are due to random effects. Here we introduce a novel hard thresholding solution based on the assumption that a coexpression network inferred by randomly generated data is expected to be empty. The threshold is theoretically derived by means of an analytic approach and, as a deterministic independent null model, it depends only on the dimensions of the starting data matrix, with assumptions on the skewness of the data distribution compatible with the structure of gene expression levels data. We show, on synthetic and array datasets, that the proposed threshold is effective in eliminating all false positive links, with an offsetting cost in terms of false negative detected edges.