Inferring interaction type in gene regulatory networks using co-expression data

Proteomic Atlas of Atherosclerosis: The Contribution of Proteoglycans to Sex Differences, Plaque Phenotypes, and Outcomes

BACKGROUND

Using proteomics, we aimed to reveal molecular types of human atherosclerotic lesions and study their associations with histology, imaging, and cardiovascular outcomes.

METHODS

Two hundred nineteen carotid endarterectomy samples were procured from 120 patients. A sequential protein extraction protocol was employed in conjunction with multiplexed, discovery proteomics. To focus on extracellular proteins, parallel reaction monitoring was employed for targeted proteomics. Proteomic signatures were integrated with bulk, single-cell, and spatial RNA-sequencing data, and validated in 200 patients from the Athero-Express Biobank study.

RESULTS

This extensive proteomics analysis identified plaque inflammation and calcification signatures, which were inversely correlated and validated using targeted proteomics. The inflammation signature was characterized by the presence of neutrophil-derived proteins, such as S100A8/9 (calprotectin) and myeloperoxidase, whereas the calcification signature included fetuin-A, osteopontin, and gamma-carboxylated proteins. The proteomics data also revealed sex differences in atherosclerosis, with large-aggregating proteoglycans versican and aggrecan being more abundant in females and exhibiting an inverse correlation with estradiol levels. The integration of RNA-sequencing data attributed the inflammation signature predominantly to neutrophils and macrophages, and the calcification and sex signatures to smooth muscle cells, except for certain plasma proteins that were not expressed but retained in plaques, such as fetuin-A. Dimensionality reduction and machine learning techniques were applied to identify 4 distinct plaque phenotypes based on proteomics data. A protein signature of 4 key proteins (calponin, protein C, serpin H1, and versican) predicted future cardiovascular mortality with an area under the curve of 75% and 67.5% in the discovery and validation cohort, respectively, surpassing the prognostic performance of imaging and histology.

CONCLUSIONS

Plaque proteomics redefined clinically relevant patient groups with distinct outcomes, identifying subgroups of male and female patients with elevated risk of future cardiovascular events.

Insights into the Synergistic Antibacterial Activity of Silver Nitrate with Potassium Tellurite against Pseudomonas aeruginosa

The constant, ever-increasing antibiotic resistance crisis leads to the announcement of "urgent, novel antibiotics needed" by the World Health Organization. Our previous works showed a promising synergistic antibacterial activity of silver nitrate with potassium tellurite out of thousands of other metal/metalloid-based antibacterial combinations. The silver-tellurite combined treatment not only is more effective than common antibiotics but also prevents bacterial recovery, decreases the risk of future resistance chance, and decreases the effective concentrations. We demonstrate that the silver-tellurite combination is effective against clinical isolates. Further, this study was conducted to address knowledge gaps in the available data on the antibacterial mechanism of both silver and tellurite, as well as to give insight into how the mixture provides synergism as a combination. Here, we defined the differentially expressed gene profile of Pseudomonas aeruginosa under silver, tellurite, and silver-tellurite combination stress using an RNA sequencing approach to examine the global transcriptional changes in the challenged cultures grown in simulated wound fluid. The study was complemented with metabolomics and biochemistry assays. Both metal ions mainly affected four cellular processes, including sulfur homeostasis, reactive oxygen species response, energy pathways, and the bacterial cell membrane (for silver). Using a Caenorhabditis elegans animal model we showed silver-tellurite has reduced toxicity over individual metal/metalloid salts and provides increased antioxidant properties to the host. This work demonstrates that the addition of tellurite would improve the efficacy of silver in biomedical applications. IMPORTANCE Metals and/or metalloids could represent antimicrobial alternatives for industrial and clinical applications (e.g., surface coatings, livestock, and topical infection control) because of their great properties, such as good stability and long half-life. Silver is the most common antimicrobial metal, but resistance prevalence is high, and it can be toxic to the host above a certain concentration. We found that a silver-tellurite composition has antibacterial synergistic effect and that the combination is beneficial to the host. So, the efficacy and application of silver could increase by adding tellurite in the recommended concentration(s). We used different methods to evaluate the mechanism for how this combination can be so incredibly synergistic, leading to efficacy against antibiotic- and silver-resistant isolates. Our two main findings are that (i) both silver and tellurite mostly target the same pathways and (ii) the coapplication of silver with tellurite tends not to target new pathways but targets the same pathways with an amplified change.

Limb Mesoderm and Head Ectomesenchyme Both Express a Core Transcriptional Program During Chondrocyte Differentiation

To explain how cartilage appeared in different parts of the vertebrate body at discrete times during evolution, we hypothesize that different embryonic populations co-opted expression of a core gene regulatory network (GRN) driving chondrocyte differentiation. To test this hypothesis, laser-capture microdissection coupled with RNA-seq was used to reveal chondrocyte transcriptomes in the developing chick humerus and ceratobranchial, which are mesoderm- and neural crest-derived, respectively. During endochondral ossification, two general types of chondrocytes differentiate. Immature chondrocytes (IMM) represent the early stages of cartilage differentiation, while mature chondrocytes (MAT) undergo additional stages of differentiation, including hypertrophy and stimulating matrix mineralization and degradation. Venn diagram analyses generally revealed a high degree of conservation between chondrocyte transcriptomes of the limb and head, including SOX9, COL2A1, and ACAN expression. Typical maturation genes, such as COL10A1, IBSP, and SPP1, were upregulated in MAT compared to IMM in both limb and head chondrocytes. Gene co-expression network (GCN) analyses of limb and head chondrocyte transcriptomes estimated the core GRN governing cartilage differentiation. Two discrete portions of the GCN contained genes that were differentially expressed in limb or head chondrocytes, but these genes were enriched for biological processes related to limb/forelimb morphogenesis or neural crest-dependent processes, respectively, perhaps simply reflecting the embryonic origin of the cells. A core GRN driving cartilage differentiation in limb and head was revealed that included typical chondrocyte differentiation and maturation markers, as well as putative novel "chondrocyte" genes. Conservation of a core transcriptional program during chondrocyte differentiation in both the limb and head suggest that the same core GRN was co-opted when cartilage appeared in different regions of the skeleton during vertebrate evolution.

Estimation of Gene Regulatory Networks from Cancer Transcriptomics Data

Cancer is a genetic disease in which multiple genes are perturbed. Thus, information about the regulatory relationships between genes is necessary for the identification of biomarkers and therapeutic targets. In this review, methods for inference of gene regulatory networks (GRNs) from transcriptomics data that are used in cancer research are introduced. The methods are classified into three categories according to the analysis model. The first category includes methods that use pair-wise measures between genes, including correlation coefficient and mutual information. The second category includes methods that determine the genetic regulatory relationship using multivariate measures, which consider the expression profiles of all genes concurrently. The third category includes methods using supervised and integrative approaches. The supervised approach estimates the regulatory relationship using a supervised learning method that constructs a regression or classification model for predicting whether there is a regulatory relationship between genes with input data of gene expression profiles and class labels of prior biological knowledge. The integrative method is an expansion of the supervised method and uses more data and biological knowledge for predicting the regulatory relationship. Furthermore, simulation and experimental validation of the estimated GRNs are also discussed in this review. This review identified that most GRN inference methods are not specific for cancer transcriptome data, and such methods are required for better understanding of cancer pathophysiology. In addition, more systematic methods for validation of the estimated GRNs need to be developed in the context of cancer biology.

PoLoBag: Polynomial Lasso Bagging for signed gene regulatory network inference from expression data

MOTIVATION

Inferring gene regulatory networks (GRNs) from expression data is a significant systems biology problem. A useful inference algorithm should not only unveil the global structure of the regulatory mechanisms but also the details of regulatory interactions such as edge direction (from regulator to target) and sign (activation/inhibition). Many popular GRN inference algorithms cannot infer edge signs, and those that can infer signed GRNs cannot simultaneously infer edge directions or network cycles.

RESULTS

To address these limitations of existing algorithms, we propose Polynomial Lasso Bagging (PoLoBag) for signed GRN inference with both edge directions and network cycles. PoLoBag is an ensemble regression algorithm in a bagging framework where Lasso weights estimated on bootstrap samples are averaged. These bootstrap samples incorporate polynomial features to capture higher-order interactions. Results demonstrate that PoLoBag is consistently more accurate for signed inference than state-of-the-art algorithms on simulated and real-world expression datasets.

AVAILABILITY AND IMPLEMENTATION

Algorithm and data are freely available at https://github.com/gourabghoshroy/PoLoBag.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Identifying the degree of genetic interactions using Restricted Boltzmann Machine-A study on colorectal cancer

The phenomenon of two or more genes affecting the expression of each other in various ways in the development of a single character of an organism is known as gene interaction. Gene interaction not only applies to normal human traits but to the diseased samples as well. Thus, an analysis of gene interaction could help us to differentiate between the normal and the diseased samples or between the two/more phases any diseased samples. At the first stage of this work we have used restricted Boltzmann machine model to find such significant interactions present in normal and/or cancer samples of every gene pairs of 20 genes of colorectal cancer data set (GDS4382) along with the weight/degree of those interactions. Later on, we are looking for those interactions present in adenoma and/or carcinoma samples of the same 20 genes of colorectal cancer data set (GDS1777). The weight/degree of those interactions represents how strong/weak an interaction is. At the end we will create a gene regulatory network with the help of those interactions, where the regulatory genes are identified by using Naïve Bayes Classifier. Experimental results are validated biologically by comparing the interactions with NCBI databases.

Physiological RNA dynamics in RNA-Seq analysis

Physiological RNA dynamics cause problems in transcriptome analysis. Physiological RNA accumulation affects the analysis of RNA quantification, and physiological RNA degradation affects the analysis of the RNA sequence length, feature site and quantification. In the present article, we review the effects of physiological degradation and accumulation of RNA on analysing RNA sequencing data. Physiological RNA accumulation and degradation probably led to such phenomena as incorrect estimations of transcription quantification, differential expressions, co-expressions, RNA decay rates, alternative splicing, boundaries of transcription, novel genes, new single-nucleotide polymorphisms, small RNAs and gene fusion. Thus, the transcriptomic data obtained up to date warrant further scrutiny. New and improved techniques and bioinformatics software are needed to produce accurate data in transcriptome research.

Genetic regulatory subnetworks and key regulating genes in rat hippocampus perturbed by prenatal malnutrition: implications for major brain disorders

OBJECTIVE

Many population studies have shown that maternal prenatal nutrition deficiency may increase the risk of neurodevelopmental disorders in their offspring, but its potential transcriptomic effects on brain development are not clear. We aimed to investigate the transcriptional regulatory interactions between genes in particular pathways responding to the prenatal nutritional deficiency and to explore their effects on neurodevelopment and related disorders.

RESULTS

We identified three modules in rat hippocampus responding to maternal prenatal nutritional deficiency and found 15 key genes (Hmgn1, Ssbp1, LOC684988, Rpl23, Gga1, Rhobtb2, Dhcr24, Atg9a, Dlgap3, Grm5, Scn2b, Furin, Sh3kbp1, Ubqln1, and Unc13a) related to the rat hippocampus developmental dysregulation, of which Hmgn1, Rhobtb2 and Unc13a related to autism, and Dlgap3, Grm5, Furin and Ubqln1 are related to Alzheimer's disease, and schizophrenia. Transcriptional alterations of the hub genes were confirmed except for Atg9a. Additionally, through modeling miRNA-mRNA-transcription factor interactions for the hub genes, we confirmed a transcription factor, Cebpa, is essential to regulate the expression of Rhobtb2. We did not find singificent singals in the prefrontal cortex responding to maternal prenatal nutritional deficiency.

CONCLUSION

These findings demonstrated that these genes with the three modules in rat hippocampus involved in synaptic development, neuronal projection, cognitive function, and learning function are significantly enriched hippocampal CA1 pyramidal neurons and suggest that three genetic regulatory subnetworks and thirteen key regulating genes in rat hippocampus perturbed by a prenatal nutrition deficiency. These genes and related subnetworks may be prenatally involved in the etiologies of major brain disorders, including Alzheimer's disease, autism, and schizophrenia.

METHODS

We compared the transcriptomic differences in the hippocampus and prefrontal cortex between 10 rats with prenatal nutritional deficiency and 10 rats with prenatal normal chow feeding by differential analysis and co-expression network analysis. A network-driven integrative analysis with microRNAs and transcription factors was performed to define significant modules and hub genes responding to prenatal nutritional deficiency. Meanwhile, the module preservation test was conducted between the hippocampus and prefrontal cortex. Expression levels of the hub genes were further validated with a quantitative real-time polymerase chain reaction based on additional 40 pairs of rats.

Co-expression network analysis identified hub genes critical to triglyceride and free fatty acid metabolism as key regulators of age-related vascular dysfunction in mice

Background: Aging has often been linked to age-related vascular disorders. The elucidation of the putative genes and pathways underlying vascular aging likely provides useful insights into vascular diseases at advanced ages. Transcriptional regulatory network analysis is the key to describing genetic interactions between molecular regulators and their target gene transcriptionally changed during vascular aging.

Results: A total of 469 differentially expressed genes were parsed into 6 modules. Among the incorporated sample traits, the most significant module related to vascular aging was associated with triglyceride and enriched with biological terms like proteolysis, blood circulation, and circulatory system process. The module associated with triglyceride was preserved in an independent microarray dataset, indicating the robustness of the identified vascular aging-related subnetwork. Additionally, Enpp5, Fez1, Kif1a, F3, H2-Q7, and their interacting miRNAs mmu-miR-449a, mmu-miR-449c, mmu-miR-34c, mmu-miR-34b-5p, mmu-miR-15a, and mmu-let-7, exhibited the most connectivity with external lipid-related traits. Transcriptional alterations of the hub genes Enpp5, Fez1, Kif1a, and F3, and the interacting microRNAs mmu-miR-34c, mmu-miR-34b-5p, mmu-let-7, mmu-miR-449a, and mmu-miR-449c were confirmed.

Conclusion: Our findings demonstrate that triglyceride and free fatty acid-related genes are key regulators of age-related vascular dysfunction in mice and show that the hub genes for Enpp5, Fez1, Kif1a, and F3 as well as their interacting miRNAs mmu-miR-34c, mmu-miR-34b-5p, mmu-let-7, mmu-miR-449a, and mmu-miR-449c, could serve as potential biomarkers in vascular aging.

Methods: The microarray gene expression profiles of aorta samples from 6-month old mice (n=6) and 20-month old mice (n=6) were processed to identify nominal differentially expressed genes. These nominal differentially expressed genes were subjected to a weighted gene co-expression network analysis. A network-driven integrative analysis with microRNAs and transcription factors was performed to define significant modules and underlying regulatory pathways associated with vascular aging, and module preservation test was conducted to validate the age-related modules based on an independent microarray gene expression dataset in mice aorta samples including three 32-week old wild-type mice (around 6-month old) and three 78-week old wild-type mice (around 20-month old). Gene ontology and protein-protein interaction analyses were conducted to determine the hub genes as potential biomarkers in the progress of vascular aging. The hub genes were further validated with quantitative real-time polymerase chain reaction in aorta samples from 20 young (6-month old) mice and 20 old (20-month old) mice.

MERIT: Systematic Analysis and Characterization of Mutational Effect on RNA Interactome Topology

The interaction between RNA-binding proteins (RBPs) and RNA plays an important role in regulating cellular function. However, decoding genome-wide protein-RNA regulatory networks as well as how cancer-related mutations impair RNA regulatory activities in hepatocellular carcinoma (HCC) remains mostly undetermined. We explored the genetic alteration patterns of RBPs and found that deleterious mutations are likely to occur on the surface of RBPs. We then constructed protein-RNA interactome networks by integration of target binding screens and expression profiles. Network analysis highlights regulatory principles among interacting RBPs. In addition, somatic mutations selectively target functionally important genes (cancer genes, core fitness genes, or conserved genes) and perturb the RBP-gene regulatory networks in cancer. These regulatory patterns were further validated using independent data. A computational method (Mutational Effect on RNA Interactome Topology) and a web-based, user-friendly resource were further proposed to analyze the RBP-gene regulatory networks across cancer types. Pan-cancer analysis also suggests that cancer cells selectively target "vulnerability" genes to perturb protein-RNA interactome that is involved in cancer hallmark-related functions. Specifically, we experimentally validated four pairs of RBP-gene interactions perturbed by mutations in HCC, which play critical roles in cell proliferation. Based on the expression of perturbed RBP and target genes, we identified three subtypes of HCC with different survival rates. Conclusion: Our results provide a valuable resource for characterizing somatic mutation-perturbed protein-RNA regulatory networks in HCC, yielding valuable insights into the genotype-phenotype relationships underlying human cancer, and potential biomarkers for precision medicine.

Deep Convolutional Neural Networks Enable Discrimination of Heterogeneous Digital Pathology Images

Pathological evaluation of tumor tissue is pivotal for diagnosis in cancer patients and automated image analysis approaches have great potential to increase precision of diagnosis and help reduce human error. In this study, we utilize several computational methods based on convolutional neural networks (CNN) and build a stand-alone pipeline to effectively classify different histopathology images across different types of cancer. In particular, we demonstrate the utility of our pipeline to discriminate between two subtypes of lung cancer, four biomarkers of bladder cancer, and five biomarkers of breast cancer. In addition, we apply our pipeline to discriminate among four immunohistochemistry (IHC) staining scores of bladder and breast cancers. Our classification pipeline includes a basic CNN architecture, Google's Inceptions with three training strategies, and an ensemble of two state-of-the-art algorithms, Inception and ResNet. Training strategies include training the last layer of Google's Inceptions, training the network from scratch, and fine-tunning the parameters for our data using two pre-trained version of Google's Inception architectures, Inception-V1 and Inception-V3. We demonstrate the power of deep learning approaches for identifying cancer subtypes, and the robustness of Google's Inceptions even in presence of extensive tumor heterogeneity. On average, our pipeline achieved accuracies of 100%, 92%, 95%, and 69% for discrimination of various cancer tissues, subtypes, biomarkers, and scores, respectively. Our pipeline and related documentation is freely available at https://github.com/ih-_lab/CNN_Smoothie.

Identification of genes and critical control proteins associated with inflammatory breast cancer using network controllability

One of the most aggressive forms of breast cancer is inflammatory breast cancer (IBC), whose lack of tumour mass also makes a prompt diagnosis difficult. Moreover, genomic differences between common breast cancers and IBC have not been completely assessed, thus substantially limiting the identification of biomarkers unique to IBC. Here, we developed a novel statistical analysis of gene expression profiles corresponding to microdissected IBC, non-IBC (nIBC) and normal samples that enabled us to identify a set of genes significantly associated with a specific disease state. Second, by using advanced methods based on controllability network theory, we identified a set of critical control proteins that uniquely and structurally control the entire proteome. By mapping high change variance genes in protein interaction networks, we found that a large statistically significant fraction of genes whose variance changed significantly between normal and IBC and nIBC disease states were among the set of critical control proteins. Moreover, this analysis identified the overlapping genes with the highest statistical significance; these genes may assist in developing future biomarkers and determining drug targets to disrupt the molecular pathways driving carcinogenesis in IBC.

Celiac disease and hepatitis C relationships in transcriptional regulatory networks

AIM

we mainly aimed to elucidate potential comorbidities between celiac disease and hepatitis c by means of data and network analysis approaches.

BACKGROUND

understanding the association among the disorders evidently has important impact on the diagnosis and therapeutic approaches. Celiac disease is the most challenging, common types of autoimmune disorders. On the other hand, hepatitis c virus genome products like some proteins are supposed to be resemble to gliadin types that in turn activates gluten intolerance in people with inclined to gluten susceptibilities. Moreover, a firm support of association between chronic hepatitis and celiac disease remains largely unclear. Henceforth exploring cross-talk among these diseases will apparently lead to the promising discoveries concerning important genes and regulators.

METHODS

321 and 1032 genes associated with celiac disease and hepatitis c retrieved from DisGeNET were subjected to build a gene regulatory network. Afterward a network-driven integrative analysis was performed to exploring prognosticates genes and related pathways.

RESULTS

105 common genes between these diseases included 11 transcription factors were identified as hallmark molecules where by further screening enriched in biological GO terms and pathways chiefly in immune systems and signaling pathways such as chemokines, cytokines and interleukins.

CONCLUSION

in silico data analysis approaches indicated that the identified selected combinations of genes covered a wide range of known functions triggering the inflammation implicated in these diseases.

Exploring conserved mRNA-miRNA interactions in colon and lung cancers

AIM

The main goal of this analysis was prioritization of co-expressed genes and miRNAs that are thought to have important influences in the pathogenesis of colon and lung cancers.

BACKGROUND

MicroRNAs (miRNAs) as small and endogenous noncoding RNAs which regulate gene expression by repressing mRNA translation or decreasing stability of mRNAs; they have proven pivotal roles in different types of cancers. Accumulating evidence indicates the role of miRNAs in a wide range of biological processes from oncogenesis and tumor suppressors to contribution to tumor progression. Colon and lung cancers are frequently encountered challenging types of cancers; therefore, exploring trade-off among underlying biological units such as miRNA with mRNAs will probably lead to identification of promising biomarkers involved in these malignancies.

METHODS

Colon cancer and lung cancer expression data were downloaded from Firehose and TCGA databases and varied genes extracted by DCGL software were subjected to build two gene regulatory networks by parmigene R package. Afterwards, a network-driven integrative analysis was performed to explore prognosticates genes, miRNAs and underlying pathways.

RESULTS

A total of 192 differentially expressed miRNAs and their target genes within gene regulatory networks were derived by ARACNE algorithm. BTF3, TP53, MYC, CALR, NEM2, miR-29b-3p and miR-145 were identified as bottleneck nodes and enriched via biological gene ontology (GO) terms and pathways chiefly in biosynthesis and signaling pathways by further screening.

CONCLUSION

Our study uncovered correlated alterations in gene expression that may relate with colon and lung cancers and highlighted the potent common biomarker candidates for the two diseases.

Predicting link directionality in gene regulation from gene expression profiles using volatility-constrained correlation

To uncover potential disease molecular pathways and signaling networks, we do not only need undirected maps but also we need to infer the directionality of functional or physical interactions between cellular components. A wide range of methods for identifying functional interactions between genes relies on correlations between experimental gene expression measurements to some extent. However, the standard Pearson or Spearman correlation-based approaches can only determine undirected correlations between cellular components. Here, we apply a volatility-constrained correlation method for gene expression profiles that offers a new metric to capture directionality of interactions between genes. To evaluate the predictions we used four datasets distributed by the DREAM5 network inference challenge including an in silico-constructed network and three organisms such as S. aureus, E. coli and S. cerevisiae. The predictions performed by our proposed method were compared to a gold standard of experimentally verified directionality of genetic regulatory links. Our findings show that our method successfully predicts the genetic interaction directionality with a success rate higher than 0.5 with high statistical significance.

Erratum to: Inferring interaction type in gene regulatory networks using co-expression data

[This corrects the article DOI: 10.1186/s13015-015-0054-4.].