Differential Network Analysis Research Articles

Differential co-expression network analysis with DCoNA reveals isomiR targeting aberrations in prostate cancer.

One of the standard methods of high-throughput RNA sequencing analysis is differential expression. However, it does not detect changes in molecular regulation. In contrast to the standard differential expression analysis, differential co-expression one aims to detect pairs or clusters whose mutual expression changes between two conditions. We developed Differential Co-expression Network Analysis (DCoNA)-an open-source statistical tool that allows one to identify pair interactions, which correlation significantly changes between two conditions. Comparing DCoNA with the state-of-the-art analog, we showed that DCoNA is a faster, more accurate and less memory-consuming tool. We applied DCoNA to prostate mRNA/miRNA-seq data collected from The Cancer Genome Atlas (TCGA) and compared predicted regulatory interactions of miRNA isoforms (isomiRs) and their target mRNAs between normal and cancer samples. As a result, almost all highly expressed isomiRs lost negative correlation with their targets in prostate cancer samples compared to ones without the pathology. One exception to this trend was the canonical isomiR of hsa-miR-93-5p acquiring cancer-specific targets. Further analysis showed that cancer aggressiveness simultaneously increased with the expression level of this isomiR in both TCGA primary tumor samples and 153 blood plasma samples of P. Hertsen Moscow Oncology Research Institute patients' cohort analyzed by miRNA microarrays. Source code and documentation of DCoNA are available at https://github.com/zhiyanov/DCoNA. Supplementary data are available at Bioinformatics online.

Lineage-specific regulatory changes in hypertrophic cardiomyopathy unraveled by single-nucleus RNA-seq and spatial transcriptomics

Hypertrophic cardiomyopathy (HCM) is the most common cardiac genetic disorder characterized by cardiomyocyte hypertrophy and cardiac fibrosis. Pathological cardiac remodeling in the myocardium of HCM patients may progress to heart failure. An in-depth elucidation of the lineage-specific changes in pathological cardiac remodeling of HCM is pivotal for the development of therapies to mitigate the progression. Here, we performed single-nucleus RNA-seq of the cardiac tissues from HCM patients or healthy donors and conducted spatial transcriptomic assays on tissue sections from patients. Unbiased clustering of 55,122 nuclei from HCM and healthy conditions revealed 9 cell lineages and 28 clusters. Lineage-specific changes in gene expression, subpopulation composition, and intercellular communication in HCM were discovered through comparative analyses. According to the results of pseudotime ordering, differential expression analysis, and differential regulatory network analysis, potential key genes during the transition towards a failing state of cardiomyocytes such as FGF12, IL31RA, and CREB5 were identified. Transcriptomic dynamics underlying cardiac fibroblast activation were also uncovered, and potential key genes involved in cardiac fibrosis were obtained such as AEBP1, RUNX1, MEOX1, LEF1, and NRXN3. Using the spatial transcriptomic data, spatial activity patterns of the candidate genes, pathways, and subpopulations were confirmed on patient tissue sections. Moreover, we showed experimental evidence that in vitro knockdown of AEBP1 could promote the activation of human cardiac fibroblasts, and overexpression of AEBP1 could attenuate the TGFβ-induced activation. Our study provided a comprehensive analysis of the lineage-specific regulatory changes in HCM, which laid the foundation for targeted drug development in HCM.

A pseudo-value regression approach for differential network analysis of co-expression data

BackgroundThe differential network (DN) analysis identifies changes in measures of association among genes under two or more experimental conditions. In this article, we introduce a pseudo-value regression approach for network analysis (PRANA). This is a novel method of differential network analysis that also adjusts for additional clinical covariates. We start from mutual information criteria, followed by pseudo-value calculations, which are then entered into a robust regression model.ResultsThis article assesses the model performances of PRANA in a multivariable setting, followed by a comparison to dnapath and DINGO in both univariable and multivariable settings through variety of simulations. Performance in terms of precision, recall, and F1 score of differentially connected (DC) genes is assessed. By and large, PRANA outperformed dnapath and DINGO, neither of which is equipped to adjust for available covariates such as patient-age. Lastly, we employ PRANA in a real data application from the Gene Expression Omnibus database to identify DC genes that are associated with chronic obstructive pulmonary disease to demonstrate its utility.ConclusionTo the best of our knowledge, this is the first attempt of utilizing a regression modeling for DN analysis by collective gene expression levels between two or more groups with the inclusion of additional clinical covariates. By and large, adjusting for available covariates improves accuracy of a DN analysis.

Over-optimism in unsupervised microbiome analysis: Insights from network learning and clustering.

In recent years, unsupervised analysis of microbiome data, such as microbial network analysis and clustering, has increased in popularity. Many new statistical and computational methods have been proposed for these tasks. This multiplicity of analysis strategies poses a challenge for researchers, who are often unsure which method(s) to use and might be tempted to try different methods on their dataset to look for the "best" ones. However, if only the best results are selectively reported, this may cause over-optimism: the "best" method is overly fitted to the specific dataset, and the results might be non-replicable on validation data. Such effects will ultimately hinder research progress. Yet so far, these topics have been given little attention in the context of unsupervised microbiome analysis. In our illustrative study, we aim to quantify over-optimism effects in this context. We model the approach of a hypothetical microbiome researcher who undertakes four unsupervised research tasks: clustering of bacterial genera, hub detection in microbial networks, differential microbial network analysis, and clustering of samples. While these tasks are unsupervised, the researcher might still have certain expectations as to what constitutes interesting results. We translate these expectations into concrete evaluation criteria that the hypothetical researcher might want to optimize. We then randomly split an exemplary dataset from the American Gut Project into discovery and validation sets multiple times. For each research task, multiple method combinations (e.g., methods for data normalization, network generation, and/or clustering) are tried on the discovery data, and the combination that yields the best result according to the evaluation criterion is chosen. While the hypothetical researcher might only report this result, we also apply the "best" method combination to the validation dataset. The results are then compared between discovery and validation data. In all four research tasks, there are notable over-optimism effects; the results on the validation data set are worse compared to the discovery data, averaged over multiple random splits into discovery/validation data. Our study thus highlights the importance of validation and replication in microbiome analysis to obtain reliable results and demonstrates that the issue of over-optimism goes beyond the context of statistical testing and fishing for significance.

Network-based identification and prioritization of key transcriptional factors of diabetic kidney disease

Diabetic nephropathy (DN) is one of the most established microvascular complications of diabetes and a key cause of end-stage renal disease. It is well established that gene susceptibility to DN plays a critical role in disease pathophysiology. Therefore, many genetic studies have been performed to categorize candidate genes in prominent diabetic cohorts, aiming to investigate DN pathogenesis and etiology. In this study, we performed a meta-analysis on the expression profiles of GSE1009, GSE30122, GSE96804, GSE99340, GSE104948, GSE104954, and GSE111154 to identify critical transcriptional factors associated with DN progression. The analysis was conducted for all individual datasets for each kidney tissue (glomerulus, tubules, and kidney cortex). We identified distinct clusters of susceptibility genes that were dysregulated in a renal compartment-specific pattern. Further, we recognized a small but a closely connected set of these susceptibility genes enriched for podocyte differentiation, several of which were characterized as genes encoding critical transcriptional factors (TFs) involved in DN development and podocyte function. To validate the role of identified TFs in DN progression, we functionally validated the three main TFs (DACH1, LMX1B, and WT1) identified through differential gene expression and network analysis using the hyperglycemic zebrafish model. We report that hyperglycemia-induced altered gene expression of the key TF genes leads to morphological abnormalities in zebrafish glomeruli, pronephric tubules, proximal and distal ducts. This study demonstrated that altered expression of these TF genes could be associated with hyperglycemia-induced nephropathy and, thus, aids in understanding the molecular drivers, essential genes, and pathways that trigger DN initiation and development.

Machine learning and network analysis of the gut microbiome from patients with schizophrenia and non-psychiatric subject controls reveal behavioral risk factors and bacterial interactions

Recent findings have supported an association between deviations in gut microbiome composition and schizophrenia. However, the extent to which the gut microbiota contributes to schizophrenia remains unclear. Moreover, studies have yet to explore variations in ecological associations among bacterial types in subjects with schizophrenia, which can reveal differences in community interactions and gut stability. We examined the dataset collected by Nguyen et al. (2021) to investigate the similarities and differences in gut microbial constituents between 48 subjects with schizophrenia and 48 matched non-psychiatric comparison cases. We re-analyzed alpha- and beta-diversity differences and completed modified differential abundance analyses and confirmed the findings of Nguyen et al. (2021) that there was little variation in alpha-diversity but significant differences in beta-diversity between individuals with schizophrenia and non-psychiatric subjects. We also conducted mediation analysis, developed a machine learning (ML) model to predict schizophrenia, and completed network analysis to examine community-level interactions among bacterial taxa. Our study offers new insights, suggesting that the gut microbiome mediates the effects between schizophrenia and smoking status, BMI, anxiety score, and depression score. Our differential abundance and network analysis findings suggest that the differential abundance of Lachnospiraceae and Ruminococcaceae taxa fosters a decrease in stabilizing competitive interactions in the gut microbiome of subjects with schizophrenia. Loss of this competition may promote ecological instability and dysbiosis, altering gut-brain axis interactions in these subjects.

Identification of PFKFB2 as a key gene for the transition from acute to old myocardial infarction in peripheral blood.

This study aims to analyze the gene expression profile of peripheral blood in different stages of myocardial infarction (MI) by transcriptome sequencing, and to study the gene expression characteristics of peripheral blood after MI. Differentially expressed genes (DEGs) and weighted gene co-expression network analysis (WGCNA) were used to identify genes and modules associated with old myocardial infarction (OMI). Gene Ontology (GO) functional annotation and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway annotation were applied to analyze the potential functions of genes. Hub genes were identified by Random Forest Classifier. CIBERSORT was used to provide an estimate of the abundance of 22 immune cells in peripheral blood. Quantitative polymerase chain reaction (qPCR) was used to detect gene expression levels in clinical samples. The cellular components (CC) of peripheral blood were counted by an automatic hematology analyzer. Through differential gene analysis and co-expression network analysis, 11 candidate genes were obtained. A random forest classifier identified 10 hub genes. Immune cell distribution of peripheral blood was found that T cell CD4 memory resting, NK cells resting, Dendritic cells activated, Mast cells resting, Monocytes and Neutrophils were correlated with OMI. Spearman correlation analysis found that PFKFB2 is related to the above immune cells. Low expression of PFKFB2 in peripheral blood of OMI was detected in clinical samples, and the relationship between PFKFB2 and peripheral blood immune cell counts was analyzed, which showed monocytes were associated with PFKFB2 in our study. PFKFB2 was low expressed in OMI, and related to the distribution of immune cells. PFKFB2 may play a key role in reflecting the transition from AMI to OMI, and predicting the distribution of immune cells, which provided a new perspective for improving myocardial fibrosis and adverse remodeling.

Altered Serum Bile Acid Profile Associated with Chronic Allograft Dysfunction in Kidney Transplant Recipients.

BACKGROUND Chronic allograft dysfunction (CAD) is the leading cause of graft loss among kidney transplant recipients (KTRs). Bile acids (BAs) play an important role in regulating inflammatory process, which is the major contributor to the development of CAD. The aim of this study was to evaluate the association between BAs metabolic dysregulation and CAD in KTRs. MATERIAL AND METHODS Fifteen serum BA species were determined in 43 healthy controls (HCs) and 131 KTRs by UPLC-MS/MS. KTRs were grouped into stable renal function (STA) and CAD1 and CAD2 groups based on eGFR levels. Circulating CYP7A1, CYP7B1, CYP27A1, and SLCO2B1 mRNA levels were determined by RT-PCR. RESULTS Total BA concentrations were comparable among the 4 groups. However, KTRs showed significantly different BAs profiling compared to HCs. KTRs with severe CAD (CAD2) had significantly lower unconjugated BAs and secondary BAs (SBAs) compared to the other 3 groups. KTRs had significantly lower SBAs/primary BAs (PBAs) ratios than HCs, which were comparable among the 3 KTR groups. Conjugated/unconjugated BAs ratios increased significantly with the deterioration of allograft function, which was further confirmed by correlation analysis. Differential correlation network analysis revealed that perturbations in intraclass and interclass BA coregulation existed during CAD progression. Moreover, relative gene expressions of CYP7B1 and CYP27A1 were positively correlated with eGFR. CONCLUSIONS BA species profiling, but not total BA concentrations, was significantly altered in KTRs with CAD. The shifts from unconjugated BAs toward conjugated BAs, SBAs toward PBAs, and distinct pairwise BAs coregulation patterns were the main characteristics of KTRs with CAD.

RF05 | PSUN341 Role of sirtuins in breast cancer cells response to Fulvestrant

Abstract Endocrine therapies reduce glucose and glutamine uptake and total cellular ATP production in breast cancer (BC) cells, resulting in changes in cellular nutrient balance that could lead to cell cycle arrest and to cell death. Understanding the ability of cells to bypass these metabolic effects is fundamental to understand acquisition of endocrine resistance and cross-resistance to other drugs by BC cells. Sirtuins (SIRTs) are NAD+-dependent deacylases involved in a variety of cellular processes, including metabolism and stress responses. SIRT1 deacetylates histones and transcription factors, and regulates glucose metabolism through TORC2, PGC1α, FOXO1. SIRT3 modulates mitochondria adaptation to conditions of low energy input by promoting the production of energy from sources different from glucose. SIRT3 mediates metabolic reprogramming by also destabilizing HIF-1α, regulating glycoytic gene expression. SIRT 3 regulation of the mitochondrial unfolded protein response and mitophagy machinery has been described. SIRT3expression is higher in early (≤3 years) vs. late (≥5 years) recurrences (measure of resistance vs. dormancy) in 4 clinical datasets. Using Differential Dependency Network analysis to compare the wiring of the SIRTs and key metabolism-related genes in matched antiestrogen- sensitive vs. resistant BC cells, guided by gene related to the nicotinate and nicotinamide metabolism, and associated with clinical outcome in ER+ patients treated with an endocrine therapy, we identified several novel signaling hubs including: CDKN2A (p16) Hub, linking SIRT3 to cell proliferation; TP53 Hub, linking SIRT5 to clinical resistance to aromatase inhibitors, cell survival and proliferation; ATM Hub that comprises ATM/SIRT2/SIRT4/HIF1A and may functionally link endocrine resistance to chemoresistance, consistent with the relatively poor responses of ER+ BC to cytotoxic drugs. Ability of SIRTs to sense and respond to changes in energy could provide a mechanism for an acquired drug-resistant phenotype enabled by epigenetically maintained changes in genes that control metabolism.Using endocrine sensitive (LCC1) and resistant (LCC9) BC cells, we confirmed the higher expression of SIRT1 and 3 in LCC9, compared to LCC1 cells. Upregulation of SIRT3 and SIRT5 expression within 24 hrs of Fulvestrant (ICI; 1 µM) treatment (p=0.0176) was observed in LCC1, but not in LCC9 cells. Conversely, treatment with 17β-estradiol (10 nM) did not significantly affect SIRT3 or SIRT5 expression within 72h in either LCC1 or LCC9 cells. Silencing of SIRT3 induced cell death, reducing growth of LCC9 after 72h of ICI treatment, compared to the untreated control, but no G1/G2 phase arrest in cell cycle was observed during same treatment. In ER+ BC patients, lower expression of SIRT3 is associated with poor relapse free survival (RFS) (HR=0.57; CI=0.43-0.76); p= 7.7x 10-5) whereas higher expression of SIRT5 is associated with poor RFS (HR=2.5; CI=0.99-6.59; p=4.4×10-2). These data indicate an involvement of SIRTs in response to Fulvestrant by BC cells and will be more investigated. Presentation: Saturday, June 11, 2022 1:12 p.m. - 1:17 p.m., Sunday, June 12, 2022 12:30 p.m. - 2:30 p.m.

Differential gene co-expression network analyses reveal novel molecules associated with transcriptional dysregulation of key biological processes in osteoarthritis knee cartilage

ObjectivesTo compare co-expression networks of normal and osteoarthritis knee cartilage to uncover molecules associated with the transcriptional misregulation compromising biological processes (BPs) critical for cartilage homeostasis. DesignNormal and osteoarthritis human knee cartilage RNA-seq GSE114007 dataset was obtained from the Gene Expression Omnibus database. Partial Correlation and Information Theory (PCIT) algorithm was used to build co-expression networks containing all nodes connecting to at least one differentially expressed gene (DEG) in normal and osteoarthritis networks. Hub and hub centrality genes were used to perform functional enrichment analysis. Enriched BPs known to be associated with both healthy and diseased cartilage were compared in depth. ResultsDifferential co-expression network analyses allowed the identification of DDX43 and USP42 as exclusively co-expressed with DEGs in normal and osteoarthritis networks, respectively. The top hub and hub centrality genes of these networks were HIST1H3A and SNHG12 (normal) and TAF9B and OTUD1 (osteoarthritis). Enrichment analysis revealed several shared BPs between the contrasting groups, which are well-known in osteoarthritis pathogenesis. Protein-protein interaction network analysis for these BPs showed a global down-regulation of transcription factors in osteoarthritis. Specific transcription factors were identified as pleiotropic mediators in articular cartilage maintenance since they take part in several BPs. In addition, chromatin organisation and modification proteins were found relevant for osteoarthritis development. ConclusionDifferential gene co-expression analysis allowed the identification of novel and high priority therapeutic candidate genes that may drive modifications in the transcriptional “status” of cartilage in osteoarthritis.

Gene co-expression network analysis revealed novel biomarkers for ovarian cancer.

Ovarian cancer is the second most common gynecologic cancer and remains the leading cause of death of all gynecologic oncologic disease. Therefore, understanding the molecular mechanisms underlying the disease, and the identification of effective and predictive biomarkers are invaluable for the development of diagnostic and treatment strategies. In the present study, a differential co-expression network analysis was performed via meta-analysis of three transcriptome datasets of serous ovarian adenocarcinoma to identify novel candidate biomarker signatures, i.e. genes and miRNAs. We identified 439 common differentially expressed genes (DEGs), and reconstructed differential co-expression networks using common DEGs and considering two conditions, i.e. healthy ovarian surface epithelia samples and serous ovarian adenocarcinoma epithelia samples. The modular analyses of the constructed networks indicated a co-expressed gene module consisting of 17 genes. A total of 11 biomarker candidates were determined through receiver operating characteristic (ROC) curves of gene expression of module genes, and miRNAs targeting these genes were identified. As a result, six genes (CDT1, CNIH4, CRLS1, LIMCH1, POC1A, and SNX13), and two miRNAs (mir-147a, and mir-103a-3p) were suggested as novel candidate prognostic biomarkers for ovarian cancer. Further experimental and clinical validation of the proposed biomarkers could help future development of potential diagnostic and therapeutic innovations in ovarian cancer.

Differential network analysis of oral microbiome metatranscriptomes identifies community scale metabolic restructuring in dental caries.

Dental caries is a microbial disease and the most common chronic health condition, affecting nearly 3.5 billion people worldwide. In this study, we used a multiomics approach to characterize the supragingival plaque microbiome of 91 Australian children, generating 658 bacterial and 189 viral metagenome-assembled genomes with transcriptional profiling and gene-expression network analysis. We developed a reproducible pipeline for clustering sample-specific genomes to integrate metagenomics and metatranscriptomics analyses regardless of biosample overlap. We introduce novel feature engineering and compositionally-aware ensemble network frameworks while demonstrating their utility for investigating regime shifts associated with caries dysbiosis. These methods can be applied when differential abundance modeling does not capture statistical enrichments or the results from such analysis are not adequate for providing deeper insight into disease. We identified which organisms and metabolic pathways were central in a coexpression network as well as how these networks were rewired between caries and caries-free phenotypes. Our findings provide evidence of a core bacterial microbiome that was transcriptionally active in the supragingival plaque of all participants regardless of phenotype, but also show highly diagnostic changes in the ways that organisms interact. Specifically, many organisms exhibit high connectedness with central carbon metabolism to Cardiobacterium and this shift serves a bridge between phenotypes. Our evidence supports the hypothesis that caries is a multifactorial ecological disease.

Applying differential network analysis to longitudinal gene expression in response to perturbations.

Differential Network (DN) analysis is a method that has long been used to interpret changes in gene expression data and provide biological insights. The method identifies the rewiring of gene networks in response to external perturbations. Our study applies the DN method to the analysis of RNA-sequencing (RNA-seq) time series datasets. We focus on expression changes: (i) in saliva of a human subject after pneumococcal vaccination (PPSV23) and (ii) in primary B cells treated ex vivo with a monoclonal antibody drug (Rituximab). The DN method enabled us to identify the activation of biological pathways consistent with the mechanisms of action of the PPSV23 vaccine and target pathways of Rituximab. The community detection algorithm on the DN revealed clusters of genes characterized by collective temporal behavior. All saliva and some B cell DN communities showed characteristic time signatures, outlining a chronological order in pathway activation in response to the perturbation. Moreover, we identified early and delayed responses within network modules in the saliva dataset and three temporal patterns in the B cell data.

Using machine learning to identify gene interaction networks associated with breast cancer

BackgroundBreast cancer (BC) is one of the most prevalent cancers worldwide but its etiology remains unclear. Obesity is recognized as a risk factor for BC, and many obesity-related genes may be involved in its occurrence and development. Research assessing the complex genetic mechanisms of BC should not only consider the effect of a single gene on the disease, but also focus on the interaction between genes. This study sought to construct a gene interaction network to identify potential pathogenic BC genes.MethodsThe study included 953 BC patients and 963 control individuals. Chi-square analysis was used to assess the correlation between demographic characteristics and BC. The joint density-based non-parametric differential interaction network analysis and classification (JDINAC) was used to build a BC gene interaction network using single nucleotide polymorphisms (SNP). The odds ratio (OR) and 95% confidence interval (95% CI) of hub gene SNPs were evaluated using a logistic regression model. To assess reliability, the hub genes were quantified by edgeR program using BC RNA-seq data from The Cancer Genome Atlas (TCGA) and identical edges were verified by logistic regression using UK Biobank datasets. Go and KEGG enrichment analysis were used to explore the biological functions of interactive genes.ResultsBody mass index (BMI) and menopause are important risk factors for BC. After adjusting for potential confounding factors, the BC gene interaction network was identified using JDINAC. LEP, LEPR, XRCC6, and RETN were identified as hub genes and both hub genes and edges were verified. LEPR genetic polymorphisms (rs1137101 and rs4655555) were also significantly associated with BC. Enrichment analysis showed that the identified genes were mainly involved in energy regulation and fat-related signaling pathways.ConclusionWe explored the interaction network of genes derived from SNP data in BC progression. Gene interaction networks provide new insight into the underlying mechanisms of BC.

DiffBrainNet: Differential analyses add new insights into the response to glucocorticoids at the level of genes, networks and brain regions

Genome-wide gene expression analyses are invaluable tools for studying biological and disease processes, allowing a hypothesis-free comparison of expression profiles. Traditionally, transcriptomic analysis has focused on gene-level effects found by differential expression. In recent years, network analysis has emerged as an important additional level of investigation, providing information on molecular connectivity, especially for diseases associated with a large number of linked effects of smaller magnitude, like neuropsychiatric disorders. Here, we describe how combined differential expression and prior-knowledge-based differential network analysis can be used to explore complex datasets. As an example, we analyze the transcriptional responses following administration of the glucocorticoid/stress receptor agonist dexamethasone in 8 mouse brain regions important for stress processing. By applying a combination of differential network- and expression-analyses, we find that these explain distinct but complementary biological mechanisms of the glucocorticoid responses. Additionally, network analysis identifies new differentially connected partners of risk genes and can be used to generate hypotheses on molecular pathways affected. With DiffBrainNet (http://diffbrainnet.psych.mpg.de), we provide an analysis framework and a publicly available resource for the study of the transcriptional landscape of the mouse brain which can identify molecular pathways important for basic functioning and response to glucocorticoids in a brain-region specific manner.

The spatial transcriptomic landscape of non-small cell lung cancer brain metastasis

Brain metastases (BrMs) are a common occurrence in lung cancer with a dismal outcome. To understand the mechanism of metastasis to inform prognosis and treatment, here we analyze primary and metastasized tumor specimens from 44 non-small cell lung cancer patients by spatial RNA sequencing, affording a whole transcriptome map of metastasis resolved with morphological markers for the tumor core, tumor immune microenvironment (TIME), and tumor brain microenvironment (TBME). Our data indicate that the tumor microenvironment (TME) in the brain, including the TIME and TBME, undergoes extensive remodeling to create an immunosuppressive and fibrogenic niche for the BrMs. Specifically, the brain TME is characterized with reduced antigen presentation and B/T cell function, increased neutrophils and M2-type macrophages, immature microglia, and reactive astrocytes. Differential gene expression and network analysis identify fibrosis and immune regulation as the major functional modules disrupted in both the lung and brain TME. Besides providing systems-level insights into the mechanism of lung cancer brain metastasis, our study uncovers potential prognostic biomarkers and suggests that therapeutic strategies should be tailored to the immune and fibrosis status of the BrMs.

NetTDP: permutation-based true discovery proportions for differential co-expression network analysis.

Existing methods for differential network analysis could only infer whether two networks of interest have differences between two groups of samples, but could not quantify and localize network differences. In this work, a novel method, permutation-based Network True Discovery Proportions (NetTDP), is proposed to quantify the number of edges (correlations) or nodes (genes) for which the co-expression networks are different. In the NetTDP method, we propose an edge-level statistic and a node-level statistic, and detect true discoveries of edges and nodes in the sense of differential co-expression network, respectively, by the permutation-based sumSome method. Furthermore, the NetTDP method could further localize the differences by inferring the TDPs for edge or gene subsets of interest, which can be selected post hoc. Our NetTDP method allows inference on data-driven modules or biology-driven gene sets, and remains valid even when these sub-networks are optimized using the same data. Experimental results on both simulation data sets and five real data sets show the effectiveness of the proposed method in inferring the quantification and localization of differential co-expression networks. The R code is available at https://github.com/LiminLi-xjtu/NetTDP.

GeneNetTools: tests for Gaussian graphical models with shrinkage.

Gaussian graphical models (GGMs) are network representations of random variables (as nodes) and their partial correlations (as edges). GGMs overcome the challenges of high-dimensional data analysis by using shrinkage methodologies. Therefore, they have become useful to reconstruct gene regulatory networks from gene-expression profiles. However, it is often ignored that the partial correlations are 'shrunk' and that they cannot be compared/assessed directly. Therefore, accurate (differential) network analyses need to account for the number of variables, the sample size, and also the shrinkage value, otherwise, the analysis and its biological interpretation would turn biased. To date, there are no appropriate methods to account for these factors and address these issues. We derive the statistical properties of the partial correlation obtained with the Ledoit-Wolf shrinkage. Our result provides a toolbox for (differential) network analyses as (i) confidence intervals, (ii) a test for zero partial correlation (null-effects) and (iii) a test to compare partial correlations. Our novel (parametric) methods account for the number of variables, the sample size and the shrinkage values. Additionally, they are computationally fast, simple to implement and require only basic statistical knowledge. Our simulations show that the novel tests perform better than DiffNetFDR-a recently published alternative-in terms of the trade-off between true and false positives. The methods are demonstrated on synthetic data and two gene-expression datasets from Escherichia coli and Mus musculus. The R package with the methods and the R script with the analysis are available in https://github.com/V-Bernal/GeneNetTools. Supplementary data are available at Bioinformatics online.

Integration of differential expression and network structure for ‘omics data analysis

Differential expression (DE) analysis has been routinely used to identify molecular features that are statistically significantly different between distinct biological groups. In recent years, differential network (DN) analysis has emerged as a powerful approach to uncover molecular network structure changes from one biological condition to the other where the molecular features with larger topological changes are selected as biomarkers. Although a large number of DE and a few DN-based methods are available, they have been usually implemented independently. DE analysis ignores the relationship among molecular features while DN analysis does not account for the expression changes at individual level. Therefore, an integrative analysis approach that accounts for both DE and DN is required to identify disease associated key features. Although, a handful of methods have been proposed, there is no method that optimizes the combination of DE and DN. We propose a novel integrative analysis method, DNrank, to identify disease-associated molecular features that leverages the strengths of both DE and DN by calculating a weight using resampling based cross validation scheme within the algorithm. First, differential expression analysis of individual molecular features is carried out. Second, a differential network structure is constructed using the differential partial correlation analysis. Third, the molecular features are ranked in the order of their significances by integrating their DE measures and DN structure using the modified Google's PageRank algorithm. In the algorithm, the optimum combination of DE and DN analyses is achieved by evaluating the prediction performance of top-ranked features utilizing support vector machine classifier with Monte Carlo cross validation. The proposed method is illustrated using both simulated data and three real data sets. The results show that the proposed method has a better performance in identifying important molecular features with respect to predictive discrimination. Also, as compared to existing feature selection methods, the top-ranked features selected by our method had a higher stability in selection. DNrank allows the researchers to identify the disease-associated features by utilizing both expression and network topology changes between two groups.

Prediction of intestinal stem cell regulatory genes from Drosophila gut damage model created using multiple inducers: Differential gene expression-based protein-protein interaction network analysis

Intestinal tissue functions in innate immunity to prevent the entry of harmful substances, and to maintain homeostasis through the constant proliferation of intestinal stem cells (ISC). To understand the mechanisms which regulate ISC in response to gut damage, we identified 81 differentially expressed genes (DEGs) through RNA-seq analysis after oral administration of three intestinal-damaging substances to Drosophila melanogaster. Through protein-protein interaction (PPI) and functional annotation studies, the top 22 DEGs ordered by the number of nodes in the PPI network were analyzed in relation to cell development. Through network topology analysis, we identified 12 essential seed genes. From this we confirmed that p53, RpL17, Fmr1, Stat92E, CG31343, Cnot4, CG9281, CG8184, Evi5, and to were essential for ISC proliferation during gut damage using knockdown RNAi Drosophila. This study presents a method for identifying candidate genes relating to intestinal damage that has scope for furthering our understanding of gut disease.