A measure of the signal-to-noise ratio of microarray samples and studies using gene correlations

A pipeline for testing drug mechanism of action and combination therapies: From microarray data to simulations via Linear-In-Flux-Expressions: Testing four-drug combinations for tuberculosis treatment

Computational methods are becoming commonly used in many areas of medical research. Recently, the modeling of biological mechanisms associated with disease pathophysiology have benefited from approaches such as Quantitative Systems Pharmacology (briefly QSP) and Physiologically Based Pharmacokinetics (briefly PBPK). These methodologies show the potential to enhance, if not substitute animal models. The main reasons for this success are the high accuracy and low cost. Solid mathematical foundations of such methods, such as compartmental systems and flux balance analysis, provide a good base on which to build computational tools. However, there are many choices to be made in model design, that will have a large impact on how these methods perform as we scale up the network or perturb the system to uncover the mechanisms of action of new compounds or therapy combinations. A computational pipeline is presented here that starts with available -omic data and utilizes advanced mathematical simulations to inform the modeling of a biochemical system. Specific attention is devoted to creating a modular workflow, including the mathematical rigorous tools to represent complex chemical reactions, and modeling drug action in terms of its impact on multiple pathways. An application to optimizing combination therapy for tuberculosis shows the potential of the approach.

LiPLike: towards gene regulatory network predictions of high certainty

MOTIVATION

High correlation in expression between regulatory elements is a persistent obstacle for the reverse-engineering of gene regulatory networks. If two potential regulators have matching expression patterns, it becomes challenging to differentiate between them, thus increasing the risk of false positive identifications.

RESULTS

To allow for gene regulation predictions of high confidence, we propose a novel method, the Linear Profile Likelihood (LiPLike), that assumes a regression model and iteratively searches for interactions that cannot be replaced by a linear combination of other predictors. To compare the performance of LiPLike with other available inference methods, we benchmarked LiPLike using three independent datasets from the Dialogue on Reverse Engineering Assessment and Methods 5 (DREAM5) network inference challenge. We found that LiPLike could be used to stratify predictions of other inference tools, and when applied to the predictions of DREAM5 participants, we observed an average improvement in accuracy of >140% compared to individual methods. Furthermore, LiPLike was able to independently predict networks better than all DREAM5 participants when applied to biological data. When predicting the Escherichia coli network, LiPLike had an accuracy of 0.38 for the top-ranked 100 interactions, whereas the corresponding DREAM5 consensus model yielded an accuracy of 0.11.

AVAILABILITY AND IMPLEMENTATION

We made LiPLike available to the community as a Python toolbox, available at https://gitlab.com/Gustafsson-lab/liplike. We believe that LiPLike will be used for high confidence predictions in studies where individual model interactions are of high importance, and to remove false positive predictions made by other state-of-the-art gene-gene regulation prediction tools.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

CRNET: an efficient sampling approach to infer functional regulatory networks by integrating large-scale ChIP-seq and time-course RNA-seq data

Motivation

NGS techniques have been widely applied in genetic and epigenetic studies. Multiple ChIP-seq and RNA-seq profiles can now be jointly used to infer functional regulatory networks (FRNs). However, existing methods suffer from either oversimplified assumption on transcription factor (TF) regulation or slow convergence of sampling for FRN inference from large-scale ChIP-seq and time-course RNA-seq data.

Results

We developed an efficient Bayesian integration method (CRNET) for FRN inference using a two-stage Gibbs sampler to estimate iteratively hidden TF activities and the posterior probabilities of binding events. A novel statistic measure that jointly considers regulation strength and regression error enables the sampling process of CRNET to converge quickly, thus making CRNET very efficient for large-scale FRN inference. Experiments on synthetic and benchmark data showed a significantly improved performance of CRNET when compared with existing methods. CRNET was applied to breast cancer data to identify FRNs functional at promoter or enhancer regions in breast cancer MCF-7 cells. Transcription factor MYC is predicted as a key functional factor in both promoter and enhancer FRNs. We experimentally validated the regulation effects of MYC on CRNET-predicted target genes using appropriate RNAi approaches in MCF-7 cells.

Availability and implementation

R scripts of CRNET are available at http://www.cbil.ece.vt.edu/software.htm.

Contact

xuan@vt.edu.

Supplementary information

Supplementary data are available at Bioinformatics online.

Identification of core genes and prediction of miRNAs associated with osteoporosis using a bioinformatics approach

Osteoporosis (OP) is an age-related disease, and osteoporotic fracture is one of the major causes of disability and mortality in elderly patients (>70 years old). As the pathogenesis and molecular mechanism of OP remain unclear, the identification of disease biomarkers is important for guiding research and providing therapeutic targets. In the present study, core genes and microRNAs (miRNAs) associated with OP were identified. Differentially expressed genes (DEGs) between human mesenchymal stem cell specimens from normal osseous tissues and OP tissues were detected using the GEO2R tool of the Gene Expression Omnibus database and Morpheus. Network topological parameters were determined using NetworkAnalyzer. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes pathway enrichment analyses were performed using the Database for Annotation, Visualization and Integrated Discovery, and ClueGO. Cytoscape with the Search Tool for the Retrieval of Interacting Genes and Molecular Complex Detection plug-in was used to visualize protein-protein interactions (PPIs). Additionally, miRNA-gene regulatory modules were predicted using CyTargetLinker in order to guide future research. In total, 915 DEGs were identified, including 774 upregulated and 141 downregulated genes. Enriched GO terms and pathways were determined, including 'nervous system development', 'regulation of molecular function', 'glutamatergic synapse pathway' and 'pathways in cancer'. The node degrees of DEGs followed power-law distributions. A PPI network with 541 nodes and 1,431 edges was obtained. Overall, 3 important modules were identified from the PPI network. The following 10 genes were identified as core genes based on high degrees of connectivity: Albumin, PH domain leucine-rich repeat-containing protein phosphatase 2 (PHLPP2), DNA topoisomerase 2-α, kininogen 1 (KNG1), interleukin 2 (IL2), leucine-rich repeats and guanylate kinase domain containing, phosphatidylinositol-4,5-bisphosphate 3-kinase catalytic subunit γ (PIK3CG), leptin, transferrin and RNA polymerase II subunit A (POLR2A). Additionally, 15 miRNA-target interactions were obtained using CyTargetLinker. Overall, 7 miRNAs co-regulated IL2, 3 regulated PHLPP2, 3 regulated KNG1, 1 regulated PIK3CG and 1 modulated POLR2A. These results indicate potential biomarkers in the pathogenesis of OP and therapeutic targets.

Specific detection of stable single nucleobase mismatch using SU-8 coated silicon nanowires platform

Novel microarray platform for single nucleotide polymorphisms (SNPs) detection has been developed using silicon nanowires (SiNWs) as support and two different surface modification methods for attaining the necessary functional groups. Accordingly, we compared the detection specificity and stability over time of the probes printed on SiNWs modified with (3-aminopropyl)triethoxysilane (APTES) and glutaraldehyde (GAD), or coated with a simpler procedure using epoxy-based SU-8 photoresist. Scanning electron microscopy (SEM) and energy dispersive X-ray spectroscopy (EDX) were used for comparative characterization of the unmodified and coated SiNWs. The hybridization efficiency was assessed by comprehensive statistical analysis of the acquired data from confocal fluorescence scanning of the manufactured biochips. The high detection specificity between the hybridized probes containing different mismatch types was demonstrated on SU-8 coating by one way ANOVA test (adjusted p value *** < .0001). The stability over time of the probes tethered on SiNWs coated with SU-8 was evaluated after 1, 4, 8 and 21 days of probe incubation, revealing values for coefficient of variation (CV) between 2.4% and 5.6%. The signal-to-both-standard-deviations ratio measured for SU-8 coated SiNWs platform was similar to the commercial support, while the APTES-GAD coated SiNWs exhibited the highest values.

Identification of genes and pathways in the synovia of women with osteoarthritis by bioinformatics analysis

Osteoarthritis (OA) has a high prevalence in female patients and sex may be a key factor affecting the progression of OA. The aim of the present study was to identify genetic signatures in the synovial membranes of female patients with OA and to elucidate the potential associated molecular mechanisms. The gene expression profiles of the GSE55457 and GSE55584 datasets were obtained from the Gene Expression Omnibus database. Data of two synovial membranes from normal female individuals (GSM1337306 and GSM1337310) and two synovial membranes from female patients affected by OA (GSM1337327 and GSM1337330) were obtained from the dataset GSE55457, and those of three synovial membranes from female patients affected by OA (GSM1339628, GSM1339629 and GSM1339632) were obtained from the dataset GSE55584. Differentially expressed genes (DEGs) were identified by using Morpheus software. Protein-protein interaction (PPI) networks of the DEGs were constructed by using Cytoscape software. Subsequently, Gene Ontology (GO) function and Kyoto Encyclopaedia of Genes and Genomes (KEGG) pathway enrichment analyses of the top module of the PPI network were performed by using ClueGo. A total of 377 DEGs were identified in the synovial membranes of OA patients compared with those of normal individuals, including 164 upregulated and 213 downregulated genes. The top 10 hub genes were ubiquitin (UB)C, ribosomal protein (RP) L23A, mammalian target of rapamycin, heat shock protein 90 α family class A member 1, RPS28, RPL37A, RPS24, RPS4X, RPS18 and UBB. The results of the GO analysis indicated that the DEGs included in the top module of the PPI were mainly enriched in the terms ‘nuclear-transcribed mRNA catabolic process’, ‘nonsense mediated decay’, and ‘cytoplasmic translation and ribosomal small subunit biogenesis’. KEGG pathway analysis indicated that the DEGs included in the top one module were mainly enriched in the ‘ribosome’ pathway. The present study provides a systematic, molecular-level understanding of the degeneration of the synovial membrane in the progression of OA in female patients. The hub genes and molecules associated with the synovial membrane may be used as biomarkers and therapeutic targets for the treatment of OA in female patients with OA.

Improving reliability and absolute quantification of human brain microarray data by filtering and scaling probes using RNA-Seq

BACKGROUND

High-throughput sequencing is gradually replacing microarrays as the preferred method for studying mRNA expression levels, providing nucleotide resolution and accurately measuring absolute expression levels of almost any transcript, known or novel. However, existing microarray data from clinical, pharmaceutical, and academic settings represent valuable and often underappreciated resources, and methods for assessing and improving the quality of these data are lacking.

RESULTS

To quantitatively assess the quality of microarray probes, we directly compare RNA-Seq to Agilent microarrays by processing 231 unique samples from the Allen Human Brain Atlas using RNA-Seq. Both techniques provide highly consistent, highly reproducible gene expression measurements in adult human brain, with RNA-Seq slightly outperforming microarray results overall. We show that RNA-Seq can be used as ground truth to assess the reliability of most microarray probes, remove probes with off-target effects, and scale probe intensities to match the expression levels identified by RNA-Seq. These sequencing scaled microarray intensities (SSMIs) provide more reliable, quantitative estimates of absolute expression levels for many genes when compared with unscaled intensities. Finally, we validate this result in two human cell lines, showing that linear scaling factors can be applied across experiments using the same microarray platform.

CONCLUSIONS

Microarrays provide consistent, reproducible gene expression measurements, which are improved using RNA-Seq as ground truth. We expect that our strategy could be used to improve probe quality for many data sets from major existing repositories.

Using expression profiling to understand the effects of chronic cadmium exposure on MCF-7 breast cancer cells

Cadmium is a metalloestrogen known to activate the estrogen receptor and promote breast cancer cell growth. Previous studies have implicated cadmium in the development of more malignant tumors; however the molecular mechanisms behind this cadmium-induced malignancy remain elusive. Using clonal cell lines derived from exposing breast cancer cells to cadmium for over 6 months (MCF-7-Cd4, -Cd6, -Cd7, -Cd8 and -Cd12), this study aims to identify gene expression signatures associated with chronic cadmium exposure. Our results demonstrate that prolonged cadmium exposure does not merely result in the deregulation of genes but actually leads to a distinctive expression profile. The genes deregulated in cadmium-exposed cells are involved in multiple biological processes (i.e. cell growth, apoptosis, etc.) and molecular functions (i.e. cadmium/metal ion binding, transcription factor activity, etc.). Hierarchical clustering demonstrates that the five clonal cadmium cell lines share a common gene expression signature of breast cancer associated genes, clearly differentiating control cells from cadmium exposed cells. The results presented in this study offer insights into the cellular and molecular impacts of cadmium on breast cancer and emphasize the importance of studying chronic cadmium exposure as one possible mechanism of promoting breast cancer progression.