Batch effect reduction of microarray data with dependent samples using an empirical Bayes approach (BRIDGE)

Batch-effects present challenges in the analysis of high-throughput molecular data and are particularly problematic in longitudinal studies when interest lies in identifying genes/features whose expression changes over time, but time is confounded with batch. While many methods to correct for batch-effects exist, most assume independence across samples; an assumption that is unlikely to hold in longitudinal microarray studies. We propose Batch effect Reduction of mIcroarray data with Dependent samples usinG Empirical Bayes (BRIDGE), a three-step parametric empirical Bayes approach that leverages technical replicate samples profiled at multiple timepoints/batches, so-called “bridge samples”, to inform batch-effect reduction/attenuation in longitudinal microarray studies. Extensive simulation studies and an analysis of a real biological data set were conducted to benchmark the performance of BRIDGE against both ComBat and longitudinal ComBat. Our results demonstrate that while all methods perform well in facilitating accurate estimates of time effects, BRIDGE outperforms both ComBat and longitudinal ComBat in the removal of batch-effects in data sets with bridging samples, and perhaps as a result, was observed to have improved statistical power for detecting genes with a time effect. BRIDGE demonstrated competitive performance in batch effect reduction of confounded longitudinal microarray studies, both in simulated and a real data sets, and may serve as a useful preprocessing method for researchers conducting longitudinal microarray studies that include bridging samples.

A Pre-Processing Pipeline to Quantify, Visualize and Reduce Technical Variation in Protein Microarray Studies

Technical variation, or variation from non-biological sources, is present in most laboratory assays. Correcting for this variation enables analysts to extract a biological signal that informs questions of interest. However, each assay has different sources and levels of technical variation and the choice of correction methods can impact downstream analyses. Compared to similar assays such as DNA microarrays, relatively few methods have been developed and evaluated for protein microarrays, a versatile tool for measuring levels of various proteins in serum samples. Here, we propose a pre-processing pipeline to correct for some common sources of technical variation in protein microarrays. The pipeline builds upon an existing normalization method by using controls to reduce technical variation. We evaluate our method using data from two protein microarray studies, and by simulation. We demonstrate that pre-processing choices impact the fluorescent-intensity based ranks of proteins, which in turn, impact downstream analysis.1Impact StatementProtein microarrays are in wide use in cancer research, infectious disease diagnostics and biomarker identification. To inform research and practice in these and other fields, technical variation must be corrected using normalization and pre-processing. Current protein microarray studies use a variety of normalization methods, many of which were developed for DNA microarrays, and therefore are based on assumptions and data that are not ideal for protein microarrays. To address this issue, we develop, evaluate, and implement a pre-processing pipeline that corrects for technical variation in protein microarrays. We show that pre-processing and normalization directly impact the validity of downstream analysis, and protein-specific approaches are essential.

Morphologic and anatomic response, catalase gene expression in drought varieties of tomato and bioinformatics analyses of microarray studies of the catalase gene

Crops in arid and semi-arid regions are exposed to adverse environmental factors such as drought. Experiments were conducted to determine the morphologic and anatomic response of drought-susceptible and tolerant varieties of tomato (Solanum lycopersicum L.) under drought conditions (100%, 75%, 50%, 25% of field capacity). To investigate the role of antioxidant enzyme, catalase gene expression was examined by real-time RT-qPCR and microarray studies of the catalase gene in tomatoes under stress examined utilizing bioinformatics. The results showed significant morphological changes under drought conditions. Anatomical studies revealed that CaljN3 is more resistant than SuperstrainB varieties under drought stress. Relative expression of the CAT1 gene did not show any significant difference in both Caljn3 and SuperstrainB varieties based on quantitative Real-Time PCR, under drought stress. The bioinformatics results from microarray analysis revealed that this gene did not show a significant difference in expression in any of the cultivars and under any of the stresses. This gene is in the conserve cluster, a cluster with 118 members and a z score of 14.26148. This showed that this cluster is fully protected between two susceptible and tolerant varieties. The enrichment gene of this cluster did not show any significant intracellular pathways. It appears that in response to stress, an activating mechanism other than catalase is necessary. The fight against oxidative stress may begin one step before that of the enzymes and seeks to combat the stressor by activating proteins, especially channels, pumps and some cellular messengers.

A merged microarray meta-dataset for transcriptionally profiling colorectal neoplasm formation and progression

Transcriptional profiling of pre- and post-malignant colorectal cancer (CRC) lesions enable temporal monitoring of molecular events underlying neoplastic progression. However, the most widely used transcriptomic dataset for CRC, TCGA-COAD, is devoid of adenoma samples, which increases reliance on an assortment of disparate microarray studies and hinders consensus building. To address this, we developed a microarray meta-dataset comprising 231 healthy, 132 adenoma, and 342 CRC tissue samples from twelve independent studies. Utilizing a stringent analytic framework, select datasets were downloaded from the Gene Expression Omnibus, normalized by frozen robust multiarray averaging and subsequently merged. Batch effects were then identified and removed by empirical Bayes estimation (ComBat). Finally, the meta-dataset was filtered for low variant probes, enabling downstream differential expression as well as quantitative and functional validation through cross-platform correlation and enrichment analyses, respectively. Overall, our meta-dataset provides a robust tool for investigating colorectal adenoma formation and malignant transformation at the transcriptional level with a pipeline that is modular and readily adaptable for similar analyses in other cancer types.

Identification of Aneuploidy in Dogs Screened by a SNP Microarray

Microarray analysis is an efficient approach for screening and identifying cytogenetic imbalances in humans. SNP arrays, in particular, are a powerful way to identify copy number gains and losses representing aneuploidy and aneusomy, but moreover, allow for the direct assessment of individual genotypes in known disease loci. Using these approaches, trisomies, monosomies and mosaicism of whole chromosomes have been identified in human microarray studies. For canines, this approach is not widely used in clinical laboratory diagnostic practice. In our laboratory, we have implemented the use of a propriety SNP array that represents approximately 650,000 loci across the domestic dog genome. During the validation of this microarray prior to clinical use, we identified three cases of aneuploidy after screening 2,053 dogs of various breeds including monosomy X, trisomy X and an apparent, mosaic trisomy of canine chromosome 38 (CFA 38). This study represents the first use of microarrays for copy number evaluation to identify cytogenetic anomalies in canines. As microarray analysis becomes more routine in canine genetic testing, more cases of chromosome aneuploidy are likely to be uncovered.

Application of proteome profiling in the clinical diagnosis of breast cancer

The early diagnosis, prognosis, and anticipation of breast cancer are crucial for proper treatment and patient survival. This disease imposes quite a severe health care encumbrance on women globally. Breast cancer classification has emphasized several global efforts, and analysis of the subtypes of the molecular basis of breast cancer has aimed to associate them with clinical outcomes and improve the current diagnostic routine. Since the last two decades, proteomics-based methods for studying breast cancer's natural history and treatment are gaining traction. In this review, some of the proteome profiling studies of tissues, plasma, serum and saliva conducted mainly by mass spectrometry-based approaches – including MALDI-TOF and SELDI-TOF are discussed. This review also emphasized tissue microarray studies and their role in identifying clinical tissues and markers in breast cancer.

Development and applications of sialoglycan-recognizing probes (SGRPs) with defined specificities: exploring the dynamic mammalian sialoglycome

Glycans that are abundantly displayed on vertebrate cell surface and secreted molecules are often capped with terminal sialic acids (Sias). These diverse 9-carbon-backbone monosaccharides are involved in numerous intrinsic biological processes. They also interact with commensals and pathogens, while undergoing dynamic changes in time and space, often influenced by environmental conditions. However, most of this sialoglycan complexity and variation remains poorly characterized by conventional techniques, which often tend to destroy or overlook crucial aspects of Sia diversity and/or fail to elucidate native structures in biological systems i.e., in the intact sialome. To date, in situ detection and analysis of sialoglycans has largely relied on the use of plant lectins, sialidases or antibodies, whose preferences (with certain exceptions) are limited and/or uncertain. We took advantage of naturally-evolved microbial molecules (bacterial adhesins, toxin subunits and viral hemagglutinin-esterases) that recognize sialoglycans with defined specificity to delineate 9 classes of Sialoglycan Recognizing Probes (SGRPs: SGRP1SGRP9) that can be used to explore mammalian sialome changes in a simple and systematic manner, using techniques common in most laboratories. SGRP candidates with specificity defined by sialoglycan microarray studies were engineered as tagged probes, each with a corresponding non-binding mutant probe as a simple and reliable negative control. The optimized panel of SGRPs can be used in methods commonly available in most bioscience labs, such as ELISA, Western Blot, flow cytometry and histochemistry. To demonstrate the utility of this approach, we provide examples of sialoglycome differences in tissues from C57BL/6 wild type mice and human-like Cmah-/- mice.

Gene Regulatory Network of ETS Domain Transcription Factors in Different Stages of Glioma

The ETS domain family of transcription factors is involved in a number of biological processes, and is commonly misregulated in various forms of cancer. Using microarray datasets from patients with different grades of glioma, we have analyzed the expression profiles of various ETS genes, and have identified ETV1, ELK3, ETV4, ELF4, and ETV6 as novel biomarkers for the identification of different glioma grades. We have further analyzed the gene regulatory networks of ETS transcription factors and compared them to previous microarray studies, where Elk-1-VP16 or PEA3-VP16 were overexpressed in neuroblastoma cell lines, and we identify unique and common regulatory networks for these ETS proteins.