Biochemistry. A postgenomic visual icon

Multiplexed MRM-based proteomics for identification of circulating proteins as biomarkers of cardiovascular damage progression associated with diabetes mellitus

BACKGROUND

Type 2 diabetes mellitus (T2DM) increases the risk of coronary heart disease (CHD) by 2-4 fold, and is associated with endothelial dysfunction, dyslipidaemia, insulin resistance, and chronic hyperglycaemia. The aim of this investigation was to assess, by a multimarker mass spectrometry approach, the predictive role of circulating proteins as biomarkers of cardiovascular damage progression associated with diabetes mellitus.

METHODS

The study considered 34 patients with both T2DM and CHD, 31 patients with T2DM and without CHD, and 30 patients without diabetes with a diagnosis of CHD. Plasma samples of subjects were analysed through a multiplexed targeted liquid chromatography mass spectrometry (LC-MS)-based assay, namely Multiple Reaction Monitoring (MRM), allowing the simultaneous detection of peptides derived from a protein of interest. Gene Ontology (GO) Analysis was employed to identify enriched GO terms in the biological process, molecular function, or cellular component categories. Non-parametric multivariate methods were used to classify samples from patients and evaluate the relevance of the analysed proteins' panel.

RESULTS

A total of 81 proteins were successfully quantified in the human plasma samples. Gene Ontology analysis assessed terms related to blood microparticles, extracellular exosomes and collagen-containing extracellular matrix. Preliminary evaluation using analysis of variance (ANOVA) of the differences in the proteomic profile among patient groups identified 13 out of the 81 proteins as significantly different. Multivariate analysis, including cluster analysis and principal component analysis, identified relevant grouping of the 13 proteins. The first main cluster comprises apolipoprotein C-III, apolipoprotein C-II, apolipoprotein A-IV, retinol-binding protein 4, lysozyme C and cystatin-C; the second one includes, albeit with sub-grouping, alpha 2 macroglobulin, afamin, kininogen 1, vitronectin, vitamin K-dependent protein S, complement factor B and mannan-binding lectin serine protease 2. Receiver operating characteristic (ROC) curves obtained with the 13 selected proteins using a nominal logistic regression indicated a significant overall distinction (p < 0.001) among the three groups of subjects, with area under the ROC curve (AUC) ranging 0.91-0.97, and sensitivity and specificity ranging from 85 to 100%.

CONCLUSIONS

Targeted mass spectrometry approach indicated 13 multiple circulating proteins as possible biomarkers of cardiovascular damage progression associated with T2DM, with excellent classification results in terms of sensitivity and specificity.

A Comparative Study Using Vestibular Mapping in Sudden Sensorineural Hearing Loss With and Without Vertigo

OBJECTIVE

To investigate the impairment patterns in peripheral vestibular organs in sudden sensorineural hearing loss (SSNHL) with and without vertigo.

STUDY DESIGN

Retrospective study.

SETTING

Single tertiary medical center.

METHODS

Data from 165 SSNHL patients in a tertiary referral center from January 2017 to December 2022 were retrospectively analyzed. All patients underwent a video head impulse test, vestibular evoked myogenic potential test, and pure-tone audiometry. Hierarchical cluster analysis was performed to investigate vestibular impairment patterns. The prognosis of the hearing was determined using American Academy of Otolaryngology-Head and Neck Surgery recommendations.

RESULTS

After excluding patients with vestibular schwannoma and Meniere's disease, 152 patients were included in this study. A total of 73 of 152 patients were categorized as SSNHL with vertigo (SSNHL_V) and showed an independent merge of the posterior semicircular canal (PSCC) in cluster analysis. A total of 79 of 152 patients were categorized as SSNHL without vertigo (SSNHL_N) and showed an independent merge of saccule in cluster analysis. The PSCC (56.2%) and saccule (20.3%) were the most frequently impaired vestibular organs in SSNHL_V and SSNHL_N, respectively. In terms of prognosis, 106 of 152 patients had partial/no recovery and showed an independent merge of the PSCC in cluster analysis. A total of 46 of 152 patients had a complete recovery and showed an independent merge of the saccule in cluster analysis.

CONCLUSION

A tendency of isolated PSCC dysfunction was seen in SSNHL_V and partial/no recovery. A tendency of isolated saccular dysfunction was seen in SSNHL_N and complete recovery. Different treatments might be needed in SSNHL depending on the presence of vertigo.

Visual representations of single- and multi-sport participation in a youth swimming sample: Implications for definitions and discussions of early specialization

Academic literature and sport policy documents have cited concerns about an increasing prevalence of early sport specialization, with associated burnout, dropout, and injury. However, evidence to support such statements is limited. Definitions of early specialization vary, but a common criterion is continued participation in a single sport, prior to adolescence. We explored the prevalence of single-sport participation and other patterns of sport involvement from ages 6-12 in a Canadian swimming sample using retrospective longitudinal methods. Parents of 236 competitive swimmers (ages 12-17) completed surveys on their children's sport backgrounds, including the number of sports participated in annually from age 6-12. A cluster heat map elucidated single- and multi-sport patterns over time. Mixed analyses of variance tested for differences by gender and club type. Fourteen percent of our sample showed stable participation in either one sport or multiple sports per year over time, 25% decreased their annual number of sports, and 60% increased. This trend of increasing, rather than decreasing the number of sports in their annual activity roster when approaching age 12 was particularly pronounced for girls. Only 10 participants (4% of the sample) consistently engaged in a single sport each year from age 6-12. Summer (seasonal) swimmers consistently did more sports than year-round swimmers. Overall, our findings showed highly idiosyncratic longitudinal patterns of sport participation that did not easily conform to current sport activity guidelines. We also found similar idiosyncrasy in an ad-hoc analysis of participants who had dropped out of swimming a year later. If single-sport participation is considered a key criterion for defining early specialization, our findings suggest the prevailing narrative around early specialization may be overstated in relation to the number of single-sport athletes. Alternatively, other components of early specialization may be more prevalent and deserving of attention due to possible associations with harmful outcomes.

Data Visualization Tools of Tobacco Product Use Patterns, Transitions and Sex Differences in the PATH Youth Data

INTRODUCTION

Evaluations of multiple tobacco product use and temporal changes in patterns of use are complicated by a large number of combinations and transitions. Visualization tools could easily identify most common patterns and transitions.

METHODS

Set intersection bar plots describe ever use of five tobacco products among 12-17 years old youth in wave 1 of the Population Assessment of Tobacco and Health (PATH) study (N = 11 497). Heat maps visualize unweighted frequencies of transitions from ever use at wave 1 (2013-2014) to past 12-month use at wave 2 (2014-2015). Weighted calibrated heat maps assess differences in relative frequencies of transitions by pattern at wave 1 and identify differences in transitions by sex.

RESULTS

The most common tobacco product ever use patterns in wave 1 were of cigarettes only, e-cigarettes only or hookah only, followed by ever use of both cigarettes and e-cigarettes. Initiation of use between waves was uncommon. The most frequent transition among those who reported use at wave 2 but not at wave 1 (N = 971) was to e-cigarette use (N = 301). However, among e-cigarette-only ever users at wave 1 (N = 260), about half did not report any product use at wave 2. Use of three or more products remained stable. Adolescent girls compared to boys appeared more likely to report hookah use at both waves.

CONCLUSION

Set intersection bar plots and heat maps are useful for visualizing tobacco product use patterns and transitions, especially for multiple products. Both techniques could identify common problematic tobacco use patterns across and within populations.

IMPLICATIONS

Given the growing complexity of the youth tobacco use landscape, approaches to efficiently communicate patterns of multiple tobacco product use should be used more often. This study introduces set intersection bar plots and modified versions of heat maps to the tobacco product literature and illustrates their use in the PATH youth sample. These techniques are useful for visualizing absolute and relative frequencies of multiple possible patterns and transitions. They also suggest targets for subsequent statistical inference such as sex differences in hookah use. The methods can be applied more generally for data visualization wherever large number of combinations occurs.

Vestibular mapping in patients with unilateral peripheral-vestibular deficits

OBJECTIVE

To test the hypothesis that patterns of semicircular canal (SCC) and otolith impairment in unilateral vestibular loss depend on the underlying disorders, we analyzed peripheral-vestibular function of all 5 vestibular sensors.

METHODS

For this retrospective case series, we screened the hospital video-head-impulse test database (n = 4,983) for patients with unilaterally impaired SCC function who also received ocular vestibular-evoked myogenic potentials and cervical vestibular-evoked myogenic potentials (n = 302). Frequency of impairment of vestibular end organs (horizontal/anterior/posterior SCC, utriculus/sacculus) was analyzed with hierarchical cluster analysis and correlated with the underlying etiology.

RESULTS

Acute vestibular neuropathy (AVN) (37.4%, 113 of 302), vestibular schwannoma (18.2%, 55 of 302), and acute cochleovestibular neuropathy (6.6%, 20 of 302) were most frequent. Horizontal SCC impairment (87.4%, 264 of 302) was more frequent (p < 0.001) than posterior (47.4%, 143 of 302) and anterior (37.8%, 114 of 302) SCC impairment. Utricular damage (58%, 175 of 302) was noted more often (p = 0.003) than saccular impairment (32%, 98 of 302). On average, 2.6 (95% confidence interval 2.48-2.78) vestibular sensors were deficient, with higher numbers (p ≤ 0.017) for acute cochleovestibular neuropathy and vestibular schwannoma than for AVN, Menière disease, and episodic vestibular syndrome. In hierarchical cluster analysis, early mergers (posterior SCC/sacculus; anterior SCC/utriculus) pointed to closer pathophysiologic association of these sensors, whereas the late merger of the horizontal canal indicated a more distinct state.

CONCLUSIONS

While the extent and pattern of vestibular impairment critically depended on the underlying disorder, more limited damage in AVN and Menière disease was noted, emphasizing the individual range of loss of function and the value of vestibular mapping. Likely, both the anatomic properties of the different vestibular end organs and their vulnerability to external factors contribute to the relative sparing of the vertical canals and the sacculus.

Interactive Clustered Heat Map Builder: An easy web-based tool for creating sophisticated clustered heat maps

Clustered heat maps are the most frequently used graphics for visualization and interpretation of genome-scale molecular profiling data in biology.  Construction of a heat map generally requires the assistance of a biostatistician or bioinformatics analyst capable of working in R or a similar programming language to transform the study data, perform hierarchical clustering, and generate the heat map.  Our web-based Interactive Heat Map Builder can be used by investigators with no bioinformatics experience to generate high-caliber, publication quality maps.  Preparation of the data and construction of a heat map is rarely a simple linear process.  Our tool allows a user to move back and forth iteratively through the various stages of map generation to try different options and approaches.  Finally, the heat map the builder creates is available in several forms, including an interactive Next-Generation Clustered Heat Map that can be explored dynamically to investigate the results more fully.

Interactive Clustered Heat Map Builder: An easy web-based tool for creating sophisticated clustered heat maps

Clustered heat maps are the most frequently used graphics for visualization and interpretation of genome-scale molecular profiling data in biology.  Construction of a heat map generally requires the assistance of a biostatistician or bioinformatics analyst capable of working in R or a similar programming language to transform the study data, perform hierarchical clustering, and generate the heat map.  Our web-based Interactive Heat Map Builder can be used by investigators with no bioinformatics experience to generate high-caliber, publication quality maps.  Preparation of the data and construction of a heat map is rarely a simple linear process.  Our tool allows a user to move back and forth iteratively through the various stages of map generation to try different options and approaches.  Finally, the heat map the builder creates is available in several forms, including an interactive Next-Generation Clustered Heat Map that can be explored dynamically to investigate the results more fully.

Survey Assessment for Decision Support Using Self-Organizing Maps Profile Characterization with an Odds and Cluster Heat Map: Application to Children's Perception of Urban School Environments

The interpretation of opinion and satisfaction surveys based exclusively on statistical analysis often faces difficulties due to the nature of the information and the requirements of the available statistical methods. These difficulties include the concurrence of categorical information with answers based on Likert scales with only a few levels, or the distancing of the necessary heuristic approach of the decision support system (DSS). The artificial neural network used for data analysis, called Kohonen or self-organizing maps (SOM), although rarely used for survey analysis, has been applied in many fields, facilitating the graphical representation and the simple interpretation of high-dimensionality data. This clustering method, based on unsupervised learning, also allows obtaining profiles of respondents without the need to provide additional information for the creation of these clusters. In this work, we propose the identification of profiles using SOM for evaluating opinion surveys. Subsequently, non-parametric chi-square tests were first conducted to contrast whether answer was independent of each profile found, and in the case of statistical significance (p ≤ 0.05), the odds ratio was evaluated as an indicator of the effect size of such dependence. Finally, all results were displayed in an odds and cluster heat map so that they could be easily interpreted and used to make decisions regarding the survey results. The methodology was applied to the analysis of a survey based on forms administered to children (N = 459) about their perception of the urban environment close to their school, obtaining relevant results, facilitating results interpretation, and providing support to the decision-process.

Systematic data analysis and data mining in CatWalk gait analysis by heat mapping exemplified in rodent models for neurodegenerative diseases

BACKGROUND

Motor impairment appears as a characteristic symptom of several diseases and injuries. Therefore, tests for analyzing motor dysfunction are widely applied across preclinical models and disease stages. Among those, gait analysis tests are commonly used, but they generate a huge number of gait parameters. Thus, complications in data analysis and reporting raise, which often leads to premature parameter selection.

NEW METHODS

In order to avoid arbitrary parameter selection, we present here a systematic initial data analysis by utilizing heat-maps for data reporting. We exemplified this approach within an intervention study, as well as applied it to two longitudinal studies in rodent models related to Parkinson's disease (PD) and Huntington disease (HD).

RESULTS

The systematic initial data analysis (IDA) is feasible for exploring gait parameters, both in experimental and longitudinal studies. The resulting heat maps provided a visualization of gait parameters within a single chart, highlighting important clusters of differences.

COMPARISON WITH EXISTING METHOD

Often, premature parameter selection is practiced, lacking comprehensiveness. Researchers often use multiple separated graphs on distinct gait parameters for reporting. Additionally, negative results are often not reported.

CONCLUSIONS

Heat mapping utilized in initial data analysis is advantageous for reporting clustered gait parameter differences in one single chart and improves data mining.

heatmaply: an R package for creating interactive cluster heatmaps for online publishing

Summary

heatmaply is an R package for easily creating interactive cluster heatmaps that can be shared online as a stand-alone HTML file. Interactivity includes a tooltip display of values when hovering over cells, as well as the ability to zoom in to specific sections of the figure from the data matrix, the side dendrograms, or annotated labels. Thanks to the synergistic relationship between heatmaply and other R packages, the user is empowered by a refined control over the statistical and visual aspects of the heatmap layout.

Availability and implementation

The heatmaply package is available under the GPL-2 Open Source license. It comes with a detailed vignette, and is freely available from: http://cran.r-project.org/package=heatmaply.

Supplementary information

Supplementary data are available at Bioinformatics online.

Covariate-adjusted heatmaps for visualizing biological data via correlation decomposition

Motivation

Heatmap is a popular visualization technique in biology and related fields. In this study, we extend heatmaps within the framework of matrix visualization (MV) by incorporating a covariate adjustment process through the estimation of conditional correlations. MV can explore the embedded information structure of high-dimensional large-scale datasets effectively without dimension reduction. The benefit of the proposed covariate-adjusted heatmap is in the exploration of conditional association structures among the subjects or variables that cannot be done with conventional MV.

Results

For adjustment of a discrete covariate, the conditional correlation is estimated by the within and between analysis. This procedure decomposes a correlation matrix into the within- and between-component matrices. The contribution of the covariate effects can then be assessed through the relative structure of the between-component to the original correlation matrix while the within-component acts as a residual. When a covariate is of continuous nature, the conditional correlation is equivalent to the partial correlation under the assumption of a joint normal distribution. A test is then employed to identify the variable pairs which possess the most significant differences at varying levels of correlation before and after a covariate adjustment. In addition, a z-score significance map is constructed to visualize these results. A simulation and three biological datasets are employed to illustrate the power and versatility of our proposed method.

Availability and implementation

GAP is available to readers and is free to non-commercial applications. The installation instructions, the user's manual, and the detailed tutorials can be found at http://gap.stat.sinica.edu.tw/Software/GAP.

Supplementary information

Supplementary Data are available at Bioinformatics online.

Discrimination and classification of extra virgin olive oil using a chemometric approach based on TMS-4,4'-desmetylsterols GC(FID) fingerprints of edible vegetable oils

A single out-line HPLC-GC (FID) analytical method is applied to acquire the chromatographic fingerprint characteristic of the TMS-4,4'-desmetylsterol derivative fraction of several marketed edible vegetable oils in order to identify and discriminate the most valuable extra-virgin olive oils from the other vegetal oils (canola, corn, grape seed, linseed, olive pomace, peanut, rapeseed, soybean, sesame, seeds (non-specified composition but usually a blend of corn and sunflower) and sunflower). The natural structure of the preprocessed data undergoes a preliminary exploration using principal component analysis and heat map-based cluster analysis. A partial least squares-discriminant model is first trained from 53 oil samples (only 3 latent variables) and externally validated from 18 test oil samples. No classification errors are found and all the test samples are correctly classified. Additional classification models are also built in order to discriminate among vegetables-oil families and excellent results have been also achieved.

Hierarchical Cluster Analysis of Semicircular Canal and Otolith Deficits in Bilateral Vestibulopathy

Background

Gait imbalance and oscillopsia are frequent complaints of bilateral vestibular loss (BLV). Video-head-impulse testing (vHIT) of all six semicircular canals (SCCs) has demonstrated varying involvement of the different canals. Sparing of anterior-canal function has been linked to aminoglycoside-related vestibulopathy and Menière’s disease. We hypothesized that utricular and saccular impairment [assessed by vestibular-evoked myogenic potentials (VEMPs)] may be disease-specific also, possibly facilitating the differential diagnosis.

Methods

We searched our vHIT database (n = 3,271) for patients with bilaterally impaired SCC function who also received ocular VEMPs (oVEMPs) and cervical VEMPs (cVEMPs) and identified 101 patients. oVEMP/cVEMP latencies above the 95th percentile and peak-to-peak amplitudes below the 5th percentile of normal were considered abnormal. Frequency of impairment of vestibular end organs (horizontal/anterior/posterior SCC, utriculus/sacculus) was analyzed with hierarchical cluster analysis and correlated with the underlying etiology.

Results

Rates of utricular and saccular loss of function were similar (87.1 vs. 78.2%, p = 0.136, Fisher’s exact test). oVEMP abnormalities were found more frequent in aminoglycoside-related bilateral vestibular loss (BVL) compared with Menière’s disease (91.7 vs. 54.6%, p = 0.039). Hierarchical cluster analysis indicated distinct patterns of vestibular end-organ impairment, showing that the results for the same end-organs on both sides are more similar than to other end-organs. Relative sparing of anterior-canal function was reflected in late merging with the other end-organs, emphasizing their distinct state. An anatomically corresponding pattern of SCC/otolith hypofunction was present in 60.4% (oVEMPs vs. horizontal SCCs), 34.7% (oVEMPs vs. anterior SCCs), and 48.5% (cVEMPs vs. posterior SCCs) of cases. Average (±1 SD) number of damaged sensors was 6.8 ± 2.2 out of 10. Significantly (p < 0.001) more sensors were impaired in patients with aminoglycoside-related BVL (8.1 ± 1.2) or inner-ear infections (8.7 ± 1.8) compared with Menière-related BVL (5.5 ± 1.5).

Discussion

Hierarchical cluster analysis may help differentiate characteristic patterns of BVL. With a prevalence of ≈80%, utricular and/or saccular impairment is frequent in BVL. The extent of SCC and otolith impairment was disease-dependent, showing most extensive damage in BVL related to inner-ear infection and aminoglycoside-exposure and more selective impairment in Menière’s disease. Specifically, assessing utricular function may help in the distinction between aminoglycoside-related BVL and bilateral Menière’s disease.

A Galaxy Implementation of Next-Generation Clustered Heatmaps for Interactive Exploration of Molecular Profiling Data

Clustered heatmaps are the most frequently used graphics for visualization of molecular profiling data in biology. However, they are generally rendered as static, or only modestly interactive, images. We have now used recent advances in web technologies to produce interactive "next-generation" clustered heatmaps (NG-CHM) that enable extreme zooming and navigation without loss of resolution. NG-CHMs also provide link-outs to additional information sources and include other features that facilitate deep exploration of the biology behind the image. Here, we describe an implementation of the NG-CHM system in the Galaxy bioinformatics platform. We illustrate the algorithm and available computational tool using RNA-seq data from The Cancer Genome Atlas program's Kidney Clear Cell Carcinoma project. Cancer Res; 77(21); e23-26. ©2017 AACR.

Unboxing cluster heatmaps

Background

Cluster heatmaps are commonly used in biology and related fields to reveal hierarchical clusters in data matrices. This visualization technique has high data density and reveal clusters better than unordered heatmaps alone. However, cluster heatmaps have known issues making them both time consuming to use and prone to error. We hypothesize that visualization techniques without the rigid grid constraint of cluster heatmaps will perform better at clustering-related tasks.

Results

We developed an approach to “unbox” the heatmap values and embed them directly in the hierarchical clustering results, allowing us to use standard hierarchical visualization techniques as alternatives to cluster heatmaps. We then tested our hypothesis by conducting a survey of 45 practitioners to determine how cluster heatmaps are used, prototyping alternatives to cluster heatmaps using pair analytics with a computational biologist, and evaluating those alternatives with hour-long interviews of 5 practitioners and an Amazon Mechanical Turk user study with approximately 200 participants. We found statistically significant performance differences for most clustering-related tasks, and in the number of perceived visual clusters. Visit git.io/vw0t3 for our results.

Conclusions

The optimal technique varied by task. However, gapmaps were preferred by the interviewed practitioners and outperformed or performed as well as cluster heatmaps for clustering-related tasks. Gapmaps are similar to cluster heatmaps, but relax the heatmap grid constraints by introducing gaps between rows and/or columns that are not closely clustered. Based on these results, we recommend users adopt gapmaps as an alternative to cluster heatmaps.

Machine-learned cluster identification in high-dimensional data

•

High-dimensional biomedical data are commonly searched for structures.

•

Common cluster algorithms may impose non-existent clusters or assign data to the wrong clusters.

•

We highlight accepted proposals of using emergent self-organizing maps for clustering.

•

The addition of the U-matrix provides a visually controllable representation of cluster structure.

Background

High-dimensional biomedical data are frequently clustered to identify subgroup structures pointing at distinct disease subtypes. It is crucial that the used cluster algorithm works correctly. However, by imposing a predefined shape on the clusters, classical algorithms occasionally suggest a cluster structure in homogenously distributed data or assign data points to incorrect clusters. We analyzed whether this can be avoided by using emergent self-organizing feature maps (ESOM).

Methods

Data sets with different degrees of complexity were submitted to ESOM analysis with large numbers of neurons, using an interactive R-based bioinformatics tool. On top of the trained ESOM the distance structure in the high dimensional feature space was visualized in the form of a so-called U-matrix. Clustering results were compared with those provided by classical common cluster algorithms including single linkage, Ward and k-means.

Results

Ward clustering imposed cluster structures on cluster-less “golf ball”, “cuboid” and “S-shaped” data sets that contained no structure at all (random data). Ward clustering also imposed structures on permuted real world data sets. By contrast, the ESOM/U-matrix approach correctly found that these data contain no cluster structure. However, ESOM/U-matrix was correct in identifying clusters in biomedical data truly containing subgroups. It was always correct in cluster structure identification in further canonical artificial data. Using intentionally simple data sets, it is shown that popular clustering algorithms typically used for biomedical data sets may fail to cluster data correctly, suggesting that they are also likely to perform erroneously on high dimensional biomedical data.

Conclusions

The present analyses emphasized that generally established classical hierarchical clustering algorithms carry a considerable tendency to produce erroneous results. By contrast, unsupervised machine-learned analysis of cluster structures, applied using the ESOM/U-matrix method, is a viable, unbiased method to identify true clusters in the high-dimensional space of complex data.

Global metabolomics reveals potential urinary biomarkers of esophageal squamous cell carcinoma for diagnosis and staging

We performed a metabolomics study using liquid chromatography-mass spectrometry (LC-MS) combined with multivariate data analysis (MVDA) to discriminate global urine profiles in urine samples from esophageal squamous cell carcinoma (ESCC) patients and healthy controls (NC). Our work evaluated the feasibility of employing urine metabolomics for the diagnosis and staging of ESCC. The satisfactory classification between the healthy controls and ESCC patients was obtained using the MVDA model, and obvious classification of early-stage and advanced-stage patients was also observed. The results suggest that the combination of LC-MS analysis and MVDA may have potential applications for ESCC diagnosis and staging. We then conducted LC-MS/MS experiments to identify the potential biomarkers with large contributions to the discrimination. A total of 83 potential diagnostic biomarkers for ESCC were screened out, and 19 potential biomarkers were identified; the variations between the differences in staging using these potential biomarkers were further analyzed. These biomarkers may not be unique to ESCCs, but instead result from any malignant disease. To further elucidate the pathophysiology of ESCC, we studied related metabolic pathways and found that ESCC is associated with perturbations of fatty acid β-oxidation and the metabolism of amino acids, purines, and pyrimidines.

InCHlib - interactive cluster heatmap for web applications

BACKGROUND

Hierarchical clustering is an exploratory data analysis method that reveals the groups (clusters) of similar objects. The result of the hierarchical clustering is a tree structure called dendrogram that shows the arrangement of individual clusters. To investigate the row/column hierarchical cluster structure of a data matrix, a visualization tool called 'cluster heatmap' is commonly employed. In the cluster heatmap, the data matrix is displayed as a heatmap, a 2-dimensional array in which the colour of each element corresponds to its value. The rows/columns of the matrix are ordered such that similar rows/columns are near each other. The ordering is given by the dendrogram which is displayed on the side of the heatmap.

RESULTS

We developed InCHlib (Interactive Cluster Heatmap Library), a highly interactive and lightweight JavaScript library for cluster heatmap visualization and exploration. InCHlib enables the user to select individual or clustered heatmap rows, to zoom in and out of clusters or to flexibly modify heatmap appearance. The cluster heatmap can be augmented with additional metadata displayed in a different colour scale. In addition, to further enhance the visualization, the cluster heatmap can be interconnected with external data sources or analysis tools. Data clustering and the preparation of the input file for InCHlib is facilitated by the Python utility script inchlib_clust.

CONCLUSIONS

The cluster heatmap is one of the most popular visualizations of large chemical and biomedical data sets originating, e.g., in high-throughput screening, genomics or transcriptomics experiments. The presented JavaScript library InCHlib is a client-side solution for cluster heatmap exploration. InCHlib can be easily deployed into any modern web application and configured to cooperate with external tools and data sources. Though InCHlib is primarily intended for the analysis of chemical or biological data, it is a versatile tool which application domain is not limited to the life sciences only.

Scientists' sense making when hypothesizing about disease mechanisms from expression data and their needs for visualization support

A common class of biomedical analysis is to explore expression data from high throughput experiments for the purpose of uncovering functional relationships that can lead to a hypothesis about mechanisms of a disease. We call this analysis expression driven, -omics hypothesizing. In it, scientists use interactive data visualizations and read deeply in the research literature. Little is known, however, about the actual flow of reasoning and behaviors (sense making) that scientists enact in this analysis, end-to-end. Understanding this flow is important because if bioinformatics tools are to be truly useful they must support it. Sense making models of visual analytics in other domains have been developed and used to inform the design of useful and usable tools. We believe they would be helpful in bioinformatics. To characterize the sense making involved in expression-driven, -omics hypothesizing, we conducted an in-depth observational study of one scientist as she engaged in this analysis over six months. From findings, we abstracted a preliminary sense making model. Here we describe its stages and suggest guidelines for developing visualization tools that we derived from this case. A single case cannot be generalized. But we offer our findings, sense making model and case-based tool guidelines as a first step toward increasing interest and further research in the bioinformatics field on scientists' analytical workflows and their implications for tool design.

CoryneBase: Corynebacterium genomic resources and analysis tools at your fingertips

Corynebacteria are used for a wide variety of industrial purposes but some species are associated with human diseases. With increasing number of corynebacterial genomes having been sequenced, comparative analysis of these strains may provide better understanding of their biology, phylogeny, virulence and taxonomy that may lead to the discoveries of beneficial industrial strains or contribute to better management of diseases. To facilitate the ongoing research of corynebacteria, a specialized central repository and analysis platform for the corynebacterial research community is needed to host the fast-growing amount of genomic data and facilitate the analysis of these data. Here we present CoryneBase, a genomic database for Corynebacterium with diverse functionality for the analysis of genomes aimed to provide: (1) annotated genome sequences of Corynebacterium where 165,918 coding sequences and 4,180 RNAs can be found in 27 species; (2) access to comprehensive Corynebacterium data through the use of advanced web technologies for interactive web interfaces; and (3) advanced bioinformatic analysis tools consisting of standard BLAST for homology search, VFDB BLAST for sequence homology search against the Virulence Factor Database (VFDB), Pairwise Genome Comparison (PGC) tool for comparative genomic analysis, and a newly designed Pathogenomics Profiling Tool (PathoProT) for comparative pathogenomic analysis. CoryneBase offers the access of a range of Corynebacterium genomic resources as well as analysis tools for comparative genomics and pathogenomics. It is publicly available at http://corynebacterium.um.edu.my/.

Employing heat maps to mine associations in structured routine care data

OBJECTIVE

Mining the electronic medical record (EMR) has the potential to deliver new medical knowledge about causal effects, which are hidden in statistical associations between different patient attributes. It is our goal to detect such causal mechanisms within current research projects which include e.g. the detection of determinants of imminent ICU readmission. An iterative statistical approach to examine each set of considered attribute pairs delivers potential answers but is difficult to interpret. Therefore, we aimed to improve the interpretation of the resulting matrices by the use of heat maps. We propose strategies to adapt heat maps for the search for associations and causal effects within routine EMR data.

METHODS

Heat maps visualize tabulated metric datasets as grid-like choropleth maps, and thus present measures of association between numerous attribute pairs clearly arranged. Basic assumptions about plausible exposures and outcomes are used to allocate distinct attribute sets to both matrix dimensions. The image then avoids certain redundant graphical elements and provides a clearer picture of the supposed associations. Specific color schemes have been chosen to incorporate preexisting information about similarities between attributes. The use of measures of association as a clustering input has been taken as a trigger to apply transformations which ensure that distance metrics always assume finite values and treat positive and negative associations in the same way. To evaluate the general capability of the approach, we conducted analyses of simulated datasets and assessed diagnostic and procedural codes in a large routine care dataset.

RESULTS

Simulation results demonstrate that the proposed clustering procedure rearranges attributes similar to simulated statistical associations. Thus, heat maps are an excellent tool to indicate whether associations concern the same attributes or different ones, and whether affected attribute sets conform to any preexisting relationship between attributes. The dendrograms help in deciding if contiguous sequences of attributes effectively correspond to homogeneous attribute associations. The exemplary analysis of a routine care dataset revealed patterns of associations that follow plausible medical constellations for several diseases and the associated medical procedures and activities. Cases with breast cancer (ICD C50), for example, appeared to be associated with radiation therapy (8-52). In cross check, approximately 60 percent of the attribute pairs in this dataset showed a strong negative association, which can be explained by diseases treated in a medical specialty which routinely does not perform the respective procedures in these cases. The corresponding diagram clearly reflects these relationships in the shape of coherent subareas.

CONCLUSION

We could demonstrate that heat maps of measures of association are effective for the visualization of patterns in routine care EMRs. The adjustable method for the assignment of attributes to image dimensions permits a balance between the display of ample information and a favorable level of graphical complexity. The scope of the search can be adapted by the use of pre-existing assumptions about plausible effects to select exposure and outcome attributes. Thus, the proposed method promises to simplify the detection of undiscovered causal effects within routine EMR data.

Abstract 5132: Interactively exploring patterns in TCGA data: a web-based compendium of ‘next-generation’ clustered heat maps

Each of the 5 TCGA marker paper published in Nature to date has included at least one clustered heat map (CHM). We introduced CHMs in the early 1990’s for pharmacogenomic analysis (1) and later for integrated visualization of genomic, transcriptomic, proteomic, pharmacological, and functional data (1). As the ubiquitous first-order way of visualizing omic data, CHMs have appeared in many thousands of publications (3–9), including those from TCGA. We have elsewhere summarized their limitations (10).

One such limitation is that CHMs are generally static images. We therefore initiated the next-generation CHM (NG-CHM) project, using an image-tiling technology similar to that in Google Maps for navigation and extreme drill-down without loss of resolution. Once the CHM has been zoomed sufficiently, labels (e.g., gene, protein, or drug names) appear on the image's axes. Clicking on a label produces a menu of link-outs (e.g., to GeneCards, Google, PubMed). For gene vs. gene maps, each pixel can represent a color-coded Pearson correlation coefficient. Clicking on the pixel pulls up the corresponding data scattergram, bootstrap statistics, literature references, or pathway relationships. Strong usability features include floating windows, flexible search tools, cluster selection tools, customizable re-coloring of the CHM, and high-quality PDF's suitable for publication. NG-CHMs are a major resource for exploratory analysis and visualization in multiple projects of TCGA and other large-scale molecular profiling programs. Explore interactive versions for TCGA breast, colorectal, lung squamous, and glioblastoma data at http://bioinformatics.mdanderson.org/main/TCGA/NGCHM.

Supported in part by NCI Grant No. U24CA143883, by a gift from the Mary K. Chapman Foundation, and by a grant from the Michael and Susan Dell Foundation honoring Lorraine Dell.

Citation Format: John N. Weinstein, David W. Kane, Rehan Akbani, Deepti Dodda, Lam Nguyen, Michael C. Ryan, Chris Wakefield, Bradley M. Broom. Interactively exploring patterns in TCGA data: a web-based compendium of ‘next-generation’ clustered heat maps. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 5132. doi:10.1158/1538-7445.AM2013-5132

Global and targeted metabolomics of esophageal squamous cell carcinoma discovers potential diagnostic and therapeutic biomarkers

Diagnostic and therapeutic biomarkers useful for esophageal squamous cell carcinoma (ESCC) have the ability to increase the long term survival of cancer patients. A metabolomics study, using plasma from four groups including ESCC patients before, during, and after chemoradiotherapy (CRT) and healthy controls, was originally carried out by LC-MS to determine global alterations in the metabolic profiles and find biomarkers potentially applicable to diagnosis and monitoring treatment effects. It is worth pointing out that a clear clustering and separation of metabolic data from the four groups was observed, which indicated that disease status and treatment intervention resulted in specific metabolic perturbations in the patients. A series of metabolites were found to be significantly altered in ESCC patients versus healthy controls and in pre- versus post-treatment patients based on multivariate statistical data analysis (MVDA). To further validate the reliability of these potential biomarkers, an independent validation was performed by using the selected reaction monitoring (SRM) based targeted approach. Finally, 18 most significantly altered plasma metabolites in ESCC patients, relative to healthy controls, were tentatively identified as lysophosphatidylcholines (lysoPCs), fatty acids, l-carnitine, acylcarnitines, organic acids, and a sterol metabolite. The classification performance of these metabolites were analyzed by receiver operating characteristic (ROC)(1) analysis and a biomarker panel was generated. Together, biological significance of these metabolites was discussed. Comparison between pre- and post-treatment patients generated 11 metabolites as potential therapeutic biomarkers that were tentatively identified as amino acids, acylcarnitines, and lysoPCs. Levels of three of these (octanoylcarnitine, lysoPC(16:1), and decanoylcarnitine) were closely correlated with treatment effect. Moreover, variation of these three potential biomarkers was investigated over the treatment course. The results suggest that these biomarkers may be useful in diagnosis, as well as in monitoring therapeutic responses and predicting outcomes of the ESCC.

TreeParser-aided Klee diagrams display taxonomic clusters in DNA barcode and nuclear gene datasets

Indicator vector analysis of a nucleotide sequence alignment generates a compact heat map, called a Klee diagram, with potential insight into clustering patterns in evolution. However, so far this approach has examined only mitochondrial cytochrome c oxidase I (COI) DNA barcode sequences. To further explore, we developed TreeParser, a freely-available web-based program that sorts a sequence alignment according to a phylogenetic tree generated from the dataset. We applied TreeParser to nuclear gene and COI barcode alignments from birds and butterflies. Distinct blocks in the resulting Klee diagrams corresponded to species and higher-level taxonomic divisions in both groups, and this enabled graphic comparison of phylogenetic information in nuclear and mitochondrial genes. Our results demonstrate TreeParser-aided Klee diagrams objectively display taxonomic clusters in nucleotide sequence alignments. This approach may help establish taxonomy in poorly studied groups and investigate higher-level clustering which appears widespread but not well understood.

A comparison of representations for discrete multi-criteria decision problems

Discrete multi-criteria decision problems with numerous Pareto-efficient solution candidates place a significant cognitive burden on the decision maker. An interactive, aspiration-based search process that iteratively progresses toward the most preferred solution can alleviate this task. In this paper, we study three ways of representing such problems in a DSS, and compare them in a laboratory experiment using subjective and objective measures of the decision process as well as solution quality and problem understanding. In addition to an immediate user evaluation, we performed a re-evaluation several weeks later. Furthermore, we consider several levels of problem complexity and user characteristics. Results indicate that different problem representations have a considerable influence on search behavior, although long-term consistency appears to remain unaffected. We also found interesting discrepancies between subjective evaluations and objective measures. Conclusions from our experiments can help designers of DSS for large multi-criteria decision problems to fit problem representations to the goals of their system and the specific task at hand.

Differentially expressed genes in adipocytokine signaling pathway of adipose tissue in pregnancy

Objective

To profile the differential gene expression of the KEGG Adipocytokine Signaling pathway in omental compared to subcutaneous tissue in normal pregnancy.

Study Design

Subjects included 14 nonobese, normal glucose tolerant, healthy pregnant women. Matched omental and subcutaneous tissue were obtained at elective cesarean delivery. Gene expression was evaluated using microarray and validated by RT-PCR. Differential gene expression was defined as ≥1.5 fold increase at p < 0.05.

Results

Six genes were significantly downregulated with two upregulated genes in omental tissue. Downregulation of Adiponectin and Insulin Receptor substrate, key genes mediating insulin sensitivity, were observed with borderline upregulation of GLUT-1. There were downregulations of CD36 and acyl-CoA Synthetase Long-chain Family Member 1which are genes involved in fatty acid uptake and activation. There was a novel expression of Carnitine palmitoyltransferase 1C.

Conclusion

Differential gene expression of Adipocytokine Signaling Pathway in omental relative to subcutaneous adipose tissue in normal pregnancy suggests a pattern of insulin resistance, hyperlipidemia, and inflammation.

Cluster analysis for identifying sub-groups and selecting potential discriminatory variables in human encephalitis

BACKGROUND

Encephalitis is an acute clinical syndrome of the central nervous system (CNS), often associated with fatal outcome or permanent damage, including cognitive and behavioural impairment, affective disorders and epileptic seizures. Infection of the central nervous system is considered to be a major cause of encephalitis and more than 100 different pathogens have been recognized as causative agents. However, a large proportion of cases have unknown disease etiology.

METHODS

We perform hierarchical cluster analysis on a multicenter England encephalitis data set with the aim of identifying sub-groups in human encephalitis. We use the simple matching similarity measure which is appropriate for binary data sets and performed variable selection using cluster heatmaps. We also use heatmaps to visually assess underlying patterns in the data, identify the main clinical and laboratory features and identify potential risk factors associated with encephalitis.

RESULTS

Our results identified fever, personality and behavioural change, headache and lethargy as the main characteristics of encephalitis. Diagnostic variables such as brain scan and measurements from cerebrospinal fluids are also identified as main indicators of encephalitis. Our analysis revealed six major clusters in the England encephalitis data set. However, marked within-cluster heterogeneity is observed in some of the big clusters indicating possible sub-groups. Overall, the results show that patients are clustered according to symptom and diagnostic variables rather than causal agents. Exposure variables such as recent infection, sick person contact and animal contact have been identified as potential risk factors.

CONCLUSIONS

It is in general assumed and is a common practice to group encephalitis cases according to disease etiology. However, our results indicate that patients are clustered with respect to mainly symptom and diagnostic variables rather than causal agents. These similarities and/or differences with respect to symptom and diagnostic measurements might be attributed to host factors. The idea that characteristics of the host may be more important than the pathogen is also consistent with the observation that for some causes, such as herpes simplex virus (HSV), encephalitis is a rare outcome of a common infection.

MulteeSum: a tool for comparative spatial and temporal gene expression data

Cells in an organism share the same genetic information in their DNA, but have very different forms and behavior because of the selective expression of subsets of their genes. The widely used approach of measuring gene expression over time from a tissue sample using techniques such as microarrays or sequencing do not provide information about the spatial position within the tissue where these genes are expressed. In contrast, we are working with biologists who use techniques that measure gene expression in every individual cell of entire fruitfly embryos over an hour of their development, and do so for multiple closely-related subspecies of Drosophila. These scientists are faced with the challenge of integrating temporal gene expression data with the spatial location of cells and, moreover, comparing this data across multiple related species. We have worked with these biologists over the past two years to develop MulteeSum, a visualization system that supports inspection and curation of data sets showing gene expression over time, in conjunction with the spatial location of the cells where the genes are expressed--it is the first tool to support comparisons across multiple such data sets. MulteeSum is part of a general and flexible framework we developed with our collaborators that is built around multiple summaries for each cell, allowing the biologists to explore the results of computations that mix spatial information, gene expression measurements over time, and data from multiple related species or organisms. We justify our design decisions based on specific descriptions of the analysis needs of our collaborators, and provide anecdotal evidence of the efficacy of MulteeSum through a series of case studies.

NeatMap--non-clustering heat map alternatives in R

Background

The clustered heat map is the most popular means of visualizing genomic data. It compactly displays a large amount of data in an intuitive format that facilitates the detection of hidden structures and relations in the data. However, it is hampered by its use of cluster analysis which does not always respect the intrinsic relations in the data, often requiring non-standardized reordering of rows/columns to be performed post-clustering. This sometimes leads to uninformative and/or misleading conclusions. Often it is more informative to use dimension-reduction algorithms (such as Principal Component Analysis and Multi-Dimensional Scaling) which respect the topology inherent in the data. Yet, despite their proven utility in the analysis of biological data, they are not as widely used. This is at least partially due to the lack of user-friendly visualization methods with the visceral impact of the heat map.

Results

NeatMap is an R package designed to meet this need. NeatMap offers a variety of novel plots (in 2 and 3 dimensions) to be used in conjunction with these dimension-reduction techniques. Like the heat map, but unlike traditional displays of such results, it allows the entire dataset to be displayed while visualizing relations between elements. It also allows superimposition of cluster analysis results for mutual validation. NeatMap is shown to be more informative than the traditional heat map with the help of two well-known microarray datasets.

Conclusions

NeatMap thus preserves many of the strengths of the clustered heat map while addressing some of its deficiencies. It is hoped that NeatMap will spur the adoption of non-clustering dimension-reduction algorithms.

Pharmacogenomics: a systems approach

Pharmacogenetics and pharmacogenomics involve the study of the role of inheritance in individual variation in drug response, a phenotype that varies from potentially life-threatening adverse drug reactions to equally serious lack of therapeutic efficacy. Pharmacogenetics-pharmacogenomics represents a major component of the movement to 'individualized medicine'. Pharmacogenetic studies originally focused on monogenic traits, often involving genetic variation in drug metabolism. However, contemporary studies increasingly involve entire 'pathways' that include both pharmacokinetics (PKs)--factors that influence the concentration of a drug reaching its target(s)--and pharmacodynamics (PDs), factors associated with the drug target(s), as well as genome-wide approaches. The convergence of advances in pharmacogenetics with rapid developments in human genomics has resulted in the evolution of pharmacogenetics into pharmacogenomics. At the same time, studies of drug response are expanding beyond genomics to encompass pharmacotranscriptomics and pharmacometabolomics to become a systems-based discipline. This discipline is also increasingly moving across the 'translational interface' into the clinic and is being incorporated into the drug development process and governmental regulation of that process. The article will provide an overview of the development of pharmacogenetics-pharmacogenomics, the scientific advances that have contributed to the continuing evolution of this discipline, the incorporation of transcriptomic and metabolomic data into attempts to understand and predict variation in drug response phenotypes as well as challenges associated with the 'translation' of this important aspect of biomedical science into the clinic.

dictyExpress: a Dictyostelium discoideum gene expression database with an explorative data analysis web-based interface

Background

Bioinformatics often leverages on recent advancements in computer science to support biologists in their scientific discovery process. Such efforts include the development of easy-to-use web interfaces to biomedical databases. Recent advancements in interactive web technologies require us to rethink the standard submit-and-wait paradigm, and craft bioinformatics web applications that share analytical and interactive power with their desktop relatives, while retaining simplicity and availability.

Results

We have developed dictyExpress, a web application that features a graphical, highly interactive explorative interface to our database that consists of more than 1000 Dictyostelium discoideum gene expression experiments. In dictyExpress, the user can select experiments and genes, perform gene clustering, view gene expression profiles across time, view gene co-expression networks, perform analyses of Gene Ontology term enrichment, and simultaneously display expression profiles for a selected gene in various experiments. Most importantly, these tasks are achieved through web applications whose components are seamlessly interlinked and immediately respond to events triggered by the user, thus providing a powerful explorative data analysis environment.

Conclusion

dictyExpress is a precursor for a new generation of web-based bioinformatics applications with simple but powerful interactive interfaces that resemble that of the modern desktop. While dictyExpress serves mainly the Dictyostelium research community, it is relatively easy to adapt it to other datasets. We propose that the design ideas behind dictyExpress will influence the development of similar applications for other model organisms.

Microarray analyses of gene expression during the Tetrahymena thermophila life cycle

BACKGROUND

The model eukaryote, Tetrahymena thermophila, is the first ciliated protozoan whose genome has been sequenced, enabling genome-wide analysis of gene expression.

METHODOLOGY/PRINCIPAL FINDINGS

A genome-wide microarray platform containing the predicted coding sequences (putative genes) for T. thermophila is described, validated and used to study gene expression during the three major stages of the organism's life cycle: growth, starvation and conjugation.

CONCLUSIONS/SIGNIFICANCE

Of the approximately 27,000 predicted open reading frames, transcripts homologous to only approximately 5900 are not detectable in any of these life cycle stages, indicating that this single-celled organism does indeed contain a large number of functional genes. Transcripts from over 5000 predicted genes are expressed at levels >5x corrected background and 95 genes are expressed at >250x corrected background in all stages. Transcripts homologous to 91 predicted genes are specifically expressed and 155 more are highly up-regulated in growing cells, while 90 are specifically expressed and 616 are up-regulated during starvation. Strikingly, transcripts homologous to 1068 predicted genes are specifically expressed and 1753 are significantly up-regulated during conjugation. The patterns of gene expression during conjugation correlate well with the developmental stages of meiosis, nuclear differentiation and DNA elimination. The relationship between gene expression and chromosome fragmentation is analyzed. Genes encoding proteins known to interact or to function in complexes show similar expression patterns, indicating that co-ordinate expression with putative genes of known function can identify genes with related functions. New candidate genes associated with the RNAi-like process of DNA elimination and with meiosis are identified and the late stages of conjugation are shown to be characterized by specific expression of an unexpectedly large and diverse number of genes not involved in nuclear functions.

Pharmacogenomics: candidate gene identification, functional validation and mechanisms

Pharmacogenetics is the study of the role of inheritance in variation in drug response phenotypes. Those phenotypes can range from life-threatening adverse drugs reactions at one end of the spectrum to equally serious lack of therapeutic efficacy at the other. Over the past half century, pharmacogenetics has--like all of medical genetics--evolved from a discipline with a focus on monogenetic traits to become pharmacogenomics, with a genome-wide perspective. This article will briefly review recent examples of the application of genome-wide techniques to clinical pharmacogenomic studies and to pharmacogenomic model systems that vary from cell line-based model systems to yeast gene deletion libraries. Functional validation of candidate genes and the use of genome-wide techniques to gain mechanistic insights will be emphasized for the establishment of biological plausibility and as essential follow-up steps after the identification of candidate genes.