Data clustering in life sciences

Kingfisher: An open-sourced web-based platform for the analysis of hydrogen exchange mass spectrometry data

Hydrogen-deuterium exchange mass spectrometry (HDX-MS) is now a critical tool in molecular biology and structural proteomics. It is routinely used to probe protein and conformational dynamics through a well-established experiment where amide hydrogens exchange with deuterium atoms in a buffer containing D2O. Although there have been numerous advances in the field, data analysis still poses challenges mainly due to the need for manual curation of the data and the lack of standardized statistics and accessible software. In response, we developed Kingfisher, an open-source, user-friendly, web-based solution that facilitates downstream analysis using well-established statistics and provides advanced high-resolution representations of the HDX results. Kingfisher is able to read data directly as exported from common software packages and usually takes less than a minute to run the analysis, without the need to download the raw code or install any software. We foresee Kingfisher as a valuable tool for both newcomers and experts in the field of Hydrogen Exchange Mass Spectrometry. Kingfisher is available to all users as an interactive web application at https://kingfisher.wustl.edu/.

Assessment of fluency dynamics in climbing

The aim of this study was to investigate the hold-by-hold climbing fluency dynamics by using an instrumented holds system that measured the contact time on each hold. Forty-four competitive climbers have been analysed in a regional lead climbing competition during a route composed of 41 instrumented holds on 11 m high artificial climbing wall and with a grade of difficulty 6b on the French scale (IRCRA reported scale: 13). After removing 10 climbers who fell before the top of the route, the 34 remaining climbers who completed the route were clustered according to their total contact time on each hold. The hierarchical cluster analysis distinguished four profiles of climbing fluency dynamics, on the basis of six 'crux' points, showing that the fastest climbers at the crux points were those with the shortest climbing time. This new instrumented-holds system appeared very innovative as it provides an instantaneous feedback to coaches regarding inter-limbs fluency and subsequent motor organisations.

Practical and comparative application of efficient data reduction - Multivariate curve resolution

The term 'Big Data' has recently attracted much attention in science. Working with big data sets can be both challenging and rewarding. The complexity and big data sets make the analysis difficult to deal with, and the increasing volume of data sets requires the development of new practical methods for their handling. In this contribution, we explored the efficient data reduction-multivariate curve resolution (EDR-MCR) strategy based on the convex hull theory for quantitative and qualitative analysis of large chemical data sets. For the quantitative example, the potential of the EDR-MCR method for selecting a representative calibration set was investigated, and the results were compared with the widely used Kennard-Stone (KS) algorithm. The EDR-MCR strategy strongly limits the number of calibration samples with a high potency of prediction performance. The priority of EDR-MCR over KS is its ability to find informative variables and eliminate redundant features. Moreover, the EDR-MCR strategy was also applied for the qualitative analysis of a large-scale metabolomic data set. The comparable analysis results of EDR-MCR with the region of interest (ROI) method confirmed the ability of this method for quantitative analysis of big mass spectrophotometer data sets.

A comparison of central-tendency and interconnectivity approaches to clustering multivariate data with irregular structure

Questions

Most clustering methods assume data are structured as discrete hyperspheroidal clusters to be evaluated by measures of central tendency. If vegetation data do not conform to this model, then vegetation data may be clustered incorrectly. What are the implications for cluster stability and evaluation if clusters are of irregular shape or density?

Location

Southeast Australia.

Methods

We define misplacement as the placement of a sample in a cluster other than (distinct from) its nearest neighbor and hypothesize that optimizing homogeneity incurs the cost of higher rates of misplacement. Chameleon is a graph-theoretic algorithm that emphasizes interconnectivity and thus is sensitive to the shape and distribution of clusters. We contrasted its solutions with those of traditional nonhierarchical and hierarchical (agglomerative and divisive) approaches.

Results

Chameleon-derived solutions had lower rates of misplacement and only marginally higher heterogeneity than those of k-means in the range of 15-60 clusters, but their metrics converged with larger numbers of clusters. Solutions derived by agglomerative clustering had the best metrics (and divisive clustering the worst) but both produced inferior high-level solutions to those of Chameleon by merging distantly-related clusters.

Conclusions

Graph-theoretic algorithms, such as Chameleon, have an advantage over traditional algorithms when data exhibit discontinuities and variable structure, typically producing more stable solutions (due to lower rates of misplacement) but scoring lower on traditional metrics of central tendency. Advantages are less obvious in the partitioning of data from continuous gradients; however, graph-based partitioning protocols facilitate the hierarchical integration of solutions.

Arm - Leg coordination profiling during the dolphin kick and the arm pull-out in elite breaststrokers

In breaststroke races, the dolphin kick could finish before, at the same time, or during the arm pull-out, but it is unclear how swimmers perform this technique. The aim of this study was to investigate whether swimmers glide between the dolphin kick and arm pull-out, favour continuity or even overlap those two phases, as it would impact the active underwater sequence. Fourteen international and national male swimmers performed 100-m breaststroke with all-out effort in a pre-calibrated 25 m swimming pool. A multi-camera system tracked the head of the swimmers. Key points of the active underwater sequence were obtained from notational analysis. A hierarchical cluster analysis identified three coordination profiles. All swimmers started their dolphin kick before the arm pull-out. However, one swimmer started the arm pull-out before the end of the dolphin kick, seven swimmers started the arm pull-out after the end of the dolphin kick, and four swimmers synchronised the beginning of the arm pull-out and the end of the dolphin kick, while two other swimmers mixed two coordination profiles among the start and the three turns. Those different profiles allow achieving similar performance outcome, suggesting individual training regarding the underwater phase.

A Shotgun Proteomic Platform for a Global Mapping of Lymphoblastoid Cells to Gain Insight into Nasu-Hakola Disease

Nasu-Hakola Disease (NHD) is a recessively inherited systemic leukodystrophy disorder characterized by a combination of frontotemporal presenile dementia and lytic bone lesions. NHD is known to be genetically related to a structural defect of TREM2 and DAP12, two genes that encode for different subunits of the membrane receptor signaling complex expressed by microglia and osteoclast cells. Because of its rarity, molecular or proteomic studies on this disorder are absent or scarce, only case reports based on neuropsychological and genetic tests being reported. In light of this, the aim of this paper is to provide evidence on the potential of a label-free proteomic platform based on the Multidimensional Protein Identification Technology (MudPIT), combined with in-house software and on-line bioinformatics tools, to characterize the protein expression trends and the most involved pathways in NHD. The application of this approach on the Lymphoblastoid cells from a family composed of individuals affected by NHD, healthy carriers and control subjects allowed for the identification of about 3000 distinct proteins within the three analyzed groups, among which proteins anomalous to each category were identified. Of note, several differentially expressed proteins were associated with neurodegenerative processes. Moreover, the protein networks highlighted some molecular pathways that may be involved in the onset or progression of this rare frontotemporal disorder. Therefore, this fully automated MudPIT platform which allowed, for the first time, the generation of the whole protein profile of Lymphoblastoid cells from Nasu-Hakola subjects, could be a valid approach for the investigation of similar neurodegenerative diseases.

Distinct changes of brain cortical thickness relate to post-treatment outcomes in children with epilepsy

PURPOSE

In the current study, we examined the potential of neuroanatomic measures to cluster patients into different subgroups and established their clinical relevance to post-treatment outcomes.

METHODS

We included seventy-two children with epilepsy (aged 14-195 months) who were treated with anti-seizure medication alone and 39 healthy participants (aged 36-60 months). High-resolution T1-weighted imaging was performed for all participants, and brain cortical thickness measurements were obtained for 68 cortical regions for each of them. Amongst the patients, data-driven hierarchical cluster analysis was performed using the selected cortical thickness measures as features. The average thickness measures in each of the 68 brain regions were then compared between patient subgroups and healthy controls.

RESULTS

Two distinct patient subgroups were identified but were not related to the clinical types. Patients within subgroup 1 (n = 56) had a significantly higher rate of recurrent seizure than those in subgroup 2 (n = 16) (41.1% vs. 14.3%, p<0.05), while the follow-up time or medication did not differ between them. This finding was further confirmed by a recent follow-up through phone calls. The demographic variables, rate of electroencephalogram abnormalities, or sleep problems did not significantly differ between patient subgroups. Compared with healthy controls, patients in subgroup 1 showed significantly increased cortical thickness in the neocortex, whereas patients in subgroup 2 only showed regional cortical thinning in the right superior temporal gyrus.

CONCLUSION

These findings suggest the potential existence of distinct subgroups of children with epilepsy that were especially relevant to the differential patterns of post-treatment outcomes, with regional cortical thinning in the temporal regions relative to controls predicting lower risk of recurrent seizure.

Shotgun Proteomics of Isolated Urinary Extracellular Vesicles for Investigating Respiratory Impedance in Healthy Preschoolers

Urine proteomic applications in children suggested their potential in discriminating between healthy subjects from those with respiratory diseases. The aim of the current study was to combine protein fractionation, by urinary extracellular vesicle isolation, and proteomics analysis in order to establish whether different patterns of respiratory impedance in healthy preschoolers can be characterized from a protein fingerprint. Twenty-one 3-5-yr-old healthy children, representative of 66 recruited subjects, were selected: 12 late preterm (LP) and 9 full-term (T) born. Children underwent measurement of respiratory impedance through Forced Oscillation Technique (FOT) and no significant differences between LP and T were found. Unbiased clustering, based on proteomic signatures, stratified three groups of children (A, B, C) with significantly different patterns of respiratory impedance, which was slightly worse in group A than in groups B and C. Six proteins (Tripeptidyl peptidase I (TPP1), Cubilin (CUBN), SerpinA4, SerpinF1, Thy-1 membrane glycoprotein (THY1) and Angiopoietin-related protein 2 (ANGPTL2)) were identified in order to type the membership of subjects to the three groups. The differential levels of the six proteins in groups A, B and C suggest that proteomic-based profiles of urinary fractionated exosomes could represent a link between respiratory impedance and underlying biological profiles in healthy preschool children.

Understanding nanoscale structural distortions in Pb(Zr0.2Ti0.8)O3 by utilizing X-ray nanodiffraction and clustering algorithm analysis

Hard X-ray nanodiffraction provides a unique nondestructive technique to quantify local strain and structural inhomogeneities at nanometer length scales. However, sample mosaicity and phase separation can result in a complex diffraction pattern that can make it challenging to quantify nanoscale structural distortions. In this work, a k-means clustering algorithm was utilized to identify local maxima of intensity by partitioning diffraction data in a three-dimensional feature space of detector coordinates and intensity. This technique has been applied to X-ray nanodiffraction measurements of a patterned ferroelectric PbZr_0.2Ti_0.8O₃ sample. The analysis reveals the presence of two phases in the sample with different lattice parameters. A highly heterogeneous distribution of lattice parameters with a variation of 0.02 Å was also observed within one ferroelectric domain. This approach provides a nanoscale survey of subtle structural distortions as well as phase separation in ferroelectric domains in a patterned sample.

GrpClassifierEC: a novel classification approach based on the ensemble clustering space

Background

Advances in molecular biology have resulted in big and complicated data sets, therefore a clustering approach that able to capture the actual structure and the hidden patterns of the data is required. Moreover, the geometric space may not reflects the actual similarity between the different objects. As a result, in this research we use clustering-based space that convert the geometric space of the molecular to a categorical space based on clustering results. Then we use this space for developing a new classification algorithm.

Results

In this study, we propose a new classification method named GrpClassifierEC that replaces the given data space with categorical space based on ensemble clustering (EC). The EC space is defined by tracking the membership of the points over multiple runs of clustering algorithms. Different points that were included in the same clusters will be represented as a single point. Our algorithm classifies all these points as a single class. The similarity between two objects is defined as the number of times that these objects were not belong to the same cluster. In order to evaluate our suggested method, we compare its results to the k nearest neighbors, Decision tree and Random forest classification algorithms on several benchmark datasets. The results confirm that the suggested new algorithm GrpClassifierEC outperforms the other algorithms.

Conclusions

Our algorithm can be integrated with many other algorithms. In this research, we use only the k-means clustering algorithm with different k values. In future research, we propose several directions: (1) checking the effect of the clustering algorithm to build an ensemble clustering space. (2) Finding poor clustering results based on the training data, (3) reducing the volume of the data by combining similar points based on the EC.

Availability and implementation

The KNIME workflow, implementing GrpClassifierEC, is available at https://malikyousef.com

Retinal Nerve Fiber Layer Features Identified by Unsupervised Machine Learning on Optical Coherence Tomography Scans Predict Glaucoma Progression

Purpose

To apply computational techniques to wide-angle swept-source optical coherence tomography (SS-OCT) images to identify novel, glaucoma-related structural features and improve detection of glaucoma and prediction of future glaucomatous progression.

Methods

Wide-angle SS-OCT, OCT circumpapillary retinal nerve fiber layer (cpRNFL) circle scans spectral-domain (SD)-OCT, standard automated perimetry (SAP), and frequency doubling technology (FDT) visual field tests were completed every 3 months for 2 years from a cohort of 28 healthy participants (56 eyes) and 93 glaucoma participants (179 eyes). RNFL thickness maps were extracted from segmented SS-OCT images and an unsupervised machine learning approach based on principal component analysis (PCA) was used to identify novel structural features. Area under the receiver operating characteristic curve (AUC) was used to assess diagnostic accuracy of RNFL PCA for detecting glaucoma and progression compared to SAP, FDT, and cpRNFL measures.

Results

The RNFL PCA features were significantly associated with mean deviation (MD) in both SAP (R² = 0.49, P < 0.0001) and FDT visual field testing (R² = 0.48, P < 0.0001), and with mean circumpapillary RNFL thickness (cpRNFLt) from SD-OCT (R² = 0.58, P < 0.0001). The identified features outperformed each of these measures in detecting glaucoma with an AUC of 0.95 for RNFL PCA compared to an 0.90 for mean cpRNFLt (P = 0.09), 0.86 for SAP MD (P = 0.034), and 0.83 for FDT MD (P = 0.021). Accuracy in predicting progression was also significantly higher for RNFL PCA compared to SAP MD, FDT MD, and mean cpRNFLt (P = 0.046, P = 0.007, and P = 0.044, respectively).

Conclusions

A computational approach can identify structural features that improve glaucoma detection and progression prediction.

ACES: a machine learning toolbox for clustering analysis and visualization

Background

Studies that aim at explaining phenotypes or disease susceptibility by genetic or epigenetic variants often rely on clustering methods to stratify individuals or samples. While statistical associations may point at increased risk for certain parts of the population, the ultimate goal is to make precise predictions for each individual. This necessitates tools that allow for the rapid inspection of each data point, in particular to find explanations for outliers.

Results

ACES is an integrative cluster- and phenotype-browser, which implements standard clustering methods, as well as multiple visualization methods in which all sample information can be displayed quickly. In addition, ACES can automatically mine a list of phenotypes for cluster enrichment, whereby the number of clusters and their boundaries are estimated by a novel method. For visual data browsing, ACES provides a 2D or 3D PCA or Heat Map view. ACES is implemented in Java, with a focus on a user-friendly, interactive, graphical interface.

Conclusions

ACES has been proven an invaluable tool for analyzing large, pre-filtered DNA methylation data sets and RNA-Sequencing data, due to its ease to link molecular markers to complex phenotypes. The source code is available from https://github.com/GrabherrGroup/ACES.

Proteome Investigation of Rat Lungs subjected to Ex Vivo Perfusion (EVLP)

Ex vivo lung perfusion (EVLP) is an emerging procedure that allows organ preservation, assessment and reconditioning, increasing the number of marginal donor lungs for transplantation. However, physiological and airflow measurements are unable to unveil the molecular mechanisms responsible of EVLP beneficial effects on lung graft and monitor the proper course of the treatment. Thus, it is urgent to find specific biomarkers that possess these requirements but also accurate and reliable techniques that identify them. The purpose of this study is to give an overview on the potentiality of shotgun proteomic platforms in characterizing the status and the evolution of metabolic pathways during EVLP in order to find new potential EVLP-related biomarkers. A nanoLC-MS/MS system was applied to the proteome analysis of lung tissues from an optimized rat model in three experimental groups: native, pre- and post-EVLP. Technical and biological repeatability were evaluated and, together with clustering analysis, underlined the good quality of data produced. In-house software and bioinformatics tools allowed the label-free extraction of differentially expressed proteins among the three examined conditions and the network visualization of the pathways mainly involved. These promising findings encourage further proteomic investigations of the molecular mechanisms behind EVLP procedure.

Adaptation of the Food Choice Questionnaire: the case of Hungary

PURPOSE

The purpose of this paper is to examine the applicability of the original 36-item Food Choice Questionnaire (FCQ) model developed by Steptoe et al. (1995) in Hungary.

DESIGN/METHODOLOGY/APPROACH

The national representative questionnaire involved 1,050 individuals in Hungary in 2015. Several multivariable statistical techniques were applied for the analysis of the data: confirmatory factor analysis, principal component analysis, and cluster and Log-linear analysis.

FINDINGS

The results indicate that the original nine-factor model is only partially applicable to Hungary. This study successfully managed to distinguish the following factors: health and natural content, mood, preparation convenience, price and purchase convenience, sensory appeal, familiarity, and ethical concern. The FCQ scales proved to be suitable for the description of clusters based on specific food choices and demographic characteristics. By using the factors, the following five clusters were identified: modern food enthusiast, tradition-oriented, optimizer, easy-choice and un-concerned, all of which could be addressed by public health policy with individually tailored messages.

ORIGINALITY/VALUE

The Hungarian testing process of the FCQ model contributes to an examination of its usability and provides the possibility of fitting the model to different cultures.

Combination of Metabolomic and Proteomic Analysis Revealed Different Features among Lactobacillus delbrueckii Subspecies bulgaricus and lactis Strains While In Vivo Testing in the Model Organism Caenorhabditis elegans Highlighted Probiotic Properties

Lactobacillus delbrueckii represents a technologically relevant member of lactic acid bacteria, since the two subspecies bulgaricus and lactis are widely associated with fermented dairy products. In the present work, we report the characterization of two commercial strains belonging to L. delbrueckii subspecies bulgaricus, lactis and a novel strain previously isolated from a traditional fermented fresh cheese. A phenomic approach was performed by combining metabolomic and proteomic analysis of the three strains, which were subsequently supplemented as food source to the model organism Caenorhabditis elegans, with the final aim to evaluate their possible probiotic effects. Restriction analysis of 16S ribosomal DNA revealed that the novel foodborne strain belonged to L. delbrueckii subspecies lactis. Proteomic and metabolomic approaches showed differences in folate, aminoacid and sugar metabolic pathways among the three strains. Moreover, evaluation of C. elegans lifespan, larval development, brood size, and bacterial colonization capacity demonstrated that L. delbrueckii subsp. bulgaricus diet exerted beneficial effects on nematodes. On the other hand, both L. delbrueckii subsp. lactis strains affected lifespan and larval development. We have characterized three strains belonging to L. delbrueckii subspecies bulgaricus and lactis highlighting their divergent origin. In particular, the two closely related isolates L. delbrueckii subspecies lactis display different galactose metabolic capabilities. Moreover, the L. delbrueckii subspecies bulgaricus strain demonstrated potential probiotic features. Combination of omic platforms coupled with in vivo screening in the simple model organism C. elegans is a powerful tool to characterize industrially relevant bacterial isolates.

Microarray-Based Gene Expression Analysis for Veterinary Pathologists: A Review

High-throughput, genome-wide transcriptome analysis is now commonly used in all fields of life science research and is on the cusp of medical and veterinary diagnostic application. Transcriptomic methods such as microarrays and next-generation sequencing generate enormous amounts of data. The pathogenetic expertise acquired from understanding of general pathology provides veterinary pathologists with a profound background, which is essential in translating transcriptomic data into meaningful biological knowledge, thereby leading to a better understanding of underlying disease mechanisms. The scientific literature concerning high-throughput data-mining techniques usually addresses mathematicians or computer scientists as the target audience. In contrast, the present review provides the reader with a clear and systematic basis from a veterinary pathologist's perspective. Therefore, the aims are (1) to introduce the reader to the necessary methodological background; (2) to introduce the sequential steps commonly performed in a microarray analysis including quality control, annotation, normalization, selection of differentially expressed genes, clustering, gene ontology and pathway analysis, analysis of manually selected genes, and biomarker discovery; and (3) to provide references to publically available and user-friendly software suites. In summary, the data analysis methods presented within this review will enable veterinary pathologists to analyze high-throughput transcriptome data obtained from their own experiments, supplemental data that accompany scientific publications, or public repositories in order to obtain a more in-depth insight into underlying disease mechanisms.

Role of route previewing strategies on climbing fluency and exploratory movements

This study examined the role of route previewing strategies on climbing fluency and on exploratory movements of the limbs, in order to understand whether previewing helps people to perceive and to realize affordances. Eight inexperienced and ten experienced climbers previewed a 10 m high route of 5b difficulty on French scale, then climbed it with a top-rope as fluently as possible. Gaze behavior was collected from an eye tracking system during the preview and allowed us to determine the number of times they scanned the route, and which of four route previewing strategies (fragmentary, ascending, zigzagging, and sequence-of-blocks) they used. Five inertial measurement units (IMU) (3D accelerometer, 3D gyroscope, 3D magnetometer) were attached to the hip, both feet, and forearms to analyze the vertical acceleration and direction of each limb and hip during the ascent. We were able to detect movement and immobility phases of each IMU using segmentation and classification processes. Depending on whether the limbs and/or hip were moving, five states of behavior were detected: immobility, postural regulation, hold exploration, hold change, and hold traction. Using cluster analysis we identified four clusters of gaze behavior during route previewing depending on route preview duration, number of scan paths, fixations duration, ascending, zigzagging, and sequence-of-blocks strategies. The number of scan paths was positively correlated with relative duration of exploration and negatively correlated with relative duration of hold changes during the ascent. Additionally, a high relative duration of sequence-of-blocks strategy and zigzagging strategy were associated with a high relative duration of immobility during the ascent. Route previewing might help to pick up functional information about reachable, graspable, and usable holds, in order to chain movements together and to find the route. In other words, route previewing might contribute to perceiving and realizing nested affordances.

Galectin-3: an early predictive biomarker of modulation of airway remodeling in patients with severe asthma treated with omalizumab for 36 months

Background

Bronchial asthma is a heterogeneous disease characterized by three cardinal features: chronic inflammation, variable airflow obstruction, and airway hyperresponsiveness. Asthma has traditionally been defined using nonspecific clinical and physiologic variables that encompass multiple phenotypes and are treated with nonspecific anti-inflammatory therapies. Based on the modulation of airway remodeling after 12 months of anti-immunoglobulin E (IgE) treatment, we identified two phenotypes (omalizumab responder, OR; and non-omalizumab responder, NOR) and performed morphometric analysis of bronchial biopsy specimens. We also found that these two phenotypes were correlated with the presence/absence of galectin-3 (Gal-3) at baseline (i.e., before treatment). The aims of the present study were to investigate the histological and molecular effects of long-term treatment (36 months) with anti-IgE and to analyze the behavior of OR and NOR patients.

Methods

All patients were treated with the monoclonal antibody anti-IgE omalizumab for 36 months. The bronchial biopsy specimens were evaluated using morphometric, eosinophilic, and proteomic analysis (MudPIT). New data were compared with previous data, and unsupervised cluster analysis of protein profiles was performed.

Results

After 36 months of treatment with omalizumab, reduction of reticular basement membrane (RBM) thickness was confirmed in OR patients (Gal-3-positive at baseline); similarly, the protein profiles (over 500 proteins identified) revealed that, in the OR group, levels of proteins specifically related to fibrosis and inflammation (e.g., smooth muscle and extracellular matrix proteins (including periostin), Gal-3, and keratins decreased by between 5- and 50-fold. Eosinophil levels were consistent with molecular data and decreased by about tenfold less in ORs and increased by twofold to tenfold more in NORs. This tendency was confirmed (p < 0.05) based on both fold change and DAVE algorithms, thus indicating a clear response to anti-IgE treatment in Gal-3-positive patients.

Conclusions

Our results showed that omalizumab can be considered a disease-modifying treatment in OR. The proteomic signatures confirmed the presence of Gal-3 at baseline to be a biomarker of long-term reduction in bronchial RBM thickness, eosinophilic inflammation, and muscular and fibrotic components in omalizumab-treated patients with severe asthma. Our findings suggest a possible relationship between Gal-3 positivity and improved pulmonary function.

Electronic supplementary material

The online version of this article (doi:10.1186/s13601-017-0143-1) contains supplementary material, which is available to authorized users.

Molybdenum and iron mutually impact their homeostasis in cucumber (Cucumis sativus) plants

Molybdenum (Mo) and iron (Fe) are essential micronutrients required for crucial enzyme activities in plant metabolism. Here we investigated the existence of a mutual control of Mo and Fe homeostasis in cucumber (Cucumis sativus). Plants were grown under single or combined Mo and Fe starvation. Physiological parameters were measured, the ionomes of tissues and the ionomes and proteomes of root mitochondria were profiled, and the activities of molybdo-enzymes and the synthesis of molybdenum cofactor (Moco) were evaluated. Fe and Mo were found to affect each other's total uptake and distribution within tissues and at the mitochondrial level, with Fe nutritional status dominating over Mo homeostasis and affecting Mo availability for molybdo-enzymes in the form of Moco. Fe starvation triggered Moco biosynthesis and affected the molybdo-enzymes, with its main impact on nitrate reductase and xanthine dehydrogenase, both being involved in nitrogen assimilation and mobilization, and on the mitochondrial amidoxime reducing component. These results, together with the identification of > 100 proteins differentially expressed in root mitochondria, highlight the central role of mitochondria in the coordination of Fe and Mo homeostasis and allow us to propose the first model of the molecular interactions connecting Mo and Fe homeostasis.

Azithromycin Attenuates Pseudomonas-Induced Lung Inflammation by Targeting Bacterial Proteins Secreted in the Cultured Medium

Background

Pseudomonas aeruginosa airway infections are a major cause of morbidity and mortality in patients with cystic fibrosis (CF). Azithromycin improves the related clinical outcomes, but its mechanisms of action remain poorly understood. We tested the hypothesis that azithromycin downregulates P. aeruginosa-induced pro-inflammatory responses by modifying release of bacterial proteins.

Methods

We monitored inflammatory markers in lungs of CF mutant mice and their littermate controls in response to conditioned media (CM) collected from the reference P. aeruginosa PAO1 strain cultured in the presence or in the absence of azithromycin. A mass spectrometry-based proteomic approach was applied to examine whether the macrolide elicits a differential release of bacterial proteins.

Results

CM collected from azithromycin-untreated PAO1 cultures induced powerful pro-inflammatory neutrophil-dominated responses. Azithromycin attenuated the responses, mainly of macrophage chemoattractant protein-1, tumor necrosis factor-α, and interferon-γ, in CF but not in wild-type mice. Proteomic analysis showed that azithromycin upregulated an array of bacterial proteins including those associated with regulation of immune functions and with repair and resolution of inflammatory responses like the chaperone DnaK and the S-adenosylmethionine synthase, while it downregulated the extracellular heme acquisition protein HasA and the catalytic enzyme lysylendopeptidase.

Conclusion

Supernatants collected from cultures of the bacterial strain PAO1 represent a novel experimental model to trigger in vivo lung inflammatory responses that should be closer to those obtained with live bacteria, but without bacterial infection. Combined with a bactericidal effect, complex regulation of bacterial innate immune and metabolic factors released in the cultured medium by the action of the macrolide can contribute to its anti-inflammatory effects.

Neurobiological degeneracy: supporting stability, flexibility and pluripotentiality in complex motor skill

This paper investigated neurobiological degeneracy of the motor system that emerged as a function of levels of environmental constraint. Fourteen participants performed a breaststroke-swimming task that required them to develop a specific biomechanically expert pattern and in turn provide the basis for a suitable task vehicle to study the functional role of movement variability. Inter-limb coordination was defined based on the computation of continuous relative phase between elbow and knee oscillators. Unsupervised cluster analysis on arm-leg coordination revealed the existence of different patterns of coordination when participants achieved the same task goal under different levels of environmental constraints (i.e. different amounts of forward resistances). In addition, clusters differed in terms of higher order derivatives (e.g., joint angular velocity, joint amplitude), suggesting an effective role for degeneracy in learning by allowing the exploration of the key relationships between motor organization and interacting constraints. There is evidence to suggest that neurobiological degeneracy supports the potential for motor re-organization to enhance motor learning.

Applying parallel factor analysis and Tucker-3 methods on sensory and instrumental data to establish preference maps: case study on sweet corn varieties

BACKGROUND

Traditional internal and external preference mapping methods are based on principal component analysis (PCA). However, parallel factor analysis (PARAFAC) and Tucker-3 methods could be a better choice. To evaluate the methods, preference maps of sweet corn varieties will be introduced.

RESULTS

A preference map of eight sweet corn varieties was established using PARAFAC and Tucker-3 methods. Instrumental data were also integrated into the maps. The triplot created by the PARAFAC model explains better how odour is separated from texture or appearance, and how some varieties are separated from others.

CONCLUSION

Internal and external preference maps were created using parallel factor analysis (PARAFAC) and Tucker-3 models employing both sensory (trained panel and consumers) and instrumental parameters simultaneously. Triplots of the applied three-way models have a competitive advantage compared to the traditional biplots of the PCA-based external preference maps. The solution of PARAFAC and Tucker-3 is very similar regarding the interpretation of the first and third factors. The main difference is due to the second factor as it differentiated the attributes better. Consumers who prefer 'super sweet' varieties (they place great emphasis especially on taste) are much younger and have significantly higher incomes, and buy sweet corn products rarely (once a month). Consumers who consume sweet corn products mainly because of their texture and appearance are significantly older and include a higher ratio of men.

Proteomics of bronchial biopsies: galectin-3 as a predictive biomarker of airway remodelling modulation in omalizumab-treated severe asthma patients

Asthma is a chronic inflammatory disease. Reticular basement membrane (RBM) thickening is considered feature of airway remodelling (AR) particularly in severe asthma (SA). Omalizumab, mAb to IgE is effective in SA and can modulate AR. Herein we describe protein profiles of bronchial biopsies to detect biomarkers of anti-IgE effects on AR and to explain potential mechanisms/pathways. We defined the bronchial biopsy protein profiles, before and after treatment. Unsupervised clustering of baseline proteomes resulted in very good agreement with the morphometric analysis of AR. Protein profiles of omalizumab responders (ORs) were significantly different from those of non-omalizumab responders (NORs). The major differences between ORs and NORs lied to smooth muscle and extra cellular matrix proteins. Notably, an IgE-binding protein (galectin-3) was reliable, stable and predictive biomarker of AR modulation. Omalizumab down-regulated bronchial smooth muscle proteins in SA. These findings suggest that omalizumab may exert disease-modifying effects on remodelling components.

Discrimination of plant-parasitic nematodes from complex soil communities using ecometagenetics

Many plant pathogens are microscopic, cryptic, and difficult to diagnose. The new approach of ecometagenetics, involving ultrasequencing, bioinformatics, and biostatistics, has the potential to improve diagnoses of plant pathogens such as nematodes from the complex mixtures found in many agricultural and biosecurity situations. We tested this approach on a gradient of complexity ranging from a few individuals from a few species of known nematode pathogens in a relatively defined substrate to a complex and poorly known suite of nematode pathogens in a complex forest soil, including its associated biota of unknown protists, fungi, and other microscopic eukaryotes. We added three known but contrasting species (Pratylenchus neglectus, the closely related P. thornei, and Heterodera avenae) to half the set of substrates, leaving the other half without them. We then tested whether all nematode pathogens-known and unknown, indigenous, and experimentally added-were detected consistently present or absent. We always detected the Pratylenchus spp. correctly and with the number of sequence reads proportional to the numbers added. However, a single cyst of H. avenae was only identified approximately half the time it was present. Other plant-parasitic nematodes and nematodes from other trophic groups were detected well but other eukaryotes were detected less consistently. DNA sampling errors or informatic errors or both were involved in misidentification of H. avenae; however, the proportions of each varied in the different bioinformatic pipelines and with different parameters used. To a large extent, false-positive and false-negative errors were complementary: pipelines and parameters with the highest false-positive rates had the lowest false-negative rates and vice versa. Sources of error identified included assumptions in the bioinformatic pipelines, slight differences in primer regions, the number of sequence reads regarded as the minimum threshold for inclusion in analysis, and inaccessible DNA in resistant life stages. Identification of the sources of error allows us to suggest ways to improve identification using ecometagenetics.

Learning from data: recognizing glaucomatous defect patterns and detecting progression from visual field measurements

A hierarchical approach to learn from visual field data was adopted to identify glaucomatous visual field defect patterns and to detect glaucomatous progression. The analysis pipeline included three stages, namely, clustering, glaucoma boundary limit detection, and glaucoma progression detection testing. First, cross-sectional visual field tests collected from each subject were clustered using a mixture of Gaussians and model parameters were estimated using expectation maximization. The visual field clusters were further estimated to recognize glaucomatous visual field defect patterns by decomposing each cluster into several axes. The glaucoma visual field defect patterns along each axis then were identified. To derive a definition of progression, the longitudinal visual fields of stable glaucoma eyes on the abnormal cluster axes were projected and the slope was approximated using linear regression (LR) to determine the confidence limit of each axis. For glaucoma progression detection, the longitudinal visual fields of each eye on the abnormal cluster axes were projected and the slope was approximated by LR. Progression was assigned if the progression rate was greater than the boundary limit of the stable eyes; otherwise, stability was assumed. The proposed method was compared to a recently developed progression detection method and to clinically available glaucoma progression detection software. The clinical accuracy of the proposed pipeline was as good as or better than the currently available methods.

Recognizing patterns of visual field loss using unsupervised machine learning

Glaucoma is a potentially blinding optic neuropathy that results in a decrease in visual sensitivity. Visual field abnormalities (decreased visual sensitivity on psychophysical tests) are the primary means of glaucoma diagnosis. One form of visual field testing is Frequency Doubling Technology (FDT) that tests sensitivity at 52 points within the visual field. Like other psychophysical tests used in clinical practice, FDT results yield specific patterns of defect indicative of the disease. We used Gaussian Mixture Model with Expectation Maximization (GEM), (EM is used to estimate the model parameters) to automatically separate FDT data into clusters of normal and abnormal eyes. Principal component analysis (PCA) was used to decompose each cluster into different axes (patterns). FDT measurements were obtained from 1,190 eyes with normal FDT results and 786 eyes with abnormal (i.e., glaucomatous) FDT results, recruited from a university-based, longitudinal, multi-center, clinical study on glaucoma. The GEM input was the 52-point FDT threshold sensitivities for all eyes. The optimal GEM model separated the FDT fields into 3 clusters. Cluster 1 contained 94% normal fields (94% specificity) and clusters 2 and 3 combined, contained 77% abnormal fields (77% sensitivity). For clusters 1, 2 and 3 the optimal number of PCA-identified axes were 2, 2 and 5, respectively. GEM with PCA successfully separated FDT fields from healthy and glaucoma eyes and identified familiar glaucomatous patterns of loss.

Development of a HIPAA-compliant environment for translational research data and analytics

High-performance computing centers (HPC) traditionally have far less restrictive privacy management policies than those encountered in healthcare. We show how an HPC can be re-engineered to accommodate clinical data while retaining its utility in computationally intensive tasks such as data mining, machine learning, and statistics. We also discuss deploying protected virtual machines. A critical planning step was to engage the university's information security operations and the information security and privacy office. Access to the environment requires a double authentication mechanism. The first level of authentication requires access to the university's virtual private network and the second requires that the users be listed in the HPC network information service directory. The physical hardware resides in a data center with controlled room access. All employees of the HPC and its users take the university's local Health Insurance Portability and Accountability Act training series. In the first 3 years, researcher count has increased from 6 to 58.

Cavities tell more than sequences: exploring functional relationships of proteases via binding pockets

Computational approaches play an increasingly important role for the analysis and prediction of selectivity profiles. As most of the successfully administered small molecule drugs bind in depressions on the surface of proteins, physicochemical properties of the pocket-exposed amino acids play a central role in ligand recognition during the binding event. Cavbase is an approach to describe binding sites in terms of the exposed physicochemical properties and to compare them independent of their sequence and fold homology. Classification of proteins by means of their binding-site properties is a promising approach to obtain information necessary for selectivity modeling. For this purpose, the workflow clusterScore has been developed to explore the important parameters of a clustering procedure, which will allow an accurate classification of proteins. It has been successfully applied on two diverse and challenging data sets. The predicted number of clusters, as suggested by clusterScore and the subsequent clustering of proteins are in agreement with the EC and Merops classifications. Furthermore, putative cross-reactivity mapped between calpain-1 and cysteine cathepsins on structural level has so far only been described based on ligand data. In a benchmark study using ligand topology, binding site, and sequence information of eleven serine proteases, the emerging clusters indicate a pronounced correlation between the cavity and ligand data. These results emphasize the importance of binding-site information which should be considered for ligand design during lead optimization cycles. The program clusterScore is freely available and can be downloaded from our Web site www.agklebe.de.

Document Sublanguage Clustering to Detect Medical Specialty in Cross-institutional Clinical Texts

This paper reports on a set of studies designed to identify sublanguages in documents for domain-specific processing across institutions. Psychological evidence indicates that humans use context-specific linguistic information when they read. Natural Language Processing (NLP) pipelines are successful within specific domains (i.e., contexts). To limit the number of domain-specific NLP systems, a natural focus would be on sublanguages. Sublanguages are identified by shared lexical and semantic features.[1] Patterson and Hurdle[2] developed a sublanguage identification system that functioned well for 12 clinical specialties at the University of Utah. The current work compares sublanguages across institutions. Using a clinical NLP pipeline augmented by a new document corpus from the University of Pittsburg (UPitt), new documents were assigned to clusters based on the minimum cosine-distance to a Utah cluster centroid. The UPitt documents were divided into a nine-group specialty corpus. Across institutions, five of the specialty groups fell within the expected clusters. We find that clustering encounters difficulty due to documents with mixed sublanguages; naming convention differences across institutions; and document types used across specialties. The findings indicate that clinical specialty sublanguages can be identified across institutions.

Non invasive clinical assessment of trunk deformities associated with scoliosis

Besides the spinal deformity, scoliosis modifies notably the general appearance of the trunk resulting in trunk rotation, imbalance and asymmetries which constitutes patients' major concern. Existing classifications of scoliosis, based on the type of spinal curve as depicted on radiographs, are currently used to guide treatment strategies. Unfortunately, even though a perfect correction of the spinal curve is achieved, some trunk deformities remain, making patients dissatisfied with the treatment received. The purpose of this study is to identify possible shape patterns of trunk surface deformity associated with scoliosis. First, trunk surface is represented by a multivariate functional trunk shape descriptor based on 3D clinical measurements computed on cross sections of the trunk. Then, the classical formulation of hierarchical clustering is adapted to the case of multivariate functional data and applied to a set of 236 trunk surface 3D reconstructions. The highest internal validity is obtained when considering 11 clusters that explain up to 65% of the variance in our dataset. Our clustering result shows a concordance with the radiographic classification of spinal curves in 68% of the cases. As opposed to radiographic evaluation, the trunk descriptor is three-dimensional and its functional nature offers a compact and elegant description of not only the type, but also the severity and extent of the trunk surface deformity along the trunk length. In future work, new management strategies based on the resulting trunk shape patterns could be thought of in order to improve the esthetic outcome after treatment, and thus patients satisfaction.

Mapping L1 ligase ribozyme conformational switch

L1 ligase (L1L) molecular switch is an in vitro optimized synthetic allosteric ribozyme that catalyzes the regioselective formation of a 5'-to-3' phosphodiester bond, a reaction for which there is no known naturally occurring RNA catalyst. L1L serves as a proof of principle that RNA can catalyze a critical reaction for prebiotic RNA self-replication according to the RNA world hypothesis. L1L crystal structure captures two distinct conformations that differ by a reorientation of one of the stems by around 80Å and are presumed to correspond to the active and inactive state, respectively. It is of great interest to understand the nature of these two states in solution and the pathway for their interconversion. In this study, we use explicit solvent molecular simulation together with a novel enhanced sampling method that utilizes concepts from network theory to map out the conformational transition between active and inactive states of L1L. We find that the overall switching mechanism can be described as a three-state/two-step process. The first step involves a large-amplitude swing that reorients stem C. The second step involves the allosteric activation of the catalytic site through distant contacts with stem C. Using a conformational space network representation of the L1L switch transition, it is shown that the connection between the three states follows different topographical patterns: the stem C swing step passes through a narrow region of the conformational space network, whereas the allosteric activation step covers a much wider region and a more diverse set of pathways through the network.

Fast structure similarity searches among protein models: efficient clustering of protein fragments

BACKGROUND

For many predictive applications a large number of models is generated and later clustered in subsets based on structure similarity. In most clustering algorithms an all-vs-all root mean square deviation (RMSD) comparison is performed. Most of the time is typically spent on comparison of non-similar structures. For sets with more than, say, 10,000 models this procedure is very time-consuming and alternative faster algorithms, restricting comparisons only to most similar structures would be useful.

RESULTS

We exploit the inverse triangle inequality on the RMSD between two structures given the RMSDs with a third structure. The lower bound on RMSD may be used, when restricting the search of similarity to a reasonably low RMSD threshold value, to speed up similarity searches significantly. Tests are performed on large sets of decoys which are widely used as test cases for predictive methods, with a speed-up of up to 100 times with respect to all-vs-all comparison depending on the set and parameters used. Sample applications are shown.

CONCLUSIONS

The algorithm presented here allows fast comparison of large data sets of structures with limited memory requirements. As an example of application we present clustering of more than 100000 fragments of length 5 from the top500H dataset into few hundred representative fragments. A more realistic scenario is provided by the search of similarity within the very large decoy sets used for the tests. Other applications regard filtering nearly-indentical conformation in selected CASP9 datasets and clustering molecular dynamics snapshots.

AVAILABILITY

A linux executable and a Perl script with examples are given in the supplementary material (Additional file 1). The source code is available upon request from the authors.

MudPIT analysis of released proteins in Pseudomonas aeruginosa laboratory and clinical strains in relation to pro-inflammatory effects

Pseudomonas aeruginosa (Pa) is the most common virulent pathogen contributing to the pathogenesis of cystic fibrosis (CF). During bacterial lung colonization, the products of its metabolism are released in the extracellular space contributing to the pathogenic events associated with its presence. To gain insights on the mechanisms involved in the Pa pathogenesis we focused our attention on proteins released by Pa using a MudPIT approach combined with cell biology assays. Conditioned medium (CM) collected under aerobic and microaerobic conditions from Pa clinical strains (in early and late colonization), unlike the laboratory strain, induced expression of IL-8 mRNA in CF airway epithelial cells. We have identified proteins released by clinically relevant Pa strains, focusing on the pro-inflammatory effects as metalloproteases (MMPs). In fact, their expression pattern was associated with the highest pro-inflammatory activity measured in the early clinically isolated strain. The relation was further supported by the result of the analysis of a larger and independent set of Pa isolates derived from sporadically and chronically infected CF patients: 76% of sporadic samples expressed protease activity (n = 44), while only 27% scored positive in the chronically infected individuals (n = 38, p < 0.0001, Fisher's exact test). Finally, looking for a possible mechanism of action of bacterial MMPs, we found that CM from early clinical isolates can cleave CXCR1 on the surface of human neutrophils, suggesting a potential role for the bacterially released MMPs in the protection of the pathogen from the host's response.

Modeling metabolic networks for mammalian cell systems: general considerations, modeling strategies, and available tools

Over the past decades, the availability of large amounts of information regarding cellular processes and reaction rates, along with increasing knowledge about the complex mechanisms involved in these processes, has changed the way we approach the understanding of cellular processes. We can no longer rely only on our intuition for interpreting experimental data and evaluating new hypotheses, as the information to analyze is becoming increasingly complex. The paradigm for the analysis of cellular systems has shifted from a focus on individual processes to comprehensive global mathematical descriptions that consider the interactions of metabolic, genomic, and signaling networks. Analysis and simulations are used to test our knowledge by refuting or validating new hypotheses regarding a complex system, which can result in predictive capabilities that lead to better experimental design. Different types of models can be used for this purpose, depending on the type and amount of information available for the specific system. Stoichiometric models are based on the metabolic structure of the system and allow explorations of steady state distributions in the network. Detailed kinetic models provide a description of the dynamics of the system, they involve a large number of reactions with varied kinetic characteristics and require a large number of parameters. Models based on statistical information provide a description of the system without information regarding structure and interactions of the networks involved. The development of detailed models for mammalian cell metabolism has only recently started to grow more strongly, due to the intrinsic complexities of mammalian systems, and the limited availability of experimental information and adequate modeling tools. In this work we review the strategies, tools, current advances, and recent models of mammalian cells, focusing mainly on metabolism, but discussing the methodology applied to other types of networks as well.

Phenotype mining for functional genomics and gene discovery

In gene prediction, studying phenotypes is highly valuable for reducing the number of locus candidates in association studies and to aid disease gene candidate prioritization. This is due to the intrinsic nature of phenotypes to visibly reflect genetic activity, making them potentially one of the most useful data types for functional studies. However, systematic use of these data has begun only recently. 'Comparative phenomics' is the analysis of genotype-phenotype associations across species and experimental methods. This is an emerging research field of utmost importance for gene discovery and gene function annotation. In this chapter, we review the use of phenotype data in the biomedical field. We will give an overview of phenotype resources, focusing on PhenomicDB--a cross-species genotype-phenotype database--which is the largest available collection of phenotype descriptions across species and experimental methods. We report on its latest extension by which genotype-phenotype relationships can be viewed as graphical representations of similar phenotypes clustered together ('phenoclusters'), supplemented with information from protein-protein interactions and Gene Ontology terms. We show that such 'phenoclusters' represent a novel approach to group genes functionally and to predict novel gene functions with high precision. We explain how these data and methods can be used to supplement the results of gene discovery approaches. The aim of this chapter is to assist researchers interested in understanding how phenotype data can be used effectively in the gene discovery field.

Cluster analysis of self-monitoring blood glucose assessments in clinical islet cell transplantation for type 1 diabetes

OBJECTIVE

Cluster analysis was performed on the results of self-monitoring of blood glucose (SMBG) to discriminate islet graft function after islet cell transplantation (ICT) in patients with type 1 diabetes.

RESEARCH DESIGN AND METHODS

Eleven islet recipients were included in this study. The patients visited our clinic monthly after ICT and provided blood samples for fasting C-peptide (n = 270), which were used to evaluate islet graft function. They also provided their SMBG data through an automatic data collection system. The SMBG data for 3 days immediately before each clinic visit were evaluated using the following assessments: M value, mean amplitude of glycemic excursions, J index, index of glycemic control, average daily risk range, and glycemic risk assessment diabetes equation. The cluster analysis was performed for both SMBG assessments and samples. Multivariate logistic regression analysis was used to evaluate the clusters of SMBG for assessing islet graft function.

RESULTS

Analysis for SMBG assessments revealed five types of clusters, which showed similar patterns according to functional or dysfunctional islet graft phase. Two clusters, the euglycemia cluster (P < 0.001) and the hypoglycemia cluster (P = 0.001), were significant factors in the logistic model for islet graft function. The SMBG clusters had significant correlations with clinical graft indexes (P < 0.001).

CONCLUSIONS

Cluster analysis of SMBG data as part of an automated data quality system could allow discrimination of islet graft dysfunction after ICT. This approach should be considered for islet recipients.

Cluster analysis of arterial involvement in Takayasu arteritis reveals symmetric extension of the lesions in paired arterial beds

OBJECTIVE

The determinants of vessel targeting are largely unknown in vasculitides. This study was undertaken to identify patterns of vascular involvement in Takayasu arteritis (TA), using objective classification of vascular beds. We postulated that cluster analysis could unveil preferential associations between vascular beds commonly affected by TA.

METHODS

Peripheral vascular Doppler, computed tomography angiography, and angio-magnetic resonance imaging data from 82 patients with TA (according to the American College of Rheumatology criteria) were studied between January 1995 and May 2006. Cross-relationships of involvement between 24 main arteries were assessed using the phi correlation coefficient. Identification of patterns of vascular involvement was performed using agglomerative hierarchical cluster analysis.

RESULTS

Data were obtained from 82 patients (68 women [82.9%] and 14 men [17.1%]). The median duration of followup was 5.1 years (range 1 month to 30 years). For 16 (80%) of 20 paired arteries, the highest correlation of involvement was observed with the contralateral artery. Conversely, disease extension was contiguous in the aorta. Cluster analysis further confirmed that all paired arterial beds, except for the internal and external carotid arteries, clustered with their contralateral counterpart and that the aortic arch, the descending thoracic aorta, and the abdominal aorta clustered together.

CONCLUSION

Our findings reveal that TA lesions mostly develop in a symmetric manner in paired vascular territories and that disease extension is contiguous in the aorta. This may prove useful for improving the radiologic followup of patients with TA and for providing a pattern for further investigations focusing on the mechanisms of vessel specificity in vasculitides.

GOModeler--a tool for hypothesis-testing of functional genomics datasets

Background

Functional genomics technologies that measure genome expression at a global scale are accelerating biological knowledge discovery. Generating these high throughput datasets is relatively easy compared to the downstream functional modelling necessary for elucidating the molecular mechanisms that govern the biology under investigation. A number of publicly available ‘discovery-based’ computational tools use the computationally amenable Gene Ontology (GO) for hypothesis generation. However, there are few tools that support hypothesis-based testing using the GO and none that support testing with user defined hypothesis terms.

Here, we present GOModeler, a tool that enables researchers to conduct hypothesis-based testing of high throughput datasets using the GO. GOModeler summarizes the overall effect of a user defined gene/protein differential expression dataset on specific GO hypothesis terms selected by the user to describe a biological experiment. The design of the tool allows the user to complement the functional information in the GO with his/her domain specific expertise for comprehensive hypothesis testing.

Results

GOModeler tests the relevance of the hypothesis terms chosen by the user for the input gene dataset by providing the individual effects of the genes on the hypothesis terms and the overall effect of the entire dataset on each of the hypothesis terms. It matches the GO identifiers (ids) of the genes with the GO ids of the hypothesis terms and parses the names of those ids that match to assign effects. We demonstrate the capabilities of GOModeler with a dataset of nine differentially expressed cytokine genes and compare the results to those obtained through manual analysis of the dataset by an immunologist. The direction of overall effects on all hypothesis terms except one was consistent with the results obtained by manual analysis. The tool’s editing capability enables the user to augment the information extracted. GOModeler is available as a part of the AgBase tool suite (http://www.agbase.msstate.edu).

Conclusions

GOModeler allows hypothesis driven analysis of high throughput datasets using the GO. Using this tool, researchers can quickly evaluate the overall effect of quantitative expression changes of gene set on specific biological processes of interest. The results are provided in both tabular and graphical formats.

Potentially amyloidogenic conformational intermediates populate the unfolding landscape of transthyretin: insights from molecular dynamics simulations

Protein aggregation into insoluble fibrillar structures known as amyloid characterizes several neurodegenerative diseases, including Alzheimer's, Huntington's and Creutzfeldt-Jakob. Transthyretin (TTR), a homotetrameric plasma protein, is known to be the causative agent of amyloid pathologies such as FAP (familial amyloid polyneuropathy), FAC (familial amyloid cardiomiopathy) and SSA (senile systemic amyloidosis). It is generally accepted that TTR tetramer dissociation and monomer partial unfolding precedes amyloid fibril formation. To explore the TTR unfolding landscape and to identify potential intermediate conformations with high tendency for amyloid formation, we have performed molecular dynamics unfolding simulations of WT-TTR and L55P-TTR, a highly amyloidogenic TTR variant. Our simulations in explicit water allow the identification of events that clearly discriminate the unfolding behavior of WT and L55P-TTR. Analysis of the simulation trajectories show that (i) the L55P monomers unfold earlier and to a larger extent than the WT; (ii) the single alpha-helix in the TTR monomer completely unfolds in most of the L55P simulations while remain folded in WT simulations; (iii) L55P forms, early in the simulations, aggregation-prone conformations characterized by full displacement of strands C and D from the main beta-sandwich core of the monomer; (iv) L55P shows, late in the simulations, severe loss of the H-bond network and consequent destabilization of the CBEF beta-sheet of the beta-sandwich; (v) WT forms aggregation-compatible conformations only late in the simulations and upon extensive unfolding of the monomer. These results clearly show that, in comparison with WT, L55P-TTR does present a much higher probability of forming transient conformations compatible with aggregation and amyloid formation.

Integrating phenotype and gene expression data for predicting gene function

Background

This paper presents a framework for integrating disparate data sets to predict gene function. The algorithm constructs a graph, called an integrated similarity graph, by computing similarities based upon both gene expression and textual phenotype data. This integrated graph is then used to make predictions about whether individual genes should be assigned a particular annotation from the Gene Ontology.

Results

A combined graph was generated from publicly-available gene expression data and phenotypic information from Saccharomyces cerevisiae. This graph was used to assign annotations to genes, as were graphs constructed from gene expression data and textual phenotype information alone. While the F-measure appeared similar for all three methods, annotations based upon the integrated similarity graph exhibited a better overall precision than gene expression or phenotype information alone can generate. The integrated approach was also able to assign almost as many annotations as the gene expression method alone, and generated significantly more total and correct assignments than the phenotype information could provide.

Conclusion

These results suggest that augmenting standard gene expression data sets with publicly-available textual phenotype data can help generate more precise functional annotation predictions while mitigating the weaknesses of a standard textual phenotype approach.

jClust: a clustering and visualization toolbox

jClust is a user-friendly application which provides access to a set of widely used clustering and clique finding algorithms. The toolbox allows a range of filtering procedures to be applied and is combined with an advanced implementation of the Medusa interactive visualization module. These implemented algorithms are k-Means, Affinity propagation, Bron–Kerbosch, MULIC, Restricted neighborhood search cluster algorithm, Markov clustering and Spectral clustering, while the supported filtering procedures are haircut, outside–inside, best neighbors and density control operations. The combination of a simple input file format, a set of clustering and filtering algorithms linked together with the visualization tool provides a powerful tool for data analysis and information extraction.

Availability:http://jclust.embl.de/

Contact:pavlopou@embl.de; rschneid@embl.de; skossida@bioacademy.gr

Supplementary information:Supplementary data are available at Bioinformatics online.

Landscape character assessment using region growing techniques in geographical information systems

Landscape character can be defined as the presence, variety and arrangement of landscape features, which give a landscape a specific identity and make it stand out from surrounding landscapes. Landscape character contributes to the esthetical and perceptional value of an area, which is important for the development of non-production functions in the countryside as demanded by society. In this paper we present a new methodology for landscape character assessment using the pattern of landscape features as stored in a GIS to delineate, characterize and evaluate landscapes using a region growing algorithm. We have applied this methodology in a case study area in the north of The Netherlands and compared the results with a series of expert classifications of the study area. The results of the region growing algorithms were good and interpretable in relation to the underlying data. The resemblance between the expert classification and the classification based on the region growing results varied between 34% and 100% for the different landscape types. The differences between the two data sets can be explained in terms of input data and knowledge about the study area. The classification of the region growing algorithm was more consistent than the expert classification throughout the study area. The presented methodology for landscape character assessment is proposed as support for spatial planning processes and policy development for landscape conservation by providing a quantitative tool to analyze landscape patterns, to discriminate between the various landscapes in a study area and by elucidating features that are important for the identity of a region.

Gene module level analysis: identification to networks and dynamics

Nature exhibits modular design in biological systems. Gene module level analysis is based on this module concept, aiming to understand biological network design and systems behavior in disease and development by emphasizing on modules of genes rather than individual genes. Module level analysis has been extensively applied in genome wide level analysis, exploring the organization of biological systems from identifying modules to reconstructing module networks and analyzing module dynamics. Such module level perspective provides a high level representation of the regulatory scenario and design of biological systems, promising to revolutionize our view of systems biology, genetic engineering as well as disease mechanisms and molecular medicine.

Conservation of small RNA pathways in platypus

Small RNA pathways play evolutionarily conserved roles in gene regulation and defense from parasitic nucleic acids. The character and expression patterns of small RNAs show conservation throughout animal lineages, but specific animal clades also show variations on these recurring themes, including species-specific small RNAs. The monotremes, with only platypus and four species of echidna as extant members, represent the basal branch of the mammalian lineage. Here, we examine the small RNA pathways of monotremes by deep sequencing of six platypus and echidna tissues. We find that highly conserved microRNA species display their signature tissue-specific expression patterns. In addition, we find a large rapidly evolving cluster of microRNAs on platypus chromosome X1, which is unique to monotremes. Platypus and echidna testes contain a robust Piwi-interacting (piRNA) system, which appears to be participating in ongoing transposon defense.

Mining phenotypes for gene function prediction

Background

Health and disease of organisms are reflected in their phenotypes. Often, a genetic component to a disease is discovered only after clearly defining its phenotype. In the past years, many technologies to systematically generate phenotypes in a high-throughput manner, such as RNA interference or gene knock-out, have been developed and used to decipher functions for genes. However, there have been relatively few efforts to make use of phenotype data beyond the single genotype-phenotype relationships.

Results

We present results on a study where we use a large set of phenotype data – in textual form – to predict gene annotation. To this end, we use text clustering to group genes based on their phenotype descriptions. We show that these clusters correlate well with several indicators for biological coherence in gene groups, such as functional annotations from the Gene Ontology (GO) and protein-protein interactions. We exploit these clusters for predicting gene function by carrying over annotations from well-annotated genes to other, less-characterized genes in the same cluster. For a subset of groups selected by applying objective criteria, we can predict GO-term annotations from the biological process sub-ontology with up to 72.6% precision and 16.7% recall, as evaluated by cross-validation. We manually verified some of these clusters and found them to exhibit high biological coherence, e.g. a group containing all available antennal Drosophila odorant receptors despite inconsistent GO-annotations.

Conclusion

The intrinsic nature of phenotypes to visibly reflect genetic activity underlines their usefulness in inferring new gene functions. Thus, systematically analyzing these data on a large scale offers many possibilities for inferring functional annotation of genes. We show that text clustering can play an important role in this process.

Evaluation of high-throughput functional categorization of human disease genes

BACKGROUND

Biological data that are well-organized by an ontology, such as Gene Ontology, enables high-throughput availability of the semantic web. It can also be used to facilitate high throughput classification of biomedical information. However, to our knowledge, no evaluation has been published on automating classifications of human diseases genes using Gene Ontology. In this study, we evaluate automated classifications of well-defined human disease genes using their Gene Ontology annotations and compared them to a gold standard. This gold standard was independently conceived by Valle's research group, and contains 923 human disease genes organized in 14 categories of protein function.

RESULTS

Two automated methods were applied to investigate the classification of human disease genes into independently pre-defined categories of protein function. One method used the structure of Gene Ontology by pre-selecting 74 Gene Ontology terms assigned to 11 protein function categories. The second method was based on the similarity of human disease genes clustered according to the information-theoretic distance of their Gene Ontology annotations. Compared to the categorization of human disease genes found in the gold standard, our automated methods can achieve an overall 56% and 47% precision with 62% and 71% recall respectively. However, approximately 15% of the studied human disease genes remain without GO annotations.

CONCLUSION

Automated methods can recapitulate a significant portion of classification of the human disease genes. The method using information-theoretic distance performs slightly better on the precision with some loss in recall. For some protein function categories, such as 'hormone' and 'transcription factor', the automated methods perform particularly well, achieving precision and recall levels above 75%. In summary, this study demonstrates that for semantic webs, methods to automatically classify or analyze a majority of human disease genes require significant progress in both the Gene Ontology annotations and particularly in the utilization of these annotations.

Assessment of autofluorescence to detect the remineralization capabilities of sodium fluoride, monofluorophosphate and non-fluoride dentifrices. A single-blind cluster randomized trial

The purpose of the study was to determine if longitudinal measurements of enamel autofluorescence (quantitative light-induced fluorescence, QLF) could detect differences in remineralization of early enamel caries on buccal surfaces of anterior teeth following supervised daily brushing with either sodium fluoride (NaF; 1,450 ppm F), sodium monofluorophosphate (MFP; 1,450 ppm F) dentifrices or a herbal, non-fluoride placebo dentifrice. The study was a pragmatic cluster-randomized controlled trial with schools as the unit of randomization. Twenty-one schools in Chengdu, China, comprised the clusters; 296 children with at least 1 visible white-spot lesion were examined using QLF at baseline and after 3 and 6 months. Each of the 21 clusters was randomly assigned 1 of the 3 dentifrices, and the children brushed under supervision once per day for 2 min. The primary outcome measure was deltaQ (product of fluorescence loss and area) at a 5% threshold after 6 months of product use. A multi-level model was fitted to the data at the site level, taking into account the hierarchical structure with baseline deltaQ, age and sex as covariates. After 3 months there was a significant difference between the MFP group and placebo (p = 0.02) and after 6 months between the NaF group (p = 0.002), MFP group (p < 0.001) and the placebo. QLF methodology could detect, within 3- and 6-month periods of supervised brushing, a difference in remineralization between fluoride-containing and non-fluoride-containing dentifrices. Typically lesions in all 3 treatment groups demonstrated improvement. Groups receiving fluoride experienced a more rapid and more substantial remineralization than those in the placebo group.