Pairwise Pearson Correlation Research Articles

Echocardiographic Parameters that Predict Outcome in Aortic Atresia Patients Undergoing Comprehensive Stage II Procedure

The hybrid procedure is one option for palliating patients with hypoplastic left heart syndrome. As experience increases with this palliation, the physiology and its influence on outcome can be better assessed. The goal of this study was to determine if echocardiographic parameters correlate with post-operative variables in patients with aortic atresia undergoing the comprehensive Stage II procedure. Retrospective chart review on all patients with aortic atresia, who underwent the comprehensive Stage II procedure from January 2002 to December 2008, was performed. Echocardiographic indices were evaluated and correlations were made with peri-operative and hospital variables. Pair-wise Pearson's correlation tests were used to analyze the associations between continuous measures. Thirty-four patients met inclusion criteria. Age at comprehensive Stage II procedure was 0.45 ± 0.13 years and body surface area was 0.31 ± 0.04 m(2). Right ventricle (RV) percentage change was 45 ± 10%, eccentricity index was 1.96 ± 0.45, estimated systemic cardiac output was 7.68 ± 2.56 L/min/m(2) and estimated effective systemic cardiac output was 5.15 ± 2.24 L/min/m(2). Retrograde patent ductus arteriosu (PDA) velocity time integral (VTI) correlated with log pre-bypass lactate and maximum lactate (r = 0.53, 0.44). PDA regurgitant fraction correlated with log post-bypass lactate, length of intubation, and urine output on day four (r = 0.39, 0.46, -0.37). RV percentage change correlated with log pre-bypass lactate, and urine output on days four and five (r =-0.38, 0.43, 0.54). No echocardiographic parameter predicted renal or liver insufficiency, dialysis, extracorporeal membrane oxygenation use, or hospital death. Retrograde PDA VTI and RV percent change correlated with some peri-operative variables though no echocardiographic parameter was associated with any major morbidities or mortality. Newer echocardiographic techniques may better predict comprehensive stage II outcomes.

Robust Methodology for Fractal Analysis of the Retinal Vasculature

We have developed a robust method to perform retinal vascular fractal analysis from digital retina images. The technique preprocesses the green channel retina images with Gabor wavelet transforms to enhance the retinal images. Fourier Fractal dimension is computed on these preprocessed images and does not require any segmentation of the vessels. This novel technique requires human input only at a single step; the allocation of the optic disk center. We have tested this technique on 380 retina images from healthy individuals aged 50+ years, randomly selected from the Blue Mountains Eye Study population. To assess its reliability in assessing retinal vascular fractals from different allocation of optic center, we performed pair-wise Pearson correlation between the fractal dimension estimates with 100 simulated region of interest for each of the 380 images. There was Gaussian distribution variation in the optic center allocation in each simulation. The resulting mean correlation coefficient (standard deviation) was 0.93 (0.005). The repeatability of this method was found to be better than the earlier box-counting method. Using this method to assess retinal vascular fractals, we have also confirmed a reduction in the retinal vasculature complexity with aging, consistent with observations from other human organ systems.

Structural validity and reliability of the integrated conflict and violence scale

The study validated structure stability, reliability and sub-scale distinctiveness of integrated conflict and violence scale (ICVS) and was cross-sectional; war-affected grade 5 school children participated. ICV internal factorial validity and reliability were evaluated; eigenvalue size and scree plots were used for factor selection. A variable retention factor load threshold of >0.30 was used: Cronbach's α tests confirmed reliability increments. Pair-wise Pearson correlation tests evaluated sub-scale distinctiveness. Gulu University granted ethical clearance. A total of 280 grade 5 children from 50 primary schools participated: 53% of them were males. Two factors accounted for 100% of variability in attitudes; 18 variables were retained. Expelled variables were: ‘If I catch some one stealing my sugar cane I will fight’ and ‘a bully should be forgiven’. Sub-scale internal consistency reliability coefficients were 0.73 and 0.65, respectively and distinctiveness correlation coefficient was −0.06. The ICVS was validated using standard criteria. Emerging two-factor scale has acceptable psychometric properties especially factorial structure, internal consistency and sub-scale distinctiveness.

Identification of p27/KIP1 expression level as a candidate biomarker of response to rapalogs therapy in human cancer

Rapamycin analogs temsirolimus and everolimus have been approved for the treatment of advanced renal cancer and are being tested in a wide spectrum of human tumors. However, objective response rates with rapalogs in clinical trials were modest and variable. Identification of biomarkers capable of predicting response to rapalogs is of increasing interest. We analyzed pairwise Pearson correlation coefficients (r) between rapalogs activity and gene expression profile for each NCI-60 cell line. p27 showed the highest positive correlation among 9,706 gene probes tested. At cellular levels, breast cancer MCF-7, T47D, and BT-474 cells, expressing high levels of p27, were sensitive to rapalogs, whereas the cells expressed low levels of p27, such as MDA-MB-231, MDA-MB-468, and MDA-MB-435 cells, exhibited resistance to rapalogs. Mechanistic study indicated that this correlation is likely determined by the basal level of p27 regardless of the phosphorylation or redistribution of p27 upon rapalogs treatment, which may provide a putative threshold to block G1/S transition. Consistently, down-regulation of p27 by siRNA conferred MCF-7 and BT-474 cells insensitive to rapalogs. Moreover, a significant positive correlation between p27 gene expression and rapamycin anti-tumor activity was also observed in mice bearing different human cancer cell xenografts. In conclusion, p27 expression level is positively correlated with the anticancer activity of rapalogs in vitro and in vivo. We propose p27 expression level may be also a candidate predictive biomarker for patient selection for rapalogs-based therapy, which requires clinical validation in a series of patients treated with rapalogs.

Dihydroartemisinin accelerates c-MYC oncoprotein degradation and induces apoptosis in c-MYC-overexpressing tumor cells

Artemisinin and its derivatives (ARTs) are effective antimalarial drugs and also possess profound anticancer activity. However, the mechanism accounted for its distinctive activity in tumor cells remains unelucidated. We computed Pair wise Pearson correlation coefficients to identify genes that show significant correlation with ARTs activity in NCI-55 cell lines using data obtained from studies with HG-U133A Affymetrix chip. We found c-myc is one of the genes that showed the highest positive correlation coefficients among the probe sets analyzed ( r = 0.585, P < 0.001). Dihydroartemisinin (DHA), the main active metabolite of ARTs, induced significant apoptosis in HL-60 and HCT116 cells that express high levels of c-MYC. Stable knockdown of c-myc abrogated DHA-induced apoptosis in HCT116 cells. Conversely, forced expression of c-myc in NIH3T3 cells sensitized these cells to DHA-induced apoptosis. Interestingly, DHA irreversibly down-regulated the protein level of c-MYC in DHA-sensitive HCT116 cells, which is consistent to persistent G1 phase arrest induced by DHA. Further studies demonstrated that DHA accelerated the degradation of c-MYC protein and this process was blocked by pretreatment with the proteasome inhibitor MG-132 or GSK 3β inhibitor LiCl in HCT116 cells. Taken together, ARTs might be useful in the treatment of c-MYC-overexpressing tumors. We also suggest that c-MYC may potentially be a biomarker candidate for prediction of the antitumor efficacies of ARTs.

ZAP-70 Expression Assessed by Immunohistochemistry Correlates with Time to First Treatment in Patients with Chronic Lymphocytic Leukemia.

Abstract 4686 BackgroundZAP-70 expression is a prognostic indicator in chronic lymphocytic leukemia (CLL). Initial studies measured intracellular ZAP-70 expression by flow cytometry and defined a cut-off of 20% as predictive of time from diagnosis to initial therapy and overall survival. However, due to technical challenges, standardization of methods for evaluation of ZAP-70 expression has been difficult. This has resulted in a limitation of applicability for a potentially useful clinical test. Immunohistochemistry (IHC) is easy to perform and interpret, and therefore may be a more reliable and accurate means for assessing ZAP-70 expression. We investigated the potential prognostic implications of ZAP-70 expression assessed by IHC. MethodsWe searched our CLL database for all patients seen at Weill Cornell Medical College between 2007 and 2009 with an identifiable date of diagnosis, date of first therapy, IgVH mutation status, and ZAP-70 status. All patients gave informed consent to participate in the research database. Paraffin-embedded cell blocks were prepared from formalin-fixed peripheral blood mononuclear cells and evaluated by IHC for PAX5, CD3, and ZAP-70. The entire cell clot was reviewed. If more than 20% of cells faintly expressed ZAP-70, the ZAP-70 status was considered to be positive. Patients had IgVH mutations status determined by a commercial laboratory with <98% sequence homology used as a cut-off. Time from diagnosis to first treatment (TT) was estimated by the Kaplan-Meier method; comparisons were performed by log-rank test. Pearson's pairwise correlation coefficient was used to evaluate the relationship between ZAP-70 and IgVH. ResultsOne-hundred thirty-seven patients with date of diagnosis and date of first treatment were identified in the database. Sixty-six patients have been treated to date. ZAP-70 status was available for 106 patients, 62 were positive, 44 were negative. In a subset of 20 patients, two blinded pathologists reviewed IHC and found perfect agreement on ZAP-70 status. IgVH mutation status was available for 86 patients, 34 were unmutated, and 52 were mutated. Eighty-two patients had both ZAP-70 status and IgVH mutation status available. There was good correlation between ZAP-70 and IgVH status (r=0.36, p=0.0008). Seventy-five percent of those with unmutated IgVH were ZAP-70 positive. Sixty-one percent of patients with mutated IgVH were ZAP-70 negative. However, only 57% of ZAP-70 positive patients were IgVH unmutated. The predicted median TT was 54 months for the whole cohort (range 0-376+ months, 95% CI 45-96 months). The median TT was 27 months (range 0-112 months) for ZAP-70-positive patients compared to 96 months (range 0-376+ months) for ZAP-70-negative patients (p=0.002). The median TT was 32 months (range 0-112 months) for IgVH unmutated patients compared to 88 months (range 0-376+ months) for IgVH mutated patients (p=0.01). ConclusionsZAP-70 as measured by IHC was feasible in a large cohort of patients and ZAP-70 status correlated with time to first treatment. These data suggest that ZAP-70 assessment by IHC may be a reliable prognostic tool in CLL and should be evaluated in larger, multicenter clinical studies. Disclosures:No relevant conflicts of interest to declare.

An integrative genomics approach for deciphering the complex interactions between ascorbate metabolism and fruit growth and composition in tomato

Very few reports have studied the interactions between ascorbate and fruit metabolism. In order to get insights into the complex relationships between ascorbate biosynthesis/recycling and other metabolic pathways in the fruit, we undertook a fruit systems biology approach. To this end, we have produced tomato transgenic lines altered in ascorbate content and redox ratio by RNAi-targeting several key enzymes involved in ascorbate biosynthesis (2 enzymes) and recycling (2 enzymes). In the VTC (ViTamin C) Fruit project, we then generated phenotypic and genomic (transcriptome, proteome, metabolome) data from wild type and mutant tomato fruit at two stages of fruit development, and developed or implemented statistical and bioinformatic tools as a web application (named VTC Tool box) necessary to store, analyse and integrate experimental data in tomato. By using Kohonen's self-organizing maps (SOMs) to cluster the biological data, pair-wise Pearson correlation analyses and simultaneous visualization of transcript/protein and metabolites (MapMan), this approach allowed us to uncover major relationships between ascorbate and other metabolic pathways. To cite this article: V. Garcia et al., C. R. Biologies 332 (2009).

STAR NET 2: a web-based tool for accelerating discovery of gene regulatory networks using microarray co-expression data

BackgroundAlthough expression microarrays have become a standard tool used by biologists, analysis of data produced by microarray experiments may still present challenges. Comparison of data from different platforms, organisms, and labs may involve complicated data processing, and inferring relationships between genes remains difficult.ResultsSTARNET 2 is a new web-based tool that allows post hoc visual analysis of correlations that are derived from expression microarray data. STARNET 2 facilitates user discovery of putative gene regulatory networks in a variety of species (human, rat, mouse, chicken, zebrafish, Drosophila, C. elegans, S. cerevisiae, Arabidopsis and rice) by graphing networks of genes that are closely co-expressed across a large heterogeneous set of preselected microarray experiments. For each of the represented organisms, raw microarray data were retrieved from NCBI's Gene Expression Omnibus for a selected Affymetrix platform. All pairwise Pearson correlation coefficients were computed for expression profiles measured on each platform, respectively. These precompiled results were stored in a MySQL database, and supplemented by additional data retrieved from NCBI. A web-based tool allows user-specified queries of the database, centered at a gene of interest. The result of a query includes graphs of correlation networks, graphs of known interactions involving genes and gene products that are present in the correlation networks, and initial statistical analyses. Two analyses may be performed in parallel to compare networks, which is facilitated by the new HEATSEEKER module.ConclusionSTARNET 2 is a useful tool for developing new hypotheses about regulatory relationships between genes and gene products, and has coverage for 10 species. Interpretation of the correlation networks is supported with a database of previously documented interactions, a test for enrichment of Gene Ontology terms, and heat maps of correlation distances that may be used to compare two networks. The list of genes in a STARNET network may be useful in developing a list of candidate genes to use for the inference of causal networks. The tool is freely available at , and does not require user registration.

DNA Methylation in Resistant Paediatric Acute Lymphoblastic Leukaemia.

Aberrant DNA methylation is a frequent phenomenon in paediatric acute lymphoblastic leukaemia (ALL). Increased methylation has been identified as an independent factor of poor prognosis in paediatric ALL patients. This study investigates the evolution of epigenetic changes in resistant and relapsed paediatric ALL to increase our understanding of the mechanisms of treatment failure in these patients. The primary objective is to establish global DNA methylation profile in ALL and whether changes in this profile are associated with chemo-resistance and relapse. We performed a genome-wide analysis of CpG island methylation in leukaemic blasts on a cohort of 22 paediatric ALL patients; 11 at presentation and 11 at relapse. Global methylation profiles were investigated using the recently developed CpG island array technology, which includes 12,192 sequenced CpG-island clones obtained from the Wellcome Trust Sanger Institute (Cambridge, UK). These clones have undergone comprehensive sequence analysis which validated the array content and confirmed that 80% of the clones have usable sequences that can be aligned to genomic DNA. The majority of CpG-island loci are localised to known distal promoter regions or close to a transcriptional start site. We have validated the application of this technology to leukaemia cells using established ALL and AML cell lines. Pairwise Pearson correlation coefficient, used to compare log2 ratios between triplicate experiments, revealed high correlation values between replicates (median=0.85), demonstrating significant reproducibility. Bisulphite sequencing was performed to confirm that this technology could reliably depict the actual methylation status of the CpG islands.We have evaluated CpG island methylation profiling in childhood ALL both at presentation and relapse. We report that ALL specific CpG island methylation patterns are emerging. We have identified a significant number of genes, involved in cell growth and differentiation and cell cycle regulation that are consistently differentially methylated between paediatric ALL patients and healthy controls. Expressly, a number of Hox genes and Hox-genes regulators, involved in cell differentiation are differentially methylated in leukaemic blasts compared to controls. Comparisons between methylation profiles at presentation and relapse demonstrate that overall genes that are methylated at presentation remain methylated in relapse; however there is an important subset of genes that become differentially methylated. At relapse, there is a significant increase in methylation levels of a set of pro-apoptotic and tumour suppressor genes, as well as several genes involved in drug metabolism. Interestingly, a group of genes involved in ErbB signalling pathway are differentially methylated in relapse, suggesting that epigenetic regulation of this pathway might be implicated in resistant disease. Our results suggest altered methylation status within a defined subset of genes may be implicated in relapse of ALL.

A common currency area for MENA countries? A VAR analysis of viability

PurposeThis paper aims to examine the feasibility of a Common Currency Area (CCA) among ten MENA (Middle East and North Africa) Countries. The ten sample countries constitute the six GCC Countries and the four Agadir nations.Design/methodology/approachMacroeconomic data for the 34‐year‐period 1970‐2003 is used. Feasibility is examined by analyzing the symmetry of response of countries within each group to a common external shock. The impulse response functions (IRF) from a Vector Autoregression Model is used. The strength of linkages within each economic bloc was examined using Pearson pairwise correlation and variance decomposition.FindingsAmong GCC countries, the results show the existence of strong linkages among the monetary variables, signifying strong monetary sector integration. Such integration however is lacking where the real sector is concerned. Despite the symmetry seen in the impulse response functions, variance decomposition showed the absence of any meaningful influence of countries on each other within the bloc. Amongst the Agadir nations, the results show no correlation in real output growth, some correlation among monetary variables but no symmetry whatsoever in response to external shocks. The variance decomposition too did not show mutual influence intra group.Practical implicationsThe lack of real sector integration will present a challenge to GCC's desired goal of a CCA by 2010. The Agadir nations appear to be simply a loosely knit economic grouping with little integration of any kind. Thus, hopes of a CCA among Agadir nations is far too premature.Originality/valueThe paper concludes that the GCC is, at present, a quasi‐monetary bloc with little real sector integration.

Quantitative quality-assessment techniques to compare fractionation and depletion methods in SELDI-TOF mass spectrometry experiments

Mass spectrometry (MS), such as the surface-enhanced laser desorption and ionization time-of-flight (SELDI-TOF) MS, provides a potentially promising proteomic technology for biomarker discovery. An important matter for such a technology to be used routinely is its reproducibility. It is of significant interest to develop quantitative measures to evaluate the quality and reliability of different experimental methods. We compare the quality of SELDI-TOF MS data using unfractionated, fractionated plasma samples and abundant protein depletion methods in terms of the numbers of detected peaks and reliability. Several statistical quality-control and quality-assessment techniques are proposed, including the Graeco-Latin square design for the sample allocation on a Protein chip, the use of the pairwise Pearson correlation coefficient as the similarity measure between the spectra in conjunction with multi-dimensional scaling (MDS) for graphically evaluating similarity of replicates and assessing outlier samples; and the use of the reliability ratio for evaluating reproducibility. Our results show that the number of peaks detected is similar among the three sample preparation technologies, and the use of the Sigma multi-removal kit does not improve peak detection. Fractionation of plasma samples introduces more experimental variability. The peaks detected using the unfractionated plasma samples have the highest reproducibility as determined by the reliability ratio. Our algorithm for assessment of SELDI-TOF experiment quality is available at http://www.biostat.harvard.edu/~xlin. Supplementary data are available at Bioinformatics online.

Distinct Expression Patterns of Natural Antisense Transcripts in Arabidopsis

It has been shown that overlapping cis-natural antisense transcripts (cis-NATs) can form a regulatory circuit in which small RNAs derived from one transcript regulate stability of the other transcript, which manifests itself as anticorrelated expression. However, little is known about how widespread antagonistic expression of cis-NATs is. We have determined how frequently cis-NAT pairs, which make up 7.4% of annotated transcription units in the Arabidopsis (Arabidopsis thaliana) genome, show anticorrelated expression patterns. Indeed, global expression profiles of pairs of cis-NATs on average have significantly lower pairwise Pearson correlation coefficients than other pairs of neighboring genes whose transcripts do not overlap. However, anticorrelated expression that is greater than expected by chance is found in only a small number of cis-NAT pairs. The degree of anticorrelation does not depend on the length of the overlap or on the distance of the 5' ends of the transcripts. Consistent with earlier findings, cis-NATs do not exhibit an increased likelihood to give rise to small RNAs, as determined from available small RNA sequences and massively parallel signature sequencing tags. However, the overlapping regions of cis-NATs appeared to be enriched for small RNA loci compared to nonoverlapping regions. Furthermore, expression of cis-NATs was not disproportionately affected in various RNA-silencing mutants. Our results demonstrate that there is a trend toward anticorrelated expression of cis-NAT pairs in Arabidopsis, but currently available data do not produce a strong signature of small RNA-mediated silencing for this process.

Universality of color names

We analyzed the World Color Survey (WCS) color-naming data set by using k-means cluster and concordance analyses. Cluster analysis relied on a similarity metric based on pairwise Pearson correlation of the complete chromatic color-naming patterns obtained from individual WCS informants. When K, the number of k-means clusters, varied from 2 to 10, we found that (i) the average color-naming patterns of the clusters all glossed easily to single or composite English patterns, and (ii) the structures of the k-means clusters unfolded in a hierarchical way that was reminiscent of the Berlin and Kay sequence of color category evolution. Gap statistical analysis showed that 8 was the optimal number of WCS chromatic categories: RED, GREEN, YELLOW-OR-ORANGE, BLUE, PURPLE, BROWN, PINK, and GRUE (GREEN-OR-BLUE). Analysis of concordance in color naming within WCS languages revealed small regions in color space that exhibited statistically significantly high concordance across languages. These regions agreed well with five of six primary focal colors of English. Concordance analysis also revealed boundary regions of statistically significantly low concordance. These boundary regions coincided with the boundaries associated with English WARM and COOL. Our results provide compelling evidence for similarities in the mechanisms that guide the lexical partitioning of color space among WCS languages and English.

Mistaken view of taxonomic validity undermines conservation of an evolutionarily distinct mouse: a response to Ramey et al. (2005)

1 Department of Biology, University of Washington, Seattle, WA, USA 2 Department of Fisheries and Wildlife Sciences, Virginia Polytechnic Institute and State University, Blacksburg, VA, USA 3 Department of Biology, Valdosta State University, Valdosta, GA, USA 4 New Mexico Museum of Natural History, Albuquerque, NM, USA 5 Department of Environmental, Population, and Organismic Biology, University of Colorado, Boulder, CO, USA 6 School of Molecular and Life Sciences, University of Limpopo, Sovenga, South Africa

Clusters of Adjacent and Similarly Expressed Genes across Normal Human Tissues Complicate Comparative Transcriptomic Discovery

Transcriptomic techniques are valuable tools with which to validate genetic and biological hypotheses and are now widely available for research. However, with the exception of tumor biology, comparative genomics analyses have been difficult to use as discovery engines to describe biologically relevant expression changes. We propose that physical proximity of human genes correlates with similar mRNA expression, so that increased expression might include a disease-relevant gene and many other genes in the adjacent region. To increase the efficiency of combining susceptibility gene mapping and interpretation of transcriptomics, we developed a method to identify clusters of adjacent and similarly expressed genes. Gene expression profiles for 28,945 genes across 101 normal human tissues were obtained from the Gene Logic BioExpress system. The expression similarity for genes in sliding-windows was measured using average pair-wise Pearson correlation coefficients. We identified 187 clusters (p < 10e-4) of co-regulated genes, including 2648 genes, or 9.1% of all genes considered and termed these "clusters of adjacent and similarly expressed genes" (CASEGs). Genes in 15 (8.2%) of these clusters demonstrate a significant co-expression enrichment (p < 10e-10). This study demonstrates the coordinate expression of neighboring genes and provides a comprehensive view of expression-based compartmentalization of the human genome, which can be overlaid on genetic susceptibility gene maps.

Local Coexpression Domains of Two to Four Genes in the Genome of Arabidopsis

Expression of genes in eukaryotic genomes is known to cluster, but cluster size is generally loosely defined and highly variable. We have here taken a very strict definition of cluster as sets of physically adjacent genes that are highly coexpressed and form so-called local coexpression domains. The Arabidopsis (Arabidopsis thaliana) genome was analyzed for the presence of such local coexpression domains to elucidate its functional characteristics. We used expression data sets that cover different experimental conditions, organs, tissues, and cells from the Massively Parallel Signature Sequencing repository and microarray data (Affymetrix) from a detailed root analysis. With these expression data, we identified 689 and 1,481 local coexpression domains, respectively, consisting of two to four genes with a pairwise Pearson's correlation coefficient larger than 0.7. This number is approximately 1- to 5-fold higher than the numbers expected by chance. A small (5%-10%) yet significant fraction of genes in the Arabidopsis genome is therefore organized into local coexpression domains. These local coexpression domains were distributed over the genome. Genes in such local domains were for the major part not categorized in the same functional category (GOslim). Neither tandemly duplicated genes nor shared promoter sequence nor gene distance explained the occurrence of coexpression of genes in such chromosomal domains. This indicates that other parameters in genes or gene positions are important to establish coexpression in local domains of Arabidopsis chromosomes.

Comparison of six calcaneal quantitative ultrasound devices: precision and hip fracture discrimination.

Quantitative ultrasound (QUS) is now accepted as a useful tool in the management of osteoporosis. There are a variety of QUS devices clinically available with a number of differences among them, including their coupling methods, parameter calculation algorithms and sites of measurement. This study evaluated the abilities of six calcaneal QUS devices to discriminate between normal and hip-fractured subjects compared with the established method of dual-energy X-ray absorptiometry (DXA). The short-term and mid-term precisions of these devices were also determined. Thirty-five women (mean age 74.5+/-7.9 years) who had sustained a hip fracture within the past 3 years, and 35 age-matched controls (75.8+/-5.6 years) were recruited. Ultrasound measurements were acquired using six ultrasound devices: three gel-coupled and three water-coupled devices. Bone mineral density was measured at the hip using DXA. Discrimination of fracture patients versus controls was assessed using logistic regression analysis (expressed as age- and BMI-adjusted odds ratios per standard deviation decrease with 95% confidence interval) and receiver operating characteristics (ROC) curve analysis. Measurement precision was standardized to the biological range (sCV). The sCV ranged from 3.14% to 5.5% for speed of sound (SOS) and from 2.45% to 6.01% for broadband ultrasound attenuation (BUA). The standardized medium-term precision ranged from 4.33% to 8.43% for SOS and from 2.77% to 6.91% for BUA. The pairwise Pearson correlation coefficients between different devices was highly significant (SOS, r = 0.79-0.93; BUA, r = 0.71-0.92). QUS variables correlated weakly, though significantly, with femoral BMD (SOS, r = 0.30-0.55; BUA, r = 0.35-0.61). The absolute BUA and SOS values varied among devices. The gel-coupled devices generally had a higher SOS than water-coupled devices. Bone mineral density (BMD) and BUA were weakly correlated with weight (r = 0.48-0.57 for BMD and r = 0.18-0.54 for BUA), whereas SOS was independent of weight. All the QUS devices gave similar, statistically significant hip fracture discrimination for both SOS and BUA measures. The odds ratios for SOS (2.1-2.8) and BUA (2.4-3.4) were comparable to those for femoral BMD (2.6-3.5), as were the area under the curve (SOS, 0.65-0.71; BUA, 0.62-0.71; BMD, 0.65-0.74) from ROC analysis. Within the limitation of the sample size all devices show similar diagnostic sensitivity.