Citations

AbstractMotivation: A central problem in biomarker discovery from large-scale gene expression or single nucleotide polymorphism (SNP) data is the computational challenge of taking into account the dependence among all the features. Methods that ignore the dependence usually identify non-reproducible biomarkers across independent datasets. We introduce a new graph-based semi-supervised feature classification algorithm to identify discriminative disease markers by learning on bipartite graphs. Our algorithm directly classifies the feature nodes in a bipartite graph as positive, negative or neutral with network propagation to capture the dependence among both samples and features (clinical and genetic variables) by exploring bi-cluster structures in a graph. Two features of our algorithm are: (1) our algorithm can find a global optimal labeling to capture the dependence among all the features and thus, generates highly reproducible results across independent microarray or other high-thoughput datasets, (2) our algorithm is capable of handling hundreds of thousands of features and thus, is particularly useful for biomarker identification from high-throughput gene expression and SNP data. In addition, although designed for classifying features, our algorithm can also simultaneously classify test samples for disease prognosis/diagnosis.Results: We applied the network propagation algorithm to study three large-scale breast cancer datasets. Our algorithm achieved competitive classification performance compared with SVMs and other baseline methods, and identified several markers with clinical or biological relevance with the disease. More importantly, our algorithm also identified highly reproducible marker genes and enriched functions from the independent datasets.Availability: Supplementary results and source code are available at http://compbio.cs.umn.edu/Feature_Class.Contact: kuang@cs.umn.eduSupplementary information: Supplementary data are available at Bioinformatics online.

The SCFFBW7 ubiquitin ligase degrades proteins involved in cell division, growth, and differentiation and is commonly mutated in cancers. The Fbw7 locus encodes three protein isoforms that occupy distinct subcellular localizations, suggesting that each has unique functions. We used gene targeting to create isoform-specific Fbw7-null mutations in human cells and found that the nucleoplasmic Fbw7α isoform accounts for almost all Fbw7 activity toward cyclin E, c-Myc, and sterol regulatory element binding protein 1. Cyclin E sensitivity to Fbw7 varies during the cell cycle, and this correlates with changes in cyclin E–cyclin-dependent kinase 2 (CDK2)–specific activity, cyclin E autophosphorylation, and CDK2 inhibitory phosphorylation. These data suggest that oscillations in cyclin E–CDK2-specific activity during the cell cycle regulate the timing of cyclin E degradation. Moreover, they highlight the utility of adeno-associated virus–mediated gene targeting in functional analyses of complex loci.

Despite the use of wrist guards during skate- and snowboard activities, fractures still occur at the wrist or at further proximal locations of the forearm. The main objectives of this study were to conduct a human subject testing under simulated falling conditions for measurement of the impact force on the hand, to model wrist guards as a shock isolator, to construct a linear mass-spring-damper model for quantification of the impact force attenuation (Q-ratio) and energy absorption (S-ratio), and to determine whether wrist guards play a role of an efficient shock isolator. While the falling direction (forward and backward) significantly influenced the impact responses, use of wrist guards provided minimal improvements in the Q- and S-ratios. It was suggested based on the results under the submaximal loading conditions that protective functions of the common wrist guard design could be enhanced with substantial increase in the damping ratio so as to maximize the energy absorption. This would bring forth minor deterioration in the impact force attenuation but significant increase in the energy absorption by 19%, which would help better protection against fall-related injuries of the upper extremity.

Abstract Oxidized low-density lipoprotein (oxLDL) has been reported as a major participant in the pathogenesis of atherosclerosis. We hypothesized that oxLDL can also interact with granulocytes during inflammatory airway diseases, such as asthma. To test the chemotactic effect of oxLDL, isolated human peripheral granulocytes were added to the upper chambers of Transwell filters and migration in response to oxLDL was determined. Cu+2-oxidized LDL stimulated neutrophil (23.4 ± 3.2% for 100 μg/ml oxLDL versus 2.9 ± 1.1% for buffer, P < 0.05) and eosinophil (19.3 ± 3.5% versus 0.6 ± 0.02% for buffer, P < 0.05) chemotaxis in a concentration-dependent manner. The magnitude of chemotaxis was dependent on the degree of LDL oxidation. Granulocyte transmigration across IL-1β–activated human pulmonary microvascular endothelial cell monolayers was similarly stimulated by oxLDL. OxLDL activated significant degranulation of both neutrophils (100.9 ± 9.8 versus 49.6 ± 8.4 ng lactoferrin released/5 × 105 neutrophils for buffer, P < 0.05) and eosinophils (342 ± 115.4 versus 85.8 ± 30.4 ng eosinophil-derived neurotoxin/1 × 106 eosinophils for buffer, P < 0.05). Therefore, in vivo influx and oxidation of LDL may be an important mediator for the initiation of bronchial inflammation where granulocytes are recruited to the lung.