Dynamic load balancing in parallel KD-tree k-means

One among the most influential and popular data mining methods is the k-Means algorithm for cluster analysis. Techniques for improving the efficiency of k-Means have been largely explored in two main directions. The amount of computation can be significantly reduced by adopting geometrical constraints and an efficient data structure, notably a multidimensional binary search tree (KD-Tree). These techniques allow to reduce the number of distance computations the algorithm performs at each iteration. A second direction is parallel processing, where data and computation loads are distributed over many processing nodes. However, little work has been done to provide a parallel formulation of the efficient sequential techniques based on KD-Trees. Such approaches are expected to have an irregular distribution of computation load and can suffer from load imbalance. This issue has so far limited the adoption of these efficient k-Means variants in parallel computing environments. In this work, we provide a parallel formulation of the KD-Tree based k-Means algorithm for distributed memory systems and address its load balancing issue. Three solutions have been developed and tested. Two approaches are based on a static partitioning of the data set and a third solution incorporates a dynamic load balancing policy.

Epidermal growth factor-stimulated extravillous cytotrophoblast motility is mediated by the activation of PI3-K, Akt and both p38 and p42/44 mitogen-activated protein kinases

BACKGROUND: Trophoblast invasion is a temporally and spatially regulated scheme of events that can dictate pregnancy outcome. Evidence suggests that the potent mitogen epidermal growth factor (EGF) regulates cytotrophoblast (CTB) differentiation and invasion during early pregnancy. METHODS AND RESULTS: In the present study, the first trimester extravillous CTB cell line SGHPL-4 was used to investigate the signalling pathways involved in the motile component of EGF-mediated CTB migration/invasion. EGF induced the phosphorylation of the phosphatidylinositol 3-kinase (PI3-K)-dependent proteins, Akt and GSK-3β as well as both p42/44 MAPK and p38 mitogen-activated protein kinases (MAPK). EGF-stimulated motility was significantly reduced following the inhibition of PI3-K (P < 0.001), Akt (P < 0.01) and both p42/44 MAPK (P < 0.001) and p38 MAPKs (P < 0.001) but not the inhibition of GSK-3β. Further analysis indicated that the p38 MAPK inhibitor SB 203580 inhibited EGF-stimulated phosphorylation of Akt on serine 473, which may be responsible for the effect SB 203580 has on CTB motility. Although Akt activation leads to GSK-3β phosphorylation and the subsequent expression of β-catenin, activation of this pathway by 1-azakenpaullone was insufficient to stimulate the motile phenotype. CONCLUSION: We demonstrate a role for PI3-K, p42/44 MAPK and p38 MAPK in the stimulation of CTB cell motility by EGF, however activation of β-catenin alone was insufficient to stimulate cell motility.

Reversible phase variation in the phnE gene, which is required for phosphonate metabolism in Escherichia coli K-12

It is known that Escherichia coli K-12 is cryptic (Phn(-)) for utilization of methyl phosphonate (MePn) and that Phn(+) variants can be selected for growth on MePn as the sole P source. Variants arise from deletion via a possible slip strand mechanism of one of three direct 8-bp repeat sequences in phnE, which restores function to a component of a putative ABC type transporter. Here we show that Phn(+) variants are present at the surprisingly high frequency of >10(-2) in K-12 strains. Amplified-fragment length polymorphism analysis was used to monitor instability in phnE in various strains growing under different conditions. This revealed that, once selection for growth on MePn is removed, Phn(+) revertants reappear and accumulate at high levels through reinsertion of the 8-bp repeat element sequence. It appears that, in K-12, phnE contains a high-frequency reversible gene switch, producing phase variation which either allows ("on" form) or blocks ("off" form) MePn utilization. The switch can also block usage of other metabolizable alkyl phosphonates, including the naturally occurring 2-aminoethylphosphonate. All K-12 strains, obtained from collections, appear in the "off" form even when bearing mutations in mutS, mutD, or dnaQ which are known to enhance slip strand events between repetitive sequences. The ability to inactivate the phnE gene appears to be unique to K-12 strains since the B strain is naturally Phn(+) and lacks the inactivating 8-bp insertion in phnE, as do important pathogenic strains for which genome sequences are known and also strains isolated recently from environmental sources.

DNA variation in a 13-Mb region including the F9 gene: inferring the genealogical history and causal role of a hemophilia B mutation (IVS 5+13 A -> G)

About 5.5% of all UK hemophilia B patients have the base substitution IVS 5+13 A-->G as the only change in their factor (F)IX gene (F9). This generates a novel donor splice site which fits the consensus better than the normal intron 5 donor splice. Use of the novel splice site should result in a missense mutation followed by the abnormal addition of four amino acids to the patients' FIX. In order to explain the prevalence of this mutation, its genealogical history is examined. Analysis of restriction fragment length polymorphism in the 21 reference UK individuals (from different families) with the above mutation showed identical haplotypes in 19 while two differed from the rest and from each other. In order to investigate the history of the mutation and to verify that it had occurred independently more than once, the sequence variation in 1.5-kb segments scattered over a 13-Mb region including F9 was examined in 18 patients and 15 controls. This variation was then analyzed with a recently developed Bayesian approach that reconstructs the genealogy of the gene investigated while providing evidence of independent mutations that contribute disconnected branches to the genealogical tree. The method also provides minimum estimates of the age of the mutation inherited by the members of coherent trees. This revealed that 17 or 18 mutant genes descend from a founder who probably lived 450 years ago, while one patient carries an independent mutation. The independent recurrence of the IVS5+13 A-->G mutation strongly supports the conclusion that it is the cause of these patients' mild hemophilia.

PDGF regulates the actin cytoskeleton through hnRNP-K-mediated activation of the ubiquitin E-3-ligase MIR

PDGF is a potent chemotactic mitogen and a strong inductor of fibroblast motility. In Swiss 3T3 fibroblasts, exposure to PDGF but not EGF or IGF-1 causes a rapid loss of actin stress fibers (SFs) and focal adhesions (FAs), which is followed by the development of retractile dendritic protrusions and induction of motility. The PDGF-specific actin reorganization was blocked by inhibition of Src-kinase and the 26S proteasome. PDGF induced Src-dependent association between the multifunctional transcription/translation regulator hnRNP-K and the mRNA-encoding myosin regulatory light-chain (MRLC)-interacting protein (MIR), a E3-ubiquitin ligase that is MRLC specific. This in turn rapidly increased MIR expression, and led to ubiquitination and proteasome-mediated degradation of MRLC. Downregulation of MIR by RNA muting prevented the reorganization of actin structures and severely reduced the migratory and wound-healing potential of PDGF-treated cells. The results show that activation of MIR and the resulting removal of diphosphorylated MRLC are essential for PDGF to instigate and maintain control over the actin-myosin-based contractile system in Swiss 3T3 fibroblasts. The PDGF induced protein destabilization through the regulation of hnRNP-K controlled ubiquitin-ligase translation identifies a novel pathway by which external stimuli can regulate phenotypic development through rapid, organelle-specific changes in the activity and stability of cytoskeletal regulators.

Stress generation in the tension wood of poplar is based on the lateral swelling power of the G-layer

The mechanism of active stress generation in tension wood is still not fully understood. To characterize the functional interdependency between the G-layer and the secondary cell wall, nanostructural characterization and mechanical tests were performed on native tension wood tissues of poplar (Populus nigra x Populus deltoids) and on tissues in which the G-layer was removed by an enzymatic treatment. In addition to the well-known axial orientation of the cellulose fibrils in the G-layer, it was shown that the microfibril angle of the S2-layer was very large (about 36 degrees). The removal of the G-layer resulted in an axial extension and a tangential contraction of the tissues. The tensile stress-strain curves of native tension wood slices showed a jagged appearance after yield that could not be seen in the enzyme-treated samples. The behaviour of the native tissue was modelled by assuming that cells deform elastically up to a critical strain at which the G-layer slips, causing a drop in stress. The results suggest that tensile stresses in poplar are generated in the living plant by a lateral swelling of the G-layer which forces the surrounding secondary cell wall to contract in the axial direction.

Agonist binding, agonist affinity and agonist efficacy at G protein-coupled receptors

Measurements of affinity and efficacy are fundamental for work on agonists both in drug discovery and in basic studies on receptors. In this review I wish to consider methods for measuring affinity and efficacy at G protein coupled receptors (GPCRs). Agonist affinity may be estimated in terms of the dissociation constant for agonist binding to a receptor using ligand binding or functional assays. It has, however, been suggested that measurements of affinity are always contaminated by efficacy so that it is impossible to separate the two parameters. Here I show that for many GPCRs, if receptor/G protein coupling is suppressed, experimental measurements of agonist affinity using ligand binding (K-obs) provide quite accurate measures of the agonist microscopic dissociation constant (K-A). Also in pharmacological functional studies, good estimates of agonist dissociation constants are possible. Efficacy can be quantitated in several ways based on functional data ( maximal effect of the agonist (E-max), ratio of agonist dissociation constant to concentration of agonist giving half maximal effect in functional assay ( K-obs/ EC50), a combined parameter EmaxKobs/EC50). Here I show that EmaxKobs/EC50 provides the best assessment of efficacy for a range of agonists across the full range of efficacy for full to partial agonists. Considerable evidence now suggests that ligand efficacy may be dependent on the pathway used to assess it. The efficacy of a ligand may, therefore, be multidimensional. It is still, however, necessary to have accurate measures of efficacy in different pathways.

Mechanisms of G protein activation via the D-2 dopamine receptor: evidence for persistent receptor/G protein interaction after agonist stimulation

Background and purpose: The aim of this report is to study mechanisms of G protein activation by agonists. Experimental approach: The association and dissociation of guanosine 5'-O-(3-[S-35] thio) triphosphate ([S-35] GTP gamma S) binding at G proteins in membranes of CHO cells stably transfected with the human dopamine D-2short receptor was studied in the presence of a range of agonists. Key results: Binding of [S-35] GTPgS was dissociable in the absence of agonist and dissociation was accelerated both in rate and extent by dopamine, an effect which was blocked by the dopamine D-2 receptor antagonist raclopride and by suramin, which inhibits receptor/G protein interaction. A range of agonists of varying efficacy increased the rate of dissociation of [S-35] GTPgS binding, with the more efficacious agonists resulting in faster dissociation. Agonists were able to dissociate about 70% of the pre-bound [S-35] GTPgS, leaving a component which may not be accessible to the agonist-bound receptor. The dissociable component of the [S-35] GTPgS binding was reduced with longer association times and increased [S-35] GTPgS concentrations. Conclusions and implications: These data are consistent with [S-35] GTPgS binding being initially to receptor-linked G proteins and then to G proteins which have separated from the agonist bound receptor. Under the conditions used typically for [S-35] GTPgS binding assays, therefore, much of the agonist-receptor complex remains in proximity to G proteins after they have been activated by agonist.

A lower bound on the size of k-neighborhood in generalized cubes

The determination of the minimum size of a k-neighborhood (i.e., a neighborhood of a set of k nodes) in a given graph is essential in the analysis of diagnosability and fault tolerance of multicomputer systems. The generalized cubes include the hypercube and most hypercube variants as special cases. In this paper, we present a lower bound on the size of a k-neighborhood in n-dimensional generalized cubes, where 2n + 1 <= k <= 3n - 2. This lower bound is tight in that it is met by the n-dimensional hypercube. Our result is an extension of two previously known results. (c) 2005 Elsevier Inc. All rights reserved.

Results of an aqueous source term model for a radiological risk assessment of the Drigg LLW Site, U.K.

A radionuclide source term model has been developed which simulates the biogeochemical evolution of the Drigg low level waste (LLW) disposal site. The DRINK (DRIgg Near field Kinetic) model provides data regarding radionuclide concentrations in groundwater over a period of 100,000 years, which are used as input to assessment calculations for a groundwater pathway. The DRINK model also provides input to human intrusion and gaseous assessment calculations through simulation of the solid radionuclide inventory. These calculations are being used to support the Drigg post closure safety case. The DRINK model considers the coupled interaction of the effects of fluid flow, microbiology, corrosion, chemical reaction, sorption and radioactive decay. It represents the first direct use of a mechanistic reaction-transport model in risk assessment calculations.

Laboratory measurements of the water vapor continuum in the 1200–8000 cm−1 region between 293 K and 351 K

Laboratory Fourier transform spectroscopy of pure water vapor and water vapor mixed with air has been conducted between 1200 and 8000 cm−1 and at temperatures between 293 and 351 K with the purpose of detecting and characterizing the water vapor continuum. The spectral features of the continuum within the major water absorption bands are presented and compared where possible to those from previous experimental studies and to the commonly used MT_CKD and CKD models. It was observed that in the main, both models adequately capture the general spectral form of the continuum; however, there were a number of exceptions. Overall, there is no evidence to indicate that MT_CKD is an improvement upon the older CKD model in these spectral regions. There was generally good agreement between our results and those of other experimental investigators. The general mathematical forms of the self-continuum temperature dependence, given by both Roberts et al. (1976) and CKD/MT_CKD, fit well to the experimental continuum in these spectral regions. However, the range of temperatures over which we made measurements is not sufficient to discriminate between these two forms or to exclude the possibility of other forms of temperature dependence being more appropriate. At the same time, the actual parameters currently used in CKD/MT_CKD to describe the temperature dependence in many spectral regions cannot reproduce the observed strong spectral variation in the temperature dependence. It has not been possible to make definitive conclusions about the magnitude of the continuum absorption in the far wings of the absorption bands investigated here.

Space partitioning for scalable K-means

K-Means is a popular clustering algorithm which adopts an iterative refinement procedure to determine data partitions and to compute their associated centres of mass, called centroids. The straightforward implementation of the algorithm is often referred to as `brute force' since it computes a proximity measure from each data point to each centroid at every iteration of the K-Means process. Efficient implementations of the K-Means algorithm have been predominantly based on multi-dimensional binary search trees (KD-Trees). A combination of an efficient data structure and geometrical constraints allow to reduce the number of distance computations required at each iteration. In this work we present a general space partitioning approach for improving the efficiency and the scalability of the K-Means algorithm. We propose to adopt approximate hierarchical clustering methods to generate binary space partitioning trees in contrast to KD-Trees. In the experimental analysis, we have tested the performance of the proposed Binary Space Partitioning K-Means (BSP-KM) when a divisive clustering algorithm is used. We have carried out extensive experimental tests to compare the proposed approach to the one based on KD-Trees (KD-KM) in a wide range of the parameters space. BSP-KM is more scalable than KDKM, while keeping the deterministic nature of the `brute force' algorithm. In particular, the proposed space partitioning approach has shown to overcome the well-known limitation of KD-Trees in high-dimensional spaces and can also be adopted to improve the efficiency of other algorithms in which KD-Trees have been used.

Determination by MAD-DM of the structure of the DNA duplex d[ACGTACG(5-BrU)]2 at 1.46 Å and 100 K

A four-wavelength MAD experiment on a new brominated octanucleotide is reported here. d[ACGTACG(5-BrU)], C77H81BrN30O32P7, (DNA) = 2235, tetragonal, P43212 (No. 96), a = 43.597, c = 26.268 Å, V = 49927.5 Å3, Z = 8, T = 100 K, R = 10.91% for 4312 reflections between 15.0 and 1.46 Å resolution. The self-complementary brominated octanucleotide d[ACGTACG(5-BrU)]2 has been crystallized and data measured to 1.45 Å at both 293 K and a second crystal flash frozen at 100 K. The latter data collection was carried out to the same resolution at the four wavelengths 0.9344, 0.9216, 0.9208 and 0.9003 Å, around the Br K edge at 0.92 Å and the structure determined from a map derived from a MAD data analysis using pseudo-MIR methodology, as implemented in the program MLPHARE. This is one of the first successful MAD phasing experiments carried out at Sincrotrone Elettra in Trieste, Italy. The structure was refined using the data measured at 0.9003 Å, anisotropic temperature factors and the restrained least-squares refinement implemented in the program SHELX96, and the helical parameters are compared with those previously determined for the isomorphous d(ACGTACGT)2 analogue. The asymmetric unit consists of a single strand of octamer with 96 water molecules. No countercations were located. The A-DNA helix geometry obtained has been analysed using the CURVES program.

Molecular modelling studies of binding of DACD derivatives into G-Quadruplex DNA: comparison of force field and quantum polarized ligand docking methods

The DNA G-qadruplexes are one of the targets being actively explored for anti-cancer therapy by inhibiting them through small molecules. This computational study was conducted to predict the binding strengths and orientations of a set of novel dimethyl-amino-ethyl-acridine (DACA) analogues that are designed and synthesized in our laboratory, but did not diffract in Synchrotron light.Thecrystal structure of DNA G-Quadruplex(TGGGGT)4(PDB: 1O0K) was used as target for their binding properties in our studies.We used both the force field (FF) and QM/MM derived atomic charge schemes simultaneously for comparing the predictions of drug binding modes and their energetics. This study evaluates the comparative performance of fixed point charge based Glide XP docking and the quantum polarized ligand docking schemes. These results will provide insights on the effects of including or ignoring the drug-receptor interfacial polarization events in molecular docking simulations, which in turn, will aid the rational selection of computational methods at different levels of theory in future drug design programs. Plenty of molecular modelling tools and methods currently exist for modelling drug-receptor or protein-protein, or DNA-protein interactionssat different levels of complexities.Yet, the capasity of such tools to describevarious physico-chemical propertiesmore accuratelyis the next step ahead in currentresearch.Especially, the usage of most accurate methods in quantum mechanics(QM) is severely restricted by theirtedious nature. Though the usage of massively parallel super computing environments resulted in a tremendous improvement in molecular mechanics (MM) calculations like molecular dynamics,they are still capable of dealing with only a couple of tens to hundreds of atoms for QM methods. One such efficient strategy that utilizes thepowers of both MM and QM are the QM/MM hybrid methods. Lately, attempts have been directed towards the goal of deploying several different QM methods for betterment of force field based simulations, but with practical restrictions in place. One of such methods utilizes the inclusion of charge polarization events at the drug-receptor interface, that is not explicitly present in the MM FF.