Pattern-based clustering of daily weigh-in trajectories using dynamic time warping

“Smart”-scales are a new tool for frequent monitoring of weight change as well as weigh-in behavior. These scales give researchers the opportunity to discover patterns in the frequency that individuals weigh themselves over time, and how these patterns are associated with overall weight loss. Our motivating data come from an 18-month behavioral weight loss study of 55 adults classified as overweight or obese who were instructed to weigh themselves daily. Adherence to daily weigh-in routines produces a binary times series for each subject, indicating whether a participant weighed in on a given day. To characterize weigh-in by time-invariant patterns rather than overall adherence, we propose using hierarchical clustering with dynamic time warping (DTW). We perform an extensive simulation study to evaluate the performance of DTW compared to Euclidean and Jaccard distances to recover underlying patterns in adherence time series. In addition, we compare cluster performance using cluster validation indices (CVIs) under the single, average, complete, and Ward linkages and evaluate how internal and external CVIs compare for clustering binary time series. We apply conclusions from the simulation to cluster our real data and summarize observed weigh-in patterns. Our analysis finds that the adherence trajectory pattern is significantly associated with weight loss.     

Chromosomal localization of uroplakin genes of cattle and mice

The asymmetric unit membrane (AUM) of the apical surface of mammalian urinary bladder epithelium contains several major integral membrane proteins, including uroplakins IA and IB (both 27 kDa), II (15 kDa), and III (47 kDa). These proteins are synthesized only in terminally differentiated bladder epithelial cells. They are encoded by separate genes and, except for uroplakins IA and IB, appear to be unrelated in their amino acid sequences. The genes encoding these uroplakins were mapped to chromosomes of cattle through their segregation in a panel of bovine x rodent somatic cell hybrids. Genes for uroplakins IA, IB, and II were mapped to bovine (BTA) Chromosomes (Chrs) 18 (UPK1A), 1 (UPK1B), and 15 (UPK2), respectively. Two bovine genomic DNA sequences reactive with a uroplakin III cDNA probe were identified and mapped to BTA 6 (UPK3A) and 5 (UPK3B). We have also mapped genes for uroplakins 1A and II in mice, to the proximal regions of mouse Chr 7 (Upk1a) and 9 (Upk2), respectively, by analyzing the inheritance of restriction fragment length variants in recombinant inbred mouse strains. These assignments are consistent with linkage relationships known to be conserved between cattle and mice. The mouse genes for uroplakins IB and III were not mapped because the mouse genomic DNA fragments reactive with each probe were invariant among the inbred strains tested. Although the stoichiometry of AUM proteins is nearly constant, the fact that the uroplakin genes are unlinked indicates that their expression must be independently regulated. Our results also suggest likely positions for two human uroplakin genes and should facilitate further analysis of their possible involvement in disease.     

Frequent nest relocation in the ant Aphaenogaster araneoides: resources, competition, and natural enemies

Sessile and vagile organisms differ from one another in some fundamental ways, including methods of resource acquisition and competition. Ant colonies are typically studied as sessile entities, even though a large fraction of ant species frequently relocate their nests in the course of their life history. Little is known about the causes and consequences of nest relocation, but it is likely that the costs and benefits of relocation are driven by nest quality, neighborhood competition, or resource availability. In this paper, we document several cycles of nest relocation in a population of the Central American ant Aphaenogaster araneoides . In our first experiment, we tracked the pattern of relocation, testing whether environmental characteristics and colony demography were associated with relocation behavior. In our second experiment, we manipulated resource availability by adding or subtracting leaf litter, which is known to predict colony growth. We found that colonies relocated their nests once per week on average and colonies often reoccupied nests from which they had once emigrated. Larger colonies relocated more frequently than smaller colonies, and quickly growing colonies utilized a greater number of nests within their home range compared to slowly growing colonies. Relocation events were most likely to occur in periods when vapor pressure deficits were greatest. Nearest neighbor distance and other measures of environmental conditions were not associated with relocation behavior and there was no significant effect of litter removal or supplementation. We found evidence that multiple natural enemies attacked A. araneoides colonies. Based on the demographic correlates of relocation and our rejection of other plausible hypotheses, we propose that nest relocation is driven by the escape from natural enemies.     

Inadequate rates are lower when FNAC samples are taken by cytopathologists

Inadequate rates (IR) in FNAC from different sources were compared. The rates were lowest when FNAC was performed by a cytopathologist (12%) and highest when done by a non-cytopathologist (32%). These differences were mirrored in high IRs in breast cancer cases. IR was not significantly improved when non-cytopathologist FNAC was attended by a cytotechnician.     

Estimating dataset size requirements for classifying DNA microarray data.

A statistical methodology for estimating dataset size requirements for classifying microarray data using learning curves is introduced. The goal is to use existing classification results to estimate dataset size requirements for future classification experiments and to evaluate the gain in accuracy and significance of classifiers built with additional data. The method is based on fitting inverse power-law models to construct empirical learning curves. It also includes a permutation test procedure to assess the statistical significance of classification performance for a given dataset size. This procedure is applied to several molecular classification problems representing a broad spectrum of levels of complexity.     

Identification of a Novel Cryptosporidium Genotype in Pigs

Over a 3-year period, a total of 646 fecal samples from pigs in 22 indoor and outdoor herds from Western Australia were screened for Cryptosporidium spp. by microscopy. Results revealed that 39 of 646 samples (6.03%) were positive for Cryptosporidium. Cryptosporidium was much more common in outdoor herds (17.2%) than in indoor herds (0.5%) and was more common in animals between the ages of 5 and 8 weeks (69.2%) than in younger animals (P < 0.0001). Molecular characterization of the positive samples at the 18S ribosomal DNA locus identified two distinct genotypes of Cryptosporidium: the previously identified pig genotype I and a novel pig genotype (pig genotype II), both of which warrant species status.     

IN VIVO EFFECTS OF CARDIOTROPHIN-1

Cardiotrophin-1 (CT-1) is a recently discovered cytokine that was isolated based on its ability to induce cardiac myocyte hypertrophy in vitro. In this study, the effects of chronic administration of CT-1 to mice (0.5 or 2 μg by intraperitoneal injection, twice a day for 14 days) were determined. A dose-dependent increase in both the heart weight and ventricular weight to body ratios was observed in the treated groups. The body weights of the animals were unaffected. These results indicate that CT-1 can induce cardiac hypertrophy in vivo. CT-1 was not specific for the heart, however. It stimulated the growth of the liver, kidney, and spleen, and caused atrophy of the thymus. CT-1 administration also increased the platelet counts by 70%, with no change in mean platelet volume. Red blood cell counts were increased in the treated animals, and there was a concomitant increase in haemoglobin concentration. Thus, CT-1 has a broad spectrum of biological activities in vivo. This observation is consistent with previous in-vitro findings showing that the mRNA for CT-1 is expressed in several tissues, and that CT-1 can function through binding to the leukaemia inhibitory factor (LIF) receptor and signalling through the gp130 pathway.     

Crystal structure of the phosphatidylinositol-specific phospholipase C from Bacillus cereus in complex with myo-inositol.

Phosphatidylinositol (PI), once regarded as an obscure component of membranes, is now recognized as an important reservoir of second messenger precursors and as an anchor for membrane enzymes. PI-specific phospholipase C (PI-PLC) is the enzyme that cleaves PI, invoking numerous cellular responses. The crystal structure of PI-PLC from Bacillus cereus (EC 3.1.4.10) has been solved at 2.6 A resolution and refined to a crystallographic R factor of 18.7%. The structure consists of an imperfect (beta alpha)8-barrel similar to that first observed for triose phosphate isomerase and does not resemble any other known phospholipase structure. The active site of the enzyme has been identified by determining the structure of PI-PLC in complex with its inhibitor, myo-inositol, at 2.6 A resolution (R factor = 19.5%). This substrate-like inhibitor interacts with a number of residues highly conserved among prokaryotic PI-PLCs. Residues His32 and His82, which are also conserved between prokaryotic and eukaryotic PI-PLCs, most likely act as general base and acid respectively in a catalytic mechanism analogous to that observed for ribonucleases.     

Analysis of Orthologous Retrovirus-Like Elements in the White-Footed Mouse,Peromyscus leucopus

Three loci in the genome of the white-footed mouse, Peromyscus leucopus, were examined for the presence or absence of orthologous copies of the retrovirus-like element mys using polymerase chain reaction. We examined these loci in 28 mice collected throughout the P. leucopus species range. Mys insertions were present in only one of the individuals examined at the mys-1 and mys-7 loci. Conversely, the mys-6 element was found in several individuals, but the presence of this element was limited to northern latitudes. Because the long terminal repeats (LTRs) of a given element are expected to be identical at the time of retrotransposition into the genome, and to accumulate changes over evolutionary time, within-element LTR sequence comparisons can be used to estimate the relative age of insertions. Within-element LTR differences are greater in mys-6 than in mys-1 or mys-7. The LTRs from orthologous mys-6 elements of six mice were sequenced. The alignment revealed 13 of the 22 differences between the right and left LTRs that were shared by all orthologous mys-6 sites, suggesting that relative to its time of insertion into the genome, mys-6 has only recently spread across the northern part of the species range. Received: 23 January 1996 / Accepted: 24 April 1996