Sparse density estimation on the multinomial manifold

A new sparse kernel density estimator is introduced based on the minimum integrated square error criterion for the finite mixture model. Since the constraint on the mixing coefficients of the finite mixture model is on the multinomial manifold, we use the well-known Riemannian trust-region (RTR) algorithm for solving this problem. The first- and second-order Riemannian geometry of the multinomial manifold are derived and utilized in the RTR algorithm. Numerical examples are employed to demonstrate that the proposed approach is effective in constructing sparse kernel density estimators with an accuracy competitive with those of existing kernel density estimators.

On-line Gaussian mixture density estimator for adaptive minimum bit-error-rate beamforming receivers

We develop an on-line Gaussian mixture density estimator (OGMDE) in the complex-valued domain to facilitate adaptive minimum bit-error-rate (MBER) beamforming receiver for multiple antenna based space-division multiple access systems. Specifically, the novel OGMDE is proposed to adaptively model the probability density function of the beamformer’s output by tracking the incoming data sample by sample. With the aid of the proposed OGMDE, our adaptive beamformer is capable of updating the beamformer’s weights sample by sample to directly minimize the achievable bit error rate (BER). We show that this OGMDE based MBER beamformer outperforms the existing on-line MBER beamformer, known as the least BER beamformer, in terms of both the convergence speed and the achievable BER.

The effects of maize planting density and weeding regimes on light and water use

Light and water are among essential resources required for production of photosynthates in plants. A study on the effects of weeding regimes and maize planting density on light and water use was conducted during the 2001/2 short and 2002 long rain seasons at Muguga in - the central highlands of Kenya. Weeding regimes were: weed free (W1), weedy (W2), herbicide (W3) and hand weeding twice (W4). Maize planting densities were 9 (D1) and 18 plants m-2 (D2) intercropped with Phaseolus vulgaris (beans). The experiment was laid as randomized complete block design replicated four times and repeated twice. All plots were thinned to 4 plants m-2 at tasseling stage (96 DAE) and thinnings quantified as forage. Soil moisture content (SMC), photosynthetically active radiation (PAR) interception, evapo-transpiration (ET crop), water use efficiency (WUE), and harvest index (HI), were determined. Percent PAR was higher in D2 than in D1 before thinning but higher in D1 than in D2 after thinning in both seasons. PAR interception was highest in W2 but similar in W1, W3 and W4 in both seasons. SMC was significantly lower in W2 but similar in W1, W3 and W4. D2 had lower SMC than D1 in season two. Weeding regime significantly influenced ET crop, while planting density and weeding regime significantly influenced WUE and HI. D2 maximizes water and light use for forage production but results to increased intra-specific plant competition for water and light severely before thinning (96 DAE) that reduce grain yield in dual purpose maize, relative to D1.

A high-density immunoblotting methodology for quantification of total protein levels and phosphorylation modifications

The components of many signaling pathways have been identified and there is now a need to conduct quantitative data-rich temporal experiments for systems biology and modeling approaches to better understand pathway dynamics and regulation. Here we present a modified Western blotting method that allows the rapid and reproducible quantification and analysis of hundreds of data points per day on proteins and their phosphorylation state at individual sites. The approach is of particular use where samples show a high degree of sample-to-sample variability such as primary cells from multiple donors. We present a case study on the analysis of >800 phosphorylation data points from three phosphorylation sites in three signaling proteins over multiple time points from platelets isolated from ten donors, demonstrating the technique's potential to determine kinetic and regulatory information from limited cell numbers and to investigate signaling variation within a population. We envisage the approach being of use in the analysis of many cellular processes such as signaling pathway dynamics to identify regulatory feedback loops and the investigation of potential drug/inhibitor responses, using primary cells and tissues, to generate information about how a cell's physiological state changes over time.

Dysregulation of granulosal bone morphogenetic protein receptor 1B density is associated with reduced ovarian reserve and the age-related decline in human fertility

Reproductive ageing is linked to the depletion of ovarian primordial follicles, which causes an irreversible change to ovarian cellular function and the capacity to reproduce. The current study aimed to profile the expression of bone morphogenetic protein receptor, (BMPR1B) in 53 IVF patients exhibiting different degrees of primordial follicle depletion. The granulosa cell receptor density was measured in 403 follicles via flow cytometry. A decline in BMPR1B density occurred at the time of dominant follicle selection and during the terminal stage of folliculogenesis in the 23-30 y good ovarian reserve patients. The 40+ y poor ovarian reserve patients experienced a reversal of this pattern. The results demonstrate an association between age-induced depletion of the ovarian reserve and BMPR1B receptor density at the two critical time points of dominant follicle selection and pre-ovulatory follicle maturation. Dysregulation of BMP receptor signalling may inhibit the normal steroidogenic differentiation required for maturation in older patients.

The urban cool island phenomenon in a high-rise high-density city and its mechanisms

The urban heat island (UHI) phenomenon has been studied extensively, but there are relatively fewer reports on the so-called urban cool island (UCI) phenomenon. We reveal here that the UCI phenomenon exists in Hong Kong during the day, and is associated with the UHI at night under all wind and cloud conditions. The possible mechanisms for the UCI phenomenon in such a high-rise compact city have been discovered using a lumped urban air temperature model. A new concept of urban cool island degree hours (UCIdh) to measure the UCI intensity and duration is proposed. Our analyses reveal that when anthropogenic heat is small or absent, a high-rise and high-density city experiences a significant daytime UCI effect. This is explained by an intensified heat storage capacity and the reduced solar radiation gain of urban surfaces. However, if anthropogenic heat in the urban area increases further, the UCI phenomenon still exists, yet UCIdh decrease dramatically in a high-rise compact city. In a low-rise, low-density city, the UCI phenomenon also occurs when there is no anthropogenic heat, but easily disappears when there is little anthropogenic heat, and the UHI phenomenon dominates. This probably explains why the UHI phenomenon is often observed, but the UCI phenomenon is rarely observed. The co-existence of urban heat/cool island phenomena implies reduction of the daily temperature range (DTR) in such cities, and its dependence on urban morphology also implies that urban morphology can be used to control the urban thermal environment.

Sparse density estimator with tunable kernels

A new sparse kernel density estimator with tunable kernels is introduced within a forward constrained regression framework whereby the nonnegative and summing-to-unity constraints of the mixing weights can easily be satisfied. Based on the minimum integrated square error criterion, a recursive algorithm is developed to select significant kernels one at time, and the kernel width of the selected kernel is then tuned using the gradient descent algorithm. Numerical examples are employed to demonstrate that the proposed approach is effective in constructing very sparse kernel density estimators with competitive accuracy to existing kernel density estimators.

Sparse density estimation on multinomial manifold combining local component analysis

A new sparse kernel density estimator is introduced based on the minimum integrated square error criterion combining local component analysis for the finite mixture model. We start with a Parzen window estimator which has the Gaussian kernels with a common covariance matrix, the local component analysis is initially applied to find the covariance matrix using expectation maximization algorithm. Since the constraint on the mixing coefficients of a finite mixture model is on the multinomial manifold, we then use the well-known Riemannian trust-region algorithm to find the set of sparse mixing coefficients. The first and second order Riemannian geometry of the multinomial manifold are utilized in the Riemannian trust-region algorithm. Numerical examples are employed to demonstrate that the proposed approach is effective in constructing sparse kernel density estimators with competitive accuracy to existing kernel density estimators.