Search

A general population model with Allee effect driven by correlated additive and multiplicative white noises is considered. This paper aims to investigate noise-induced phenomenological bifurcation (P-bifurcation) and the influence of noises on the population model. With the help of Fokker-Planck equation, we obtain the stationary probability distribution (SPD) of the model, and find that the shape of SPD experiences a transition from one structure to another when the noise intensity passes through a critical value, i.e., the P-bifurcation occurs. Moreover, detailed analysis and simulations for stochastic logistic-like models with weak and strong Allee effects show that the correlated noises have complex effects on the eventual distribution of population size. This paper aims to investigate noise-induced phenomenological bifurcation (P-bifurcation) for a general population model with Allee effect driven by correlated additive and multiplicative white noises. We find that the shape of SPD experiences a transition from one structure to another when the noise intensity passes through a critical value, i.e., the P-bifurcation occurs.

Haihe River is one of the 7 largest rivers in China. The problem of water pollution in Haihe River Basin is serious. The water quality of Haihe River Basin is generally the best in the north and the worst in the south. And the water quality of the Luanhe River is the best, the proportion of I - III is about 60%, and the water quality of the Tuhaimajia River is the worst, the proportion of >V exceeds 60%. According to the trend of water quality change, the improvement of Shandong Province in Tuhaimajia River system is the most obvious, and proportion of >V water decreased from 100% to about 30%.

Community detection is an important data analysis problem in many different areas, and how to enhance the quality of community detection in complicated real applications is still a challenge. Current community detection enhancement methods often take the enhancement as a preprocess of community detection. They mainly focus on how to design the suitable topological similarity of nodes to adjust the original network, but did not consider how to make use of this topological similarity more effectively. In order to better utilize the similarity information, we propose a model framework which integrates the enhancement into the whole community detection procedure. First, we calculate the structural similarity of nodes based on network topology. Second, we present a stochastic model to describe the community memberships of nodes; we then model the strong constraint based on structural similarity, i.e., we make each node have the same community membership distribution with its most similar neighbors; and then we model the weak constraint, i.e., if two nodes have a high similarity we will make their community membership distributions close, otherwise we will make them not close. Finally, we present a nonnegative matrix factorization approach to learn the model parameters. We evaluate our method on both synthetic and real-world networks with ground-truths, and compare it with five comparable methods. The experimental results demonstrate the superior performance of our new method over the competing ones for community detection and enhancement.

Representation learning is believed to produce high-level representations of underlying dynamics in temporal sequences. A three-dimensional convolutional neural network trained to predict future frames in ultrasound tongue and optical lip images creates features for a continuous hidden Markov model based speech recognition system. Predictive tongue features are found to generate lower word error rates than those obtained from an auto-encoder without future frames, or from discrete cosine transforms. Improvement is apparent for the monophone/triphone Gaussian mixture model and deep neural network acoustic models. When tongue and lip modalities are combined, the advantage of the predictive features is reduced.

Zinc is an essential metal in bacteria. One important bacterial zinc transporter is AdcA, and most bacteria possess AdcA homologs that are single-domain small proteins due to better efficiency of protein biogenesis. However, a double-domain AdcA with two zinc-binding sites is significantly overrepresented in Streptococcus species, many of which are major human pathogens. Using molecular simulation and experimental validations of AdcA from Streptococcus pyogenes, we found here that the two AdcA domains sequentially stabilize the structure upon zinc binding, indicating an organization required for both increased zinc affinity and transfer speed. This structural organization appears to endow Streptococcus species with distinct advantages in zinc-depleted environments, which would not be achieved by each single AdcA domain alone. This enhanced zinc transport mechanism sheds light on the significance of the evolution of the AdcA domain fusion, provides new insights into double-domain transporter proteins with two binding sites for the same ion, and indicates a potential target of antimicrobial drugs against pathogenic Streptococcus species.

Ultrasound imaging of the tongue has been used for decades in studies of speech production and speech motor control, for silent speech interfaces, and in numerous other areas. Despite substantial efforts, however, extraction of reliable features from ultrasound tongue data remains a challenge due to speckle noise and acoustic propagation issues. Recently, Representation Learning has emerged in a variety of fields as a powerful means of generating useful representations of underlying structure in raw, high-dimensional data. In its unsupervised form, Representation Learning discovers structures in unlabelled data, thereby eliminating the need for a time-consuming labelling step. The present work is believed to be the first use of unsupervised Representation Learning to reveal structures related to tongue dynamics in unlabelled ultrasound video. A 3-D Convolutional Neural Network examining a series of unlabelled 60 Hz tongue images is found to accurately predict unseen future images even for large interframe tongue displacements. By comparing the 3DCNN prediction error to that of a simple previous-frame predictor, tongue trajectories containing transitions between regions of acoustic stability can be identified and correlated with formant trajectories in a spectrogram. Prospects for leveraging the tongue dynamic representation for use in subsequent speech processing tasks will be discussed.Ultrasound imaging of the tongue has been used for decades in studies of speech production and speech motor control, for silent speech interfaces, and in numerous other areas. Despite substantial efforts, however, extraction of reliable features from ultrasound tongue data remains a challenge due to speckle noise and acoustic propagation issues. Recently, Representation Learning has emerged in a variety of fields as a powerful means of generating useful representations of underlying structure in raw, high-dimensional data. In its unsupervised form, Representation Learning discovers structures in unlabelled data, thereby eliminating the need for a time-consuming labelling step. The present work is believed to be the first use of unsupervised Representation Learning to reveal structures related to tongue dynamics in unlabelled ultrasound video. A 3-D Convolutional Neural Network examining a series of unlabelled 60 Hz tongue images is found to accurately predict unseen future images even for large interframe...

The interaction between vincamine (VCM) and human serum albumin (HSA) has been studied using a fluorescence quenching technique in combination with UV/vis absorption spectroscopy, Fourier transform infrared (FT-IR) spectroscopy, circular dichroism (CD) spectroscopy and molecular modeling under conditions similar to human physiological conditions. VCM effectively quenched the intrinsic fluorescence of HSA via static quenching. The binding constants were calculated from the fluorescence data. Thermodynamic analysis by Van't Hoff equation revealed enthalpy change (ΔH) and entropy change (ΔS) were -4.57 kJ/mol and 76.26 J/mol/K, respectively, which indicated that the binding process was spontaneous and the hydrophobic interaction was the predominant force. The distance r between the donor (HSA) and acceptor (VCM) was obtained according to the Forster's theory of non-radiative energy transfer and found to be 4.41 nm. Metal ions, viz., Na(+), K(+), Li(+), Ni(2+), Ca(2+), Zn(2+) and Al(3+) were found to influence binding of the drug to protein. The 3D fluorescence, FT-IR and CD spectral results revealed changes in the secondary structure of the protein upon interaction with VCM. Furthermore, molecular modeling indicated that VCM could bind to the subdomain IIA (site I) of HSA.

Higher fullerenes (C84, C90, C92, C94 and C96) were successfully isolated from the Soxhlet extract of the synthetical “Graphote Smokes” soot (GS sample) by using a big Cosmosil Buckysep (Phenomenex) column (250 mm × 10 mm) with a large injection. The fractions isolated have been determined by high performance liquid chromatography (HPLC) and laser desorption mass spectrometry (LDMS). It is found that there are different fullerenes molecules in different fractions with retention time. The result indicates that fullerenes do exist in GS samples. Also, it excludes the suspicion to some extent that fullerene molecules might be generated by the laser desorption process in the LDMS. In addition, it also provides the experimental basis for the study of natural higher fullerenes and might be helpful to figure out the question if higher fullerenes do exist in the natural samples.