The utility of multilevel models for continuous-time feature selection of spatio-temporal networks

AbstractAutoregressive and moving average models for temporally dynamic networks treat time as a series of discrete steps which assumes even intervals between data measurements and can introduce bias if this assumption is not met. Using real and simulated data from the London Underground network, this paper illustrates the use of continuous time multilevel models to capture temporal trajectories of edge properties without the need for simultaneous measurements, along with two methods for producing interpretable summaries of model results. These including extracting ‘features’ of temporal patterns (e.g. maxima, time of maxima) which have utility in understanding the network properties of each connection and summarising whole-network properties as a continuous function of time which allows estimation of network properties at any time without temporal aggregation of non-simultaneous measurements. Results for temporal pattern features in the response variable were captured with reasonable accuracy. Variation in the temporal pattern features for the exposure variable was underestimated by the models. The models showed some lack of precision. Both model summaries provided clear ‘real-world’ interpretations and could be applied to data from a range of spatio-temporal network structures (e.g. rivers, social networks). These models should be tested more extensively in a range of scenarios, with potential improvements such as random effects in the exposure variable dimension.

Download Full-text

Feature selection of the armature winding broken coils in synchronous motor using genetic algorithm and mahalanobis distance

Archives of Metallurgy and Materials ◽

10.2478/v10172-012-0091-7 ◽

2012 ◽

Vol 57 (3) ◽

pp. 829-835 ◽

Cited By ~ 1

Author(s):

Z. Głowacz ◽

J. Kozik

Keyword(s):

Genetic Algorithm ◽

Feature Selection ◽

Mahalanobis Distance ◽

Distance Measure ◽

Synchronous Motor ◽

Medical Diagnostics ◽

Motor Current ◽

Feature Spaces ◽

Multidimensional Feature Spaces ◽

Selection Of

The paper describes a procedure for automatic selection of symptoms accompanying the break in the synchronous motor armature winding coils. This procedure, called the feature selection, leads to choosing from a full set of features describing the problem, such a subset that would allow the best distinguishing between healthy and damaged states. As the features the spectra components amplitudes of the motor current signals were used. The full spectra of current signals are considered as the multidimensional feature spaces and their subspaces are tested. Particular subspaces are chosen with the aid of genetic algorithm and their goodness is tested using Mahalanobis distance measure. The algorithm searches for such a subspaces for which this distance is the greatest. The algorithm is very efficient and, as it was confirmed by research, leads to good results. The proposed technique is successfully applied in many other fields of science and technology, including medical diagnostics.

Download Full-text

Short-Time Recurrences of Plasmodium vivax Malaria as a Public Health Proxy for Chloroquine-Resistance Surveillance: A Spatio-Temporal Study in the Brazilian Amazon

International Journal of Environmental Research and Public Health ◽

10.3390/ijerph18105061 ◽

2021 ◽

Vol 18 (10) ◽

pp. 5061

Author(s):

Antonio A. S. Balieiro ◽

Andre M. Siqueira ◽

Gisely C. Melo ◽

Wuelton M. Monteiro ◽

Vanderson S. Sampaio ◽

...

Keyword(s):

Plasmodium Vivax ◽

Multilevel Models ◽

Malaria Incidence ◽

Chloroquine Resistance ◽

Spatial Clusters ◽

Temporal Study ◽

Plasmodium Vivax Malaria ◽

Spatio Temporal ◽

Spatially Varying ◽

Short Time

In Brazil, malaria caused by Plasmodium vivax presents control challenges due to several reasons, among them the increasing possibility of failure of P. vivax treatment due to chloroquine-resistance (CQR). Despite limited reports of CQR, more extensive studies on the actual magnitude of resistance are still needed. Short-time recurrences of malaria cases were analyzed in different transmission scenarios over three years (2005, 2010, and 2015), selected according to malaria incidence. Multilevel models (binomial) were used to evaluate association of short-time recurrences with variables such as age. The zero-inflated Poisson scan model (scanZIP) was used to detect spatial clusters of recurrences up to 28 days. Recurrences compose less than 5% of overall infection, being more frequent in the age group under four years. Recurrences slightly increased incidence. No fixed clusters were detected throughout the period, although there are clustering sites, spatially varying over the years. This is the most extensive analysis of short-time recurrences worldwide which addresses the occurrence of P. vivax CQR. As an important step forward in malaria elimination, policymakers should focus their efforts on young children, with an eventual shift in the first line of malaria treatment to P. vivax.

Download Full-text

Spatio-temporal selection of reference genes in the two congeneric species of Glycyrrhiza

Scientific Reports ◽

10.1038/s41598-020-79298-8 ◽

2021 ◽

Vol 11 (1) ◽

Author(s):

Yuping Li ◽

Xiaoju Liang ◽

Xuguo Zhou ◽

Yu An ◽

Ming Li ◽

...

Keyword(s):

Gene Expression ◽

Developmental Stage ◽

Reference Genes ◽

Developmental Stages ◽

Individual Factors ◽

Congeneric Species ◽

Spatio Temporal ◽

The Stability ◽

Different Tissues ◽

Selection Of

AbstractGlycyrrhiza, a genus of perennial medicinal herbs, has been traditionally used to treat human diseases, including respiratory disorders. Functional analysis of genes involved in the synthesis, accumulation, and degradation of bioactive compounds in these medicinal plants requires accurate measurement of their expression profiles. Reverse transcription quantitative real-time PCR (RT-qPCR) is a primary tool, which requires stably expressed reference genes to serve as the internal references to normalize the target gene expression. In this study, the stability of 14 candidate reference genes from the two congeneric species G. uralensis and G. inflata, including ACT, CAC, CYP, DNAJ, DREB, EF1, RAN, TIF1, TUB, UBC2, ABCC2, COPS3, CS, R3HDM2, were evaluated across different tissues and throughout various developmental stages. More importantly, we investigated the impact of interactions between tissue and developmental stage on the performance of candidate reference genes. Four algorithms, including geNorm, NormFinder, BestKeeper, and Delta Ct, were used to analyze the expression stability and RefFinder, a comprehensive software, provided the final recommendation. Based on previous research and our preliminary data, we hypothesized that internal references for spatio-temporal gene expression are different from the reference genes suited for individual factors. In G. uralensis, the top three most stable reference genes across different tissues were R3HDM2, CAC and TUB, while CAC, CYP and ABCC2 were most suited for different developmental stages. CAC is the only candidate recommended for both biotic factors, which is reflected in the stability ranking for the spatio (tissue)-temporal (developmental stage) interactions (CAC, R3HDM2 and DNAJ). Similarly, in G. inflata, COPS3, R3HDM2 and DREB were selected for tissues, while RAN, COPS3 and CS were recommended for developmental stages. For the tissue-developmental stage interactions, COPS3, DREB and ABCC2 were the most suited reference genes. In both species, only one of the top three candidates was shared between the individual factors and their interactions, specifically, CAC in G. uralensis and COPS3 in G. inflata, which supports our overarching hypothesis. In summary, spatio-temporal selection of reference genes not only lays the foundation for functional genomics research in Glycyrrhiza, but also facilitates these traditional medicinal herbs to reach/maximize their pharmaceutical potential.

Download Full-text