Westfield State University

Monroe J.G., Srikant T., Carbonell-Bejerano P., Becker C., Lensink M., Exposito-Alonso M., Klein M., Hildebrandt J., Neumann M., Kliebenstein D., Weng M., Imbert E., Ågren J., Rutter M.T., Fenster C.B., et. al.

Since the first half of the twentieth century, evolutionary theory has been dominated by the idea that mutations occur randomly with respect to their consequences1. Here we test this assumption with large surveys of de novo mutations in the plant Arabidopsis thaliana. In contrast to expectations, we find that mutations occur less often in functionally constrained regions of the genome—mutation frequency is reduced by half inside gene bodies and by two-thirds in essential genes. With independent genomic mutation datasets, including from the largest Arabidopsis mutation accumulation experiment conducted to date, we demonstrate that epigenomic and physical features explain over 90% of variance in the genome-wide pattern of mutation bias surrounding genes. Observed mutation frequencies around genes in turn accurately predict patterns of genetic polymorphisms in natural Arabidopsis accessions (r = 0.96). That mutation bias is the primary force behind patterns of sequence evolution around genes in natural accessions is supported by analyses of allele frequencies. Finally, we find that genes subject to stronger purifying selection have a lower mutation rate. We conclude that epigenome-associated mutation bias2 reduces the occurrence of deleterious mutations in Arabidopsis, challenging the prevailing paradigm that mutation is a directionless force in evolution. Data on de novo mutations in Arabidopsis thaliana reveal that mutations do not occur randomly; instead, epigenome-associated mutation bias reduces the occurrence of deleterious mutations.

Santangelo J.S., Ness R.W., Cohan B., Fitzpatrick C.R., Innes S.G., Koch S., Miles L.S., Munim S., Peres-Neto P.R., Prashad C., Tong A.T., Aguirre W.E., Akinwole P.O., Alberti M., Álvarez J., et. al.

Urbanization transforms environments in ways that alter biological evolution. We examined whether urban environmental change drives parallel evolution by sampling 110,019 white clover plants from 6169 populations in 160 cities globally. Plants were assayed for a Mendelian antiherbivore defense that also affects tolerance to abiotic stressors. Urban-rural gradients were associated with the evolution of clines in defense in 47% of cities throughout the world. Variation in the strength of clines was explained by environmental changes in drought stress and vegetation cover that varied among cities. Sequencing 2074 genomes from 26 cities revealed that the evolution of urban-rural clines was best explained by adaptive evolution, but the degree of parallel adaptation varied among cities. Our results demonstrate that urbanization leads to adaptation at a global scale.

Shan L., Liu Y., Tang M., Yang M., Bai X.

Well logging is a significant method of formation description and resource assessment in exploration and development of oil, natural gas, minerals, groundwater, and sub-surface thermal energy, as well as geotechnical engineering and environmental research. However, the shortage problem of well logging data always exists because well logs can only be measured through a drilling process involving costly and time-consuming field trials. To address this issue, bidirectional long short-term memory (BiLSTM), attention mechanism, and convolutional neural network (CNN) were coupled to build hybrid neural networks for predicting missing well logs. The proposed architecture is a structure of two branches. One branch uses CNN to capture the spatial properties of well logs, and the other one conducts the feature selections by utilizing two-layer BiLSTM with attention mechanism. The spatio-temporal correlations from two branches are merged to forecast the target well logs. The performance of the proposed method is evaluated within a highly heterogeneous reservoir at the Gangdong oilfield in China. In our experiments, six models were trained and used for generating synthetic well logs including compensated neutron logs (CNL), acoustic (AC), spontaneous potential (SP), gamma-ray (GR), density (DEN), and formation resistivity (RT). Moreover, traditional machine learning models, CNN, BiLSTM, and other deep learning benchmark models were developed to compare with the presented models. Results show that the proposed method achieves higher prediction accuracy because it takes into account the spatio-temporal information of well logs. • Hybrid neural network models for the well logging prediction in heterogeneous reservoirs were developed. • The proposed architecture considers the effects of the spatial and temporal properties of well logs in the feature selection. • The proposed method is superior to deep learning benchmarks and traditional machine learning methods in predicting well logs. • Our method can achieve a good fitting degree between the predicted and actual measurements of all kinds of well logs.

Xia W., Wang S., Yang M., Gao Q., Han J., Gao X.

Multi-view clustering has become an active topic in artificial intelligence. Yet, similar investigation for graph-structured data clustering has been absent so far. To fill this gap, we present a Multi-View Graph embedding Clustering network (MVGC). Specifically, unlike traditional multi-view construction methods, which are only suitable to describe Euclidean structure data, we leverage Euler transform to augment the node attribute, as a new view descriptor, for non-Euclidean structure data. Meanwhile, we impose block diagonal representation constraint, which is measured by the ℓ1,2-norm, on self-expression coefficient matrix to well explore the cluster structure. By doing so, the learned view-consensus coefficient matrix well encodes the discriminative information. Moreover, we make use of the learned clustering labels to guide the learnings of node representation and coefficient matrix, where the latter is used in turn to conduct the subsequent clustering. In this way, clustering and representation learning are seamlessly connected, with the aim to achieve better clustering performance. Extensive experimental results indicate that MVGC is superior to 11 state-of-the-art methods on four benchmark datasets. In particular, MVGC achieves an Accuracy of 96.17% (53.31%) on the ACM (IMDB) dataset, which is an up to 2.85% (1.97%) clustering performance improvement compared with the strongest baseline.

Yang H., Gao Q., Xia W., Yang M., Gao X.

Multi-view spectral clustering has become appealing due to its good performance in capturing the correlations among all views. However, on one hand, many existing methods usually require a quadratic or cubic complexity for graph construction or eigenvalue decomposition of Laplacian matrix; on the other hand, they are inefficient and unbearable burden to be applied to large scale data sets, which can be easily obtained in the era of big data. Moreover, the existing methods cannot encode the complementary information between adjacency matrices, i.e. , similarity graphs of views and the low-rank spatial structure of adjacency matrix of each view. To address these limitations, we develop a novel multi-view spectral clustering model. Our model well encodes the complementary information by Schatten $p$ -norm regularization on the third tensor whose lateral slices are composed of the adjacency matrices of the corresponding views. To further improve the computational efficiency, we leverage anchor graphs of views instead of full adjacency matrices of the corresponding views, and then present a fast model that encodes the complementary information embedded in anchor graphs of views by Schatten $p$ -norm regularization on the tensor bipartite graph. Finally, an efficient alternating algorithm is derived to optimize our model. The constructed sequence was proved to converge to the stationary KKT point. Extensive experimental results indicate that our method has good performance.

Lv Z., Gao Q., Zhang X., Li Q., Yang M.

In this article, we present a novel general framework for incomplete multi-view clustering by integrating graph learning and spectral clustering. In our model, a tensor low-rank constraint are introduced to learn a stable low-dimensional representation, which encodes the complementary information and takes into account the cluster structure between different views. A corresponding algorithm associated with augmented Lagrangian multipliers is established. In particular, tensor Schatten $p$ -norm is used as a tighter approximation to the tensor rank function. Besides, both consistency and specificity are jointly exploited for subspace representation learning. Extensive experiments on benchmark datasets demonstrate that our model outperforms several baseline methods in incomplete multi-view clustering.

Fetters A.M., Cantalupo P.G., Wei N., Robles M.T., Stanley A., Stephens J.D., Pipas J.M., Ashman T.

Pollen is a unique vehicle for viral spread. Pollen-associated viruses hitchhike on or within pollen grains and are transported to other plants by pollinators. They are deposited on flowers and have a direct pathway into the plant and next generation via seeds. To discover the diversity of pollen-associated viruses and identify contributing landscape and floral features, we perform a species-level metagenomic survey of pollen from wild, visually asymptomatic plants, located in one of four regions in the United States of America varying in land use. We identify many known and novel pollen-associated viruses, half belonging to the Bromoviridae, Partitiviridae, and Secoviridae viral families, but many families are represented. Across the regions, species harbor more viruses when surrounded by less natural and more human-modified environments than the reverse, but we note that other region-level differences may also covary with this. When examining the novel connection between virus richness and floral traits, we find that species with multiple, bilaterally symmetric flowers and smaller, spikier pollen harbored more viruses than those with opposite traits. The association of viral diversity with floral traits highlights the need to incorporate plant-pollinator interactions as a driver of pollen-associated virus transport into the study of plant-viral interactions. Pollen can be a vehicle for viral spread among plants. Here, Fetters et al. apply viral metagenomics to characterize the pollen virome of a diverse set of wild plants, find known and previously un-known viruses and show that wild plant species harbor more viruses when surrounded by less natural vegetation and when they have traits that promote increased plant-pollinator vector interactions.

Shu X., Zhang X., Gao Q., Yang M., Wang R., Gao X.

Graph-based multi-view clustering method has attracted considerable attention in multi-media data analyse community due to its good clustering performance and efficiency in characterizing the relationship between data. But the existing graph-based clustering methods still have many shortcomings. Firstly, they have high computational complexity due to the eigenvalue decomposition. Secondly, the complementary information and spatial structure embedded in different views can affect the clustering performance. However, some existing graph-based clustering methods do not consider these two points. In this article, we use the anchor graphs of different views as input, which effectively reduces the computational complexity. And then we explicitly consider the complementary information and spatial structure between anchor graphs of different views by minimizing the tensor Schatten $p$-norm, aiming to achieve a better tensor with low-rank approximation. Finally, we learn the view-consensus anchor graph with connectivity constraints, which can directly indicate clusters by self-weighted strategy. An efficient alternating algorithm is then derived to optimize the proposed multi-view special clustering model. Furthermore, the constructed sequence was proved to converge to the stationary KKT point. Experiments show that our proposed method not only reduces the time cost, but also outperforms the most advanced methods.

ALESSIO H.M., BASSETT D.R., BOPP M.J., PARR B.B., PATCH G.S., RANKIN J.W., ROJAS-RUEDA D., ROTI M.W., WOJCIK J.R.

Active transportation is defined as self-propelled, human-powered transportation modes, such as walking and bicycling. In this article, we review the evidence that reliance on gasoline-powered transportation is contributing to global climate change, air pollution, and physical inactivity and that this is harmful to human health. Global climate change poses a major threat to human health and in the future could offset the health gains achieved over the last 100 yr. Based on hundreds of scientific studies, there is strong evidence that human-caused greenhouse gas emissions are contributing to global climate change. Climate change is associated with increased severity of storms, flooding, rising sea levels, hotter climates, and drought, all leading to increased morbidity and mortality. Along with increases in atmospheric CO2, other pollutants such as nitrogen dioxide, ozone, and particulate matter (e.g., PM2.5) are released by combustion engines and industry, which can lead to pulmonary and cardiovascular diseases. Also, as car ownership and vehicle miles traveled have increased, the shift toward motorized transport has contributed to physical inactivity. Each of these global challenges has resulted in, or is projected to result in, millions of premature deaths each year. One of the ways that nations can mitigate the health consequences of climate change, air pollution, and chronic diseases is through the use of active transportation. Research indicates that populations that rely heavily on active transportation enjoy better health and increased longevity. In summary, active transportation has tremendous potential to simultaneously address three global public health challenges of the 21st century.

Ngo A., Duong H., Nguyen T., Nguyen L.

This paper examines the effects of ownership structure on dividend policy, specifically the role of controlling shareholders in shaping dividend policy in a sample of firms that pay dividends and issue new equity simultaneously. The results show that managers in weakly governed firms are more likely to initiate customized dividends to meet outside large shareholders' needs while simultaneously using costly external capital to finance new investment projects. This paper contributes to the existing literature on agency problems by explaining why firms engage in this suboptimal dividend policy: it allows large shareholders to extract private benefits.