CNNDLP: A Method Based on Convolutional Autoencoder and Convolutional Neural Network with Adjacent Edge Attention for Predicting lncRNA–Disease Associations

It is well known that the unusual expression of long non-coding RNAs (lncRNAs) is closely related to the physiological and pathological processes of diseases. Therefore, inferring the potential lncRNA–disease associations are helpful for understanding the molecular pathogenesis of diseases. Most previous methods have concentrated on the construction of shallow learning models in order to predict lncRNA-disease associations, while they have failed to deeply integrate heterogeneous multi-source data and to learn the low-dimensional feature representations from these data. We propose a method based on the convolutional neural network with the attention mechanism and convolutional autoencoder for predicting candidate disease-related lncRNAs, and refer to it as CNNDLP. CNNDLP integrates multiple kinds of data from heterogeneous sources, including the associations, interactions, and similarities related to the lncRNAs, diseases, and miRNAs. Two different embedding layers are established by combining the diverse biological premises about the cases that the lncRNAs are likely to associate with the diseases. We construct a novel prediction model based on the convolutional neural network with attention mechanism and convolutional autoencoder to learn the attention and the low-dimensional network representations of the lncRNA–disease pairs from the embedding layers. The different adjacent edges among the lncRNA, miRNA, and disease nodes have different contributions for association prediction. Hence, an attention mechanism at the adjacent edge level is established, and the left side of the model learns the attention representation of a pair of lncRNA and disease. A new type of lncRNA similarity and a new type of disease similarity are calculated by incorporating the topological structures of multiple bipartite networks. The low-dimensional network representation of the lncRNA-disease pairs is further learned by the autoencoder based convolutional neutral network on the right side of the model. The cross-validation experimental results confirm that CNNDLP has superior prediction performance compared to the state-of-the-art methods. Case studies on stomach cancer, breast cancer, and prostate cancer further show the ability of CNNDLP for discovering the potential disease lncRNAs.

AG-Index: Adjacent Edge Hash Index for Graph Databases

Many queries have been proposed to retrieve graphs. Among others, subgraph query is a fundamental one: given a graph database and a query graph, find the graphs in the database containing the query graph. Most existing works follow the filtering-and-verification framework, where a core task is to reduce the number of candidate graphs. This paper follows the framework and we propose a novel index, namely AG-Index. It indexes adjacent edge pairs of data graphs and can significantly reduce the number of candidate graphs. Our experiments show that our AG-Index outperforms several existing techniques on real-life datasets and synthetic datasets.

Greedy Random Walk

We study a discrete time self-interacting random process on graphs, which we call greedy random walk. The walker is located initially at some vertex. As time evolves, each vertex maintains the set of adjacent edges touching it that have not yet been crossed by the walker. At each step, the walker, being at some vertex, picks an adjacent edge among the edges that have not traversed thus far according to some (deterministic or randomized) rule. If all the adjacent edges have already been traversed, then an adjacent edge is chosen uniformly at random. After picking an edge the walker jumps along it to the neighbouring vertex. We show that the expected edge cover time of the greedy random walk is linear in the number of edges for certain natural families of graphs. Examples of such graphs include the complete graph, even degree expanders of logarithmic girth, and the hypercube graph. We also show that GRW is transient in$\mathbb{Z}^d$for alld≥ 3.

Direct Numerical Simulation of Single and Multiple Square Jets in Cross-Flow

Direct numerical simulations (DNSs) have been carried out for single and multiple square jets issuing normally into a cross-flow, with the primary aim of studying the flow structures and interaction mechanisms associated with the jet in cross-flow (JICF) problems. The single JICF configuration follows a similar study previously done by Sau et al. (2004, Phys. Rev. E, 69, p. 066302) and the multiple JICF configurations are arranged side-by-side in the spanwise direction with a jet-to-jet adjacent edge distance (H) for the twin-jet case and an additional third jet downstream along the centerline with a jet-to-jet adjacent edge distance (L) for the triple-jet case. Simulations are performed for two twin-jet cases with H=1D,2D, respectively, and for one triple-jet case with H=1D, L=2D, where D is the jet exit width. Flow conditions similar to Sau et al. are considered, i.e., the jet to the cross-flow velocity ratio R=2.5 and the Reynolds number 225, based on the freestream velocity and the jet exit width. For the single jet in cross-flow, the vortical structures from our DNS are in good qualitative agreement with the findings of Sau et al. For the side-by-side twin-jet configuration, results have shown that the merging process of the two initially separated counter-rotating vortex pairs (CRVPs) from each jet hole exit is strongly dependent on the jet-to-jet adjacent edge distance H with earlier merging observed for the case H=1D. Downstream, the flow is dominated by a larger CRVP structure, accompanied by a smaller inner vortex pair. The inner vortex pair is found not to survive in the far-field as it rapidly dissipates before exiting the computational domain. These observations are in good agreement with the experimental findings in the literature. Simulations of the triple-jet in cross-flow case have shown some complicated jet-jet and jet-cross-flow interactions with three vortex pairs observed downstream, significantly different from that seen in the twin-jet cases. The evidence of these flow structures and interaction characteristics could provide a valuable reference database for future in-depth flow physics studies of laboratory experimental and numerical investigations.

Theoretical morphology of the crinoid cup

Two simple plate parameters, P, the height of the plate measured normal to the plate base, and α, the angle formed between the plate base and the adjacent edge of that plate, serve to model crinoid aboral cup morphology. With few exceptions, the resulting theoretical geometries replicate the range of calyx morphology observed in the natural world. A theoretical morphospace, derived from these parameters, encompasses both the realized and unrealized possibilities of crinoid calyx construction. The model and the associated morphospace demonstrate that the occupation of crinoid cup space varies non-uniformly in time and space and suggest that functional constraints and/or ecological habit are important components of the distribution of cup morphology in time.

Studies of the Byron Bog in Southwestern Ontario: IX. Insects Trapped as Adults Emerging from Redmond's Pond

In the description of the Byron Bog (Judd, 1957a) it was pointed out that Redmond's Pond is situated in the northwest corner of the bog and that during 1956 a tent-trap was anchored in a small bay in the northeast corner of the pond to trap insects emerging as adults from the water. The position of the trap on the pond is shown in the map accompanying the description of the bog (Judd, 1957a) and the structure and use of the trap are also described by Judd (1957b). The trap was placed on the water on May 15, 1956 and remained there until November 8. It was about four feet from the edge of the pond in water about two feet deep. At this point the bottom of the pond was composed of a thick layer of loose, brown peat and the adjacent edge of the pond was occupied by a dense growth of leatherleaf, Chamaedaphne calyculata, growing in Sphagnum (Judd, 1957a). The branches of the bushes of leatherleaf extended out over the water of the small bay in which the trap floated. The only rooted plant growing in and around the trap was spatterdock, Nuphar advena. Floating in the water was a sparse growth of bladderwort, Utricularia vulgaris, and on the surface of the water there were a few scattered fronds of duckweed, Lemna minor, and water flax-seed, Spirodela polyrhiza.