GraphChainer: Co-linear Chaining for Accurate Alignment of Long Reads to Variation Graphs

Aligning reads to a variation graph is a standard task in pangenomics, with downstream applications in e.g., improving variant calling. While the vg toolkit (Garrison et al., Nature Biotechnology, 2018) is a popular aligner of short reads, GraphAligner (Rautiainen and Marschall, Genome Biology, 2020) is the state-of-the-art aligner of long reads. GraphAligner works by finding candidate read occurrences based on individually extending the best seeds of the read in the variation graph. However, a more principled approach recognized in the community is to co-linearly chain multiple seeds. We present a new algorithm to co-linearly chain a set of seeds in an acyclic variation graph, together with the first efficient implementation of such a co-linear chaining algorithm into a new aligner of long reads to variation graphs, GraphChainer. Compared to GraphAligner, at a normalized edit distance threshold of 40%, it aligns 9% to 12% more reads, and 15% to 19% more total read length, on real PacBio reads from human chromosomes 1 and 22. On both simulated and real data, GraphChainer aligns between 97% and 99% of all reads, and of total read length. At the more stringent normalized edit distance threshold of 30%, GraphChainer aligns up to 29% more total real read length than GraphAligner. GraphChainer is freely available at https://github.com/algbio/GraphChainer

Generative hypergraph clustering: From blockmodels to modularity

Hypergraphs are a natural modeling paradigm for networked systems with multiway interactions. A standard task in network analysis is the identification of closely related or densely interconnected nodes. We propose a probabilistic generative model of clustered hypergraphs with heterogeneous node degrees and edge sizes. Approximate maximum likelihood inference in this model leads to a clustering objective that generalizes the popular modularity objective for graphs. From this, we derive an inference algorithm that generalizes the Louvain graph community detection method, and a faster, specialized variant in which edges are expected to lie fully within clusters. Using synthetic and empirical data, we demonstrate that the specialized method is highly scalable and can detect clusters where graph-based methods fail. We also use our model to find interpretable higher-order structure in school contact networks, U.S. congressional bill cosponsorship and committees, product categories in copurchasing behavior, and hotel locations from web browsing sessions.

Improvement in Task Scheduling Capabilities for SaaS Cloud Deployments Using Intelligent Schedulers

Task scheduling on the cloud involves processing a large set of variables from both the task side and the scheduling machine side. This processing often results in a computational model that produces efficient task to machine maps. The efficiency of such models is decided based on various parameters like computational complexity, mean waiting time for the task, effectiveness to utilize the machines, etc. In this paper, a novel Q-Dynamic and Integrated Resource Scheduling (DAIRS-Q) algorithm is proposed which combines the effectiveness of DAIRS with Q-Learning in order to reduce the task waiting time, and improve the machine utilization efficiency. The DAIRS algorithm produces an initial task to machine mapping, which is optimized with the help of a reward & penalty model using Q-Learning, and a final task-machine map is obtained. The performance of the proposed algorithm showcases a 15% reduction in task waiting time, and a 20% improvement in machine utilization when compared to DAIRS and other standard task scheduling algorithms.

Alcohol unleashes homo economicus by inhibiting cooperation

Human behavior lies somewhere between purely self-interested homo economicus and socially-motivated homo reciprocans. The factors that cause people to choose self-interest over costly cooperation can provide insights into human nature and are essential when designing institutions and policies that are meant to influence behavior. Alcohol consumption can shed light on the inflection point between selfish and selfless because it is commonly consumed and has global effects on the brain. The present study administered alcohol or placebo (N = 128), titrated to sex and weight, to examine its effect on cooperation in a standard task in experimental economics, the public goods game (PGG). Alcohol, compared to placebo, doubled the number of free-riders who contributed nothing to the public good and reduced average PGG contributions by 32% (p = .005). This generated 64% higher average profits in the PGG for those who consumed alcohol. The degree of intoxication, measured by blood alcohol concentration, linearly reduced PGG contributions (r = -0.18, p = .05). The reduction in cooperation was traced to a deterioration in mood and an increase in physiologic stress as measured by adrenocorticotropic hormone. Our findings indicate that moderate alcohol consumption inhibits the motivation to cooperate and that homo economicus is stressed and unhappy.

Determination of normal flow depth in cas maxima system

The development of personal mobile microprocessor gadgets, computer mathematics systems and interactive online supplements allow victoriousness in the initial process by algorithms of symbolic mathematics, numerical methods and hard functional graphical. The article shows a choice of options for engineering development of a standard task of assigning a normal depth in line with a trapezoidal living process. Discernible shortcomings of symbolic and numerical algorithms in the development of tasks in the CAS MAXIMA system. The article presents a visualization and method of a simple iterative solution of tasks. An assessment of the accuracy of the result was carried out using the graphical method. In parallel, the solution of the tasks is taken from the Web-interface to the on-line service of the CAS MAXIMA system on the CESGA server.

What Explains the Socioeconomic Status-Health Gradient? Evidence from Workplace COVID-19 Infections

This paper studies the contribution of the workplace to the SES-health gradient. Our analysis is based on a unique dataset that tracks various health outcomes and workplace risks among healthcare workers during the first four months of the coronavirus 2019 (COVID-19) pandemic. The setting provides an exceptional opportunity to test for work-related disparities in health, while controlling for confounding determinants of the SES-health gradient. We find that low-SES nurses were systematically more likely to contract COVID-19 as a result of workplace exposure. These differentials existed in all healthcare institutions, but were particularly large in non-hospital settings. In contrast, we find no relationship between SES and non work-related infection rates. The differences in workplace infection rates are substantially larger than those implied by standard 'task-based' indices of transmission risk, and cannot be attributable to easily identifiable metrics of workplace risk. Together, our results show how subtle differences in work conditions or job duties can substantially contribute to the SES-health gradient.

A Kind of Reinforcement Learning to Improve Genetic Algorithm for Multiagent Task Scheduling

It is difficult to coordinate the various processes in the process industry. We built a multiagent distributed hierarchical intelligent control model for manufacturing systems integrating multiple production units based on multiagent system technology. The model organically combines multiple intelligent agent modules and physical entities to form an intelligent control system with certain functions. The model consists of system management agent, workshop control agent, and equipment agent. For the task assignment problem with this model, we combine reinforcement learning to improve the genetic algorithm for multiagent task scheduling and use the standard task scheduling dataset in OR-Library for simulation experiment analysis. Experimental results show that the algorithm is superior.

Gamifying Questions by Adding Contextin the Creative Development Marketing Research

Gamifi cation is defi ned as using game design elements in non-gaming contexts: education, management, marketing and also market research. Gamifi ed research tools help to increase respondents’ engagement and obtain more in-depth results. Up till now the eff ects of gamifi cations were tested in the domains of brand strategy and consumer experience. The article shows the results of the experiment proving the eff ectiveness of a gamifi ed approach to the qualitative advertisement testing. The experimental group with a narrative context added to a question regarding the fi rst impression performed better than the control group with a standard task. Also gender diff erences were observed: the eff ect was valid only for men – there were no signifi cant diff erences in the performance of women in both groups. Due to an uneven split of men and women and a small sample in general, this eff ect needs further examination.

EXPERIMENTAL ANALYSIS OF THE HELICOPTER DITCHING

The paper presents the method and results of the model tests of helicopter ditching, aimed at adaptation of the helicopter's construction for marine mis-sions. The experiment, realized by Maritime Advanced Research Centre S.A. (CTO) required elaboration of dedicated measurement stand, and solving a number of specific technical problems resulting from the necessity of assuring the repeatability, and required accuracy of the measurements and scalability of the results. Such kind of experiment is not a standard task of a hydrodynamic model testing institution, so it brings an innovation in the testing methodology. The paper presents the details of the test stand and the model itself with rotating rotor, design of the experiment, as well as an overview of the results and main conclusions

INVESTIGATIONS ON SKIP-CONNECTIONS WITH AN ADDITIONAL COSINE SIMILARITY LOSS FOR LAND COVER CLASSIFICATION

Pixel-based land cover classification of aerial images is a standard task in remote sensing, whose goal is to identify the physical material of the earth’s surface. Recently, most of the well-performing methods rely on encoder-decoder structure based convolutional neural networks (CNN). In the encoder part, many successive convolution and pooling operations are applied to obtain features at a lower spatial resolution, and in the decoder part these features are up-sampled gradually and layer by layer, in order to make predictions in the original spatial resolution. However, the loss of spatial resolution caused by pooling affects the final classification performance negatively, which is compensated by skip-connections between corresponding features in the encoder and the decoder. The most popular ways to combine features are element-wise addition of feature maps and 1x1 convolution. In this work, we investigate skip-connections. We argue that not every skip-connections are equally important. Therefore, we conducted experiments designed to find out which skip-connections are important. Moreover, we propose a new cosine similarity loss function to utilize the relationship of the features of the pixels belonging to the same category inside one mini-batch, i.e. these features should be close in feature space. Our experiments show that the new cosine similarity loss does help the classification. We evaluated our methods using the Vaihingen and Potsdam dataset of the ISPRS 2D semantic labelling challenge and achieved an overall accuracy of 91.1% for both test sites.