Wearable devices, such as smartwatches, have been widely used by the public for health monitoring, exercise tracking, and message alerts. Regarding localization applications, although smartwatches have been explored by many researchers, current solutions typically require combining data from multiple sensors or multiple mobile devices for reliable position estimation, which is inconvenient for practical usage. In this paper, we propose a novel indoor localization method that relies solely on an inertial measurement unit (IMU) of a smartwatch. First, we design a smartwatch-based neural inertial odometry (WNIO) that adapts to multiple motion patterns and achieves low-drift dead reckoning. Second, we detect interactions between the arm and the environment, such as opening a door, to formulate absolute position measurements that help constrain error drift. A multi-hypothesis Kalman filter (MHKF) is employed to ensure reliable matching and localization. We conducted real-world experiments to validate the effectiveness of the proposed method. The results demonstrate that our method achieves reliable dead reckoning across various motion patterns, including walking and running at different speeds. Furthermore, the method effectively mitigates error drift without relying on any external signals. The video (BiliBili) is also shared to display our research.

Round-Trip Time (RTT) -based Wi-Fi indoor positioning technology is a critical component of the Internet of Things (IoT), enabling various location-based services in smart buildings. However, the positioning accuracy of RTT will be significantly degraded in environments with severe non-line-of-sight (NLOS) reception and a limited number of access points (APs). To solve this challenge, we propose an indoor positioning method based on visibility matching. Compared to the RTT, which is easily affected by NLOS reception, AP visibility is a new positioning feature determined by RTT and signal strength. Additional constraints can be imposed on the positioning system based on the consistency between the observed visibility and the simulated visibility. To improve the accuracy of RTT ranging, a multi-wall ranging model is adopted that considers the influence of signal penetration through walls. The proposed positioning method combines the refined RTT ranging result and visibility constraint. It mitigates the impact of NLOS receptions on positioning accuracy through the multi-wall ranging model and makes full use of the characteristics of NLOS signal through visibility matching. In different experiments, the proposed method improves median positioning accuracy from 2.3 meters to 0.64 meters compared to other methods.

Wi-Fi is vulnerable to non-line-of-sight (NLOS) propagation and multipath effects in indoor positioning. This paper proposes a novel access point (AP) visibility matching method for Wi-Fi round-trip-time (RTT) ranging, aiming at reducing positioning errors in complex NLOS indoor environments. This method identifies NLOS signals to determine observation visibility based on the RTT and received signal strength indicator (RSSI). This method also projects the floor plan onto the polar coordinate system to derive simulation visibility. By comparing the consistency between simulation and observation visibility, the method pinpoints where users are most likely located. Experimental evaluations demonstrate that the proposed method achieves positioning accuracy comparable to fingerprinting, without the need for extensive pre-work.

Single object tracking (SOT) is one of the most active research directions in the field of computer vision. Compared with the 2-D image-based SOT which has already been well-studied, SOT on 3-D point clouds is a relatively emerging research field. In this article, a novel approach, namely, the contextual-aware tracker (CAT), is investigated to achieve a superior 3-D SOT through spatially and temporally contextual learning from the LiDAR sequence. More precisely, in contrast to the previous 3-D SOT methods merely exploiting point clouds in the target bounding box as the template, CAT generates templates by adaptively including the surroundings outside the target box to use available ambient cues. This template generation strategy is more effective and rational than the previous area-fixed one, especially when the object has only a small number of points. Moreover, it is deduced that LiDAR point clouds in 3-D scenes are often incomplete and significantly vary from frame to another, which makes the learning process more difficult. To this end, a novel cross-frame aggregation (CFA) module is proposed to enhance the feature representation of the template by aggregating the features from a historical reference frame. Leveraging such schemes enables CAT to achieve a robust performance, even in the case of extremely sparse point clouds. The experiments confirm that the proposed CAT outperforms the state-of-the-art methods on both the KITTI and NuScenes benchmarks, achieving 3.9% and 5.6% improvements in terms of precision.

Data-intensive application workflows benefit by leveraging cloud services to decrease execution times and increase data sharing. Cloud service providers (CSPs) have distinct capabilities and policies, and performance/cost of the cloud services are amongst the prime factors for CSP selection. However, workflow users who need brokering of cloud resources often lack expert guidance to handle the problem of overwhelming choice in CSP selection, and optimization to compensate for service dynamics. In this paper, we address the optimal resource selection problem using a multi-cloud resource broker viz., OnTimeURB that uses knowledge-engineering of user requirements and service capabilities across multiple CSPs. OnTimeURB is powered by integer linear programming and a Naive Bayes classifier to recommend optimal cloud template solutions by weighting performance, agility, cost, and security (PACS) factors. We evaluate the OnTimeURB recommendations with a catalog of bioinformatics application workflows using four CSP resources featuring more than 300 different instance configurations. Our evaluation results show the efficacy of OnTimeURB in creating consistently cost-effective and agile solutions compared to a state-of-the-art k-nearest neighbors (k-NN) approach. We also show that OnTimeURB has 91% success rate improvement in workflow execution times via cloud template recommendations over approaches that do not use knowledge-engineered multi-CSP resource brokering.

Multi-cloud resource brokering is becoming a critical requirement for applications that require high scale, diversity, and resilience. Applications demand timely selection of distributed data storage and computation platforms that span local private cloud resources as well as resources from multiple cloud service providers (CSPs). The distinct capabilities and policies, as well as performance/cost of the cloud services, are amongst the prime factors for CSP selection. However, application owners who need suitable cyber resources in community/public clouds, often have preliminary knowledge and preferences of certain CSPs. They also lack expert guidance to handle the problem of overwhelming resource choice from CSPs, and optimization to compensate for service dynamics. In this paper, we address this challenge of optimal resource selection while also leveraging limited user's expertise and preferences towards CSPs through multi-level fuzzy logic modeling based on convoluted factors of performance, agility, cost, and security. We evaluate the efficiency of our fuzzy-engineered resource brokering in improving allocation of resources as well as user satisfiability by using case studies and independent validations of CSPs evaluation.

Big data computation tools are vital for researchers and educators from various domains such as plant science, animal science, biomedical science and others. With the growing computational complexity of biology big data, advanced analytic systems, known as prescriptive systems, are being built using machine learning models to intelligently predict optimum computation solutions for users for better data analysis. However, lack of user-friendly prescriptive systems poses a critical roadblock to facilitating informed decision-making by users. In this paper, we detail a formative usability study to address the complexities faced by users while using prescriptive systems. Our usability research approach considers bioinformatics workflows and community cloud resources in the KBCommons framework's science gateway. The results show that recommendations from usability studies performed in iterations during the development of prescriptive systems can improve user experience, user satisfaction and help novice as well as expert users to make decisions in a well-informed manner.

Single nucleotide polymorphisms (SNPs) and insertions/deletions (Indels) are widely spread across all chromosomes of the genome and act as biological markers, which aid in identification of genes associated with traits or phenotypes. With the advances in next-generation sequencing (NGS) technology, large amounts of SNPs and Indels data have become available, making it difficult to effectively perform analysis across multiple samples and intuitively integrate, compare and/or visualize them simultaneously. Genome-wide association studies (GWAS) is a widely used method to find genetic variations associated with a trait, but it lacks an efficient way to investigate genomic variant functions. To tackle these issues, we have developed SNPViz v2.0, a web-based tool to visualize large-scale haplotype blocks with detailed SNPs and Indels grouped by their chromosomal coordinates, along with their overlapping gene models, phenotype to genotype accuracies, Gene Ontology (GO) annotations, protein families (Pfam) annotations, genomic variant annotations, and their functional effects. Moreover, SNPViz v2.0 integrates several large scale soybean SNPs and Indels datasets including G. Soja, GWAS, NAM41, USB-15x, USB-40x, MSMC and Zhou302, available from multiple studies. SNPViz v2.0 is deployed for all organisms and available in both SoyKB and KBCommons frameworks. For soybean data only, the SNPViz 2.0 is publicly available at http://soykb.org/SNPViz2/. For other organisms such as Arabidopsis thaliana, Mus musculus and Zea mays, SNPViz 2.0 is publicly available in their respective knowledge bases at https://kbcommons.org.

The precise mechanism behind treatment resistance in cancer is still not fully understood. Despite advances in precision oncology, there is a lack of tools that help to understand a mechanistic picture of treatment resistance in cancer patients. Existing enrichment methods heavily rely on quantitative data and limited to analysis of differentially expressed genes, ignoring crucial players that might be involved in this process. In order to tackle treatment resistance, the precise identification of deregulated flow of signal transduction is critical. Here, we introduce a bioinformatics framework that is capable of inferring deregulated flow of signal transduction given evidence-based knowledge about pathway topology and patient-specific mutations. While testing the proposed pipeline on a case study, our algorithm was able to confirm findings from biological experiment, where KRAS mutant cells developed treatment resistance to MEK inhibitor. Our model provides a framework for mechanistic understanding of acquired treatment resistance, thus, equipped clinicians with tool for searching more accurate diagnostic clues in patients with non-trivial disease representations.

Scientific workflows due to their data and memory intensive requirements are among the prime applications which benefit by leveraging cloud computing. However, Cloud service providers (CSPs) have distinct policies and service dynamics that present a problem of excess choice for users. Performance and cost of the cloud services are among the principal factors in CSP selection for scientific bioinformatics workflows. The workflows typically are based on private data, and require diverse cloud resources, thus often requiring synergistic services from multiple CSPs. In this paper, we address this challenge of multi-cloud resource selection using cloud template solutions based on user specifications. We propose an optimizer that incorporates a combinatorial optimization model built on performance, cost and CSPs interoperability factors. The optimizer is integrated within a novel resource broker (i.e., OnTimeURB) for prescriptive recommendations of template solutions with intuitive choices for users. We implement and evaluate the OnTimeURB recommendations framework with a catalog of bioinformatics workflow applications integrated within a KBCommons science gateway. The evaluation considered four CSP resources featuring more than 300 different machine configuration instances. Our evaluation results show that our OnTimeURB creates consistently more economical, performance optimized and practical cloud solutions compared to a k-nearest neighbors (k-NN) approach.

Data-intensive bioinformatics applications often use federated multi-cloud infrastructures to support compute-intensive processing needs. In this paper, we propose a Multi-Cloud Performance and Security (MCPS) Brokering framework within such federated multi-cloud infrastructures to allocate cloud resources to applications by satisfying their performance and security requirements.

Precision Medicine is one of the most rapidly evolving areas within medicine, which allows utilization of patient's own genomic information in guiding personalized treatments and customized therapy. Applying the public data obtained from human genomic and cancer cell line studies in the diagnosis and treatment of diseases is a challenging task as these data are spread across different repositories with diverse formats and omics data types. There is an immediate need for platforms to make this integration and analysis seamless and readily accessible to researchers and clinicians. To achieve this, we have implemented Cancer Cell Line KB, a branch of the KBCommons for homoSapiens which focuses on automatically establishing a web-based resource for comprehensive multi-omics data access using our in-house developed KBCommons framework. It currently provides data for human genes / proteins, miRNA, and publicly available information for 1046 cancer cell lines and drug response datasets. Additionally, gene expression and de-identified clinical data from The Cancer Genome Atlas has also been added which includes 7706 patient samples for 20 cancer types, and 23368 genes expression as FPKM and FeatureCounts. Users can view these datasets using various analytical and graphical visualization tools in Cancer Cell Line KB and search by genes or cancer cell lines. We have also developed Principal Component Analysis (PCA) tool and patient clinical data browser to provide access to patient's genomic tests information and gene expression. In the future, heatmap will be developed to analysis the microarray gene expression data with hierarchical clustering method and more de-identified clinical data from cBioportal can be imported through WEB-API. We also plan to implement a tool to show how the patient's own genomic tests information compares with the population based datasets for that disease traits.

Multi-level ‘OMICS’ data integration for multiple organisms has been one of the major challenges in the era of advanced next generation sequencing and high performance technologies. However, these data are often stored individually across different web resources based on data type and organism making it difficult to find and integrate them. There are many websites which stores different data types and display data in pie charts or plain text format but limit their data to only one fixed organism. Making it difficult for researchers working on other biological organisms including plants, animals, humans, and microbes have similar needs with multi-level omics data. These complex omics data requires extensive data management, exhaustive computational analysis, and effective integration to have a one-stop interactive, web-based portal to browse, access, analyze, integrate, visualize and share knowledge about genomics and molecular mechanisms, with ultimate links to phenotypes and traits. To achieve this, we have developed Knowledge Base Commons (KBCommons), a platform that automates the process of establishing the database and making the tools for organisms available via a dedicated web resource.

Soybean tissue-specific network helps identify and visualize the gene-gene relationships in various tissues [1]. We have built SoyTSN, a web-based tool for tissue-specific network prediction in soybean using 14 tissues RNA-Seq datasets including flower, root, nodule, leaf, stem, seed, etc. SoyTSN first combines multiple tissue specific RNA-Seq studies, and later Cross-Conditions Cluster Detection (C3D) algorithm [2] was applied to detect modules based on co-expression relationships across all the tissues. Following this soybean tissue-specific interactomes were inferred by combining tissue-specific expression and protein-protein interaction from the STRING database [3]. All these relationships between various soybean genes were collected and stored in Soybean Knowledge Base (SoyKB)[4] and can be queried using SoyTSN. For every query gene, SoyTSN computes and visualizes any of the 14 tissue specific networks both at the expression and interactome level. Users can compare the gene-gene relationship differences at different confidence levels across all the soybean tissues.

Advancement of next generation sequencing and high-throughput technologies has resulted in generation of multi-level of `OMICS' data for many organisms. However, these data are often individually scattered across different repositories based on data type, making it difficult to integrate them. We have addressed this issue through our in-house developed Soybean Knowledge Base[1,2](SoyKB) framework, a comprehensive web-based resource. It acts as a centralized repository for soybean multi-omics data, and is equipped with an array of bioinformatics analytical and graphical visualization tools. It is available at http://soykb.org and has proven to be a great success with more than 500 registered users. Users working on other biological organisms including plants, animals and biomedical diseases have similar needs and the developed framework can be expanded to make the visualization and analysis tools function for other organisms, without having to reinvent the wheel. To achieve this we have developed KBCommons, a platform that automates the process of establishing the database and making the tools for other organisms available via a dedicated web resource. It provides information for six entities including genes/proteins, microRNAs/sRNAs, metabolites, SNP, traits as well as plant introduction or strains/populations. It also incorporates several multi-omics datasets including transcriptomics, proteomics, metabolomics, epigenomics, molecular breeding and other types. We have currently expanded KBCommons framework and tools to Zea mays, Arabidopsis, Mus musculus and Homo sapiens. We have integrated various genomics dataset for maize including RNAseq B73 mutants and Tassel meristem from our collaborators. It provides a suite of tools such as the gene/metabolite pathway viewer, Protein Bio-Viewer, heatmaps, scatter plots and hierarchical clustering. It also provides access to PGen, Pegasus analytics workflows developed for genomics variations analysis. It also has suite of tools for differential expression analysis of transcriptomics and other multi-omics datasets including venn diagrams, volcano plots, function enrichment and gene modules.