MOTIVATION: There are high demands for joint genotyping of structural variations with short-read sequencing, but efficient and accurate genotyping in population scale is a challenging task. RESULTS: We developed muCNV that aggregates per-sample summary pileups for joint genotyping of > 100,000 samples. Pilot results show very low Mendelian inconsistencies. Applications to large-scale projects in cloud show the computational efficiencies of muCNV genotyping pipeline. AVAILABILITY: muCNV is publicly available for download at: https://github.com/gjun/muCNV. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

SUMMARY: Although the 1000 Genomes haplotypes are the most commonly used reference panel for imputation, medical sequencing projects are generating large alternate sets of sequenced samples. Imputation in African Americans using 3384 haplotypes from the Exome Sequencing Project, compared with 2184 haplotypes from 1000 Genomes Project, increased effective sample size by 8.3-11.4% for coding variants with minor allele frequency <1%. No loss of imputation quality was observed using a panel built from phenotypic extremes. We recommend using haplotypes from Exome Sequencing Project alone or concatenation of the two panels over quality score-based post-imputation selection or IMPUTE2's two-panel combination. CONTACT: [email protected]. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Both supervised and semisupervised algorithms for hyperspectral data analysis typically assume that all unlabeled data belong to the same set of land-cover classes that is represented by labeled data. This is not true in general, however, since there may be new classes in the unexplored regions within an image or in areas that are geographically near but topographically distinct. This problem is more likely to occur when one attempts to build classifiers that cover wider areas; such classifiers also need to address spatial variations in acquired spectral signatures if they are to be accurate and robust. This paper presents a semisupervised spatially adaptive mixture model (SESSAMM) to identify land covers from hyperspectral images in the presence of previously unknown land-cover classes and spatial variation of spectral responses. SESSAMM uses a nonparametric Bayesian framework to apply spatially adaptive mechanisms to the mixture model with (potentially) infinitely many components. In this method, each component in the mixture has spatially adapted parameters estimated by Gaussian process regression, and spatial correlations between indicator variables are also considered. The proposed SESSAMM algorithm is applied to hyperspectral data from Botswana and from the DC Mall, where some classes are present only in the unlabeled data. SESSAMM successfully differentiates unlabeled instances of previously known classes from unknown classes and provides better results than the standard Dirichlet process mixture model and other alternatives.

Graphs play a role in many semi-supervised learning algorithms, where unlabeled samples are used to find useful structural properties in the data. Dimensionality reduction and regularization based on preserving smoothness over a graph are common in these settings, and they perform particularly well if proximity in the original feature space closely reflects similarity in the classification problem of interest. However, many real-world problem spaces are overwhelmed by noise in the form of features that have no useful relevance to the concept that is being learned. This leads to a lack of robustness in these methods that limits their applicability to new domains. We present a graph-construction method that uses a collection of task-specific random subspaces to promote smoothness with respect to the problem of interest. Application of this method in a graph-based semi-supervised setting demonstrates improvements in both the effectiveness and robustness of the learning algorithms in noisy problem domains.

This paper proposes a novel framework called Gaussian process maximum likelihood for spatially adaptive classification of hyperspectral data. In hyperspectral images, spectral responses of land covers vary over space, and conventional classification algorithms that result in spatially invariant solutions are fundamentally limited. In the proposed framework, each band of a given class is modeled by a Gaussian random process indexed by spatial coordinates. These models are then used to characterize each land cover class at a given location by a multivariate Gaussian distribution with parameters adapted for that location. Experimental results show that the proposed method effectively captures the spatial variations of hyperspectral data, significantly outperforming a variety of other classification algorithms on three different hyperspectral data sets.

Manifold models for nonlinear dimensionality reduction provide useful low-dimensional representations of high-dimensional data. Most manifold models are unsupervised algorithms and map the entire data onto a single manifold. Heterogeneous data with multiple classes are often better modeled by multiple manifolds rather than by a single global manifold, but there is no explicit way to compare instances embedded in different subspaces. We propose a novel low-to-high dimensional mapping using Gaussian processes that offers comparisons in the original space. Based on the mapping, we propose a nearest-manifold classification algorithm for high-dimensional data. Experimental results show that the proposed algorithm provides good classification accuracies for problems well-modeled by multiple manifolds.

Automated classification of land cover types based on hyper-spectral imagery often involves a large geographical area, but class labels are available for only small portions of the entire area. Moreover, the spectral signature of the same land cover class may vary substantially over different locations. When a classifier is trained on a specific geographical location and applied to other areas, it often performs poorly because of such spatial variation of spectral signatures. In this paper, we propose a novel framework for classification of hyper-spectral data: a Gaussian-Process Maximum-Likelihood (GP-ML) model where the mean of each spectral band is spatially modeled using a Gaussian process. Our framework provides a practical and effective way to model spatial variations of high dimensional data such as hyperspectral images for classification problems.

Supervised learners can be used to automatically classify many types of spatially distributed data. For example, land cover classification by hyperspectral image data analysis is an important remote sensing task where a supervised learner is trained on a large set of labeled data. However, while gathering unlabeled samples may be relatively easy, labeling large amounts of data can be very costly. Acting learning is one approach to reduce the amount of labeled data required to build a supervised learner that performs well. However, most active learning approaches assume that the cost of acquiring labels for all points is uniform. For spatially distributed data that requires physical access to spatial locations in order to assign labels, label acquisition costs become proportional to distance traveled in order to label a point. In this paper, we present results for applying a novel active learning method which takes variable label acquisition costs into account on two hyperspectral datasets.

In active learning, one attempts to maximize classifier performance for a given number of labeled training points by allowing the active learning algorithm to choose which points should be labeled.Typically, when the active learner requests labels for the selected points, it assumes that all points require the same amount of effort to label and that the cost of labeling a point is independent of other selected points.In spatially distributed data such as hyperspectral imagery for land-cover classification, the act of labeling a point (i.e., determining the land-type) may involve physically traveling to a location and determining ground truth.In this case, both assumptions about label acquisition costs made by traditional active learning are broken, since costs will depend on physical locations and accessibility of all the visited points.This paper formulates and analyzes the novel problem of performing active learning on spatial data where label acquisition costs are proportional to distance traveled.

We propose an active learning algorithm with knowledge transfer for classification of hyperspectral remote sensing data. The proposed method is based on a previously proposed algorithm, but yields faster learning curves by adjusting distributions of labeled data differently for the old and the new data. With the proposed method, the classifier can effectively transfer its knowledge learned from one region to a spatially or temporally separated region whose spectral signature is different. Empirical evaluation of the proposed algorithm is performed for two different hyperspectal datasets.

Monitoring highway traffic is an important application of computer vision research. In this paper, we analyze congested highway situations where it is difficult to track individual vehicles in heavy traffic because vehicles either occlude each other or are connected together by shadow. Moreover, scenes from traffic monitoring videos are usually noisy due to weather conditions and/or video compression. We present a method that can separate occluded vehicles by tracking movements of feature points and assigning over-segmented image fragments to the motion vector that best represents the fragment's movement. Experiments were conducted on traffic videos taken from highways in Turkey, and the proposed method can successfully separate vehicles in overpopulated and cluttered scenes.

In this paper we present home media center and media clients for multi-room audio and video entertainment and describe the remote user interface protocol, which is based on remote framebuffer (RFB) protocol and remote user interface technology. Users in another room can watch live digital TV programs, view EPG information, utilize PVR functions, enjoy multimedia contents, and access the Internet with full GUI information using low-cost media client connected to home media center via home network, and these media clients can be easily embedded into display devices. Home media center and media clients are UPnP compatible devices.