No abstract

Abstract-Three-dimensional digital terrain models are of fundamental importance in many areas such as the geo-sciences and outdoor robotics. Accurate modeling requires the ability to deal with a varying data density and to balance smoothing against the preservation of discontinuities. The latter is particularly important for robotics applications, as discontinuities that arise, for example, at steps, stairs, or building walls are important features for path planning or terrain segmentation tasks. In this paper, we present an extension of the well-established Gaussian process regression approach that utilizes non-stationary covariance functions to locally adapt to the structure of the terrain data. In this way, we achieve strong smoothing in flat areas and along edges and at the same time preserve edges and corners. The derived model yields predictive distributions for terrain elevations at arbitrary locations and thus allows to fill gaps in the data and to perform conservative predictions in occluded areas.

Live-virus vaccines activate both humoral and cell-mediated immunity, require only a single boosting, and generally provide longer immune protection than killed or subunit vaccines. However, growth of live-virus vaccines must be attenuated to minimize their potential pathogenic effects, and mechanisms of attenuation by conventional serial-transfer viral adaptation are not well-understood. New methods of attenuation based on rational engineering of viral genomes may offer a potentially greater control if one can link defined genetic modifications to changes in virus growth. To begin to establish such links between genotype and growth phenotype, we developed a computer model for the intracellular growth of vesicular stomatitis virus (VSV), a well-studied, nonsegmented, negative-stranded RNA virus. Our model incorporated established regulatory mechanisms of VSV while integrating key wild-type infection steps: hijacking of host resources, transcription, translation, and replication, followed by assembly and release of progeny VSV particles. Generalization of the wild-type model to allow for genome rearrangements matched the experimentally observed attenuation ranking for recombinant VSV strains that altered the genome position of their nucleocapsid gene. Finally, our simulations captured previously reported experimental results showing how altering the positions of other VSV genes has the potential to attenuate the VSV growth while overexpressing the immunogenic VSV surface glycoprotein. Such models will facilitate the engineering of new live-virus vaccines by linking genomic manipulations to controlled changes in virus gene-expression and growth.

No abstract

A common task in the hit-to-lead process is classifying sets of compounds into multiple, usually structural classes, which build the groundwork for subsequent SAR studies. Machine learning techniques can be used to automate this process by learning classification models from training compounds of each class. Gathering class information for compounds can be cost-intensive as the required data needs to be provided by human experts or experiments. This paper studies whether active machine learning can be used to reduce the required number of training compounds. Active learning is a machine learning method which processes class label data in an iterative fashion. It has gained much attention in a broad range of application areas. In this paper, an active learning method for multiclass compound classification is proposed. This method selects informative training compounds so as to optimally support the learning progress. The combination with human feedback leads to a semiautomated interactive multiclass classification procedure. This method was investigated empirically on 15 compound classification tasks containing 86-2870 compounds in 3-38 classes. The empirical results show that active learning can solve these classification tasks using 10-80% of the data which would be necessary for standard learning techniques.