A hierarchical multiclass support vector machine incorporated with holistic triple learning units

This paper proposes a new hierarchical learning structure, namely the holistic triple learning (HTL), for extending the binary support vector machine (SVM) to multi-classification problems. For an N-class problem, a HTL constructs a decision tree up to a depth of A leaf node of the decision tree is allowed to be placed with a holistic triple learning unit whose generalisation abilities are assessed and approved. Meanwhile, the remaining nodes in the decision tree each accommodate a standard binary SVM classifier. The holistic triple classifier is a regression model trained on three classes, whose training algorithm is originated from a recently proposed implementation technique, namely the least-squares support vector machine (LS-SVM). A major novelty with the holistic triple classifier is the reduced number of support vectors in the solution. For the resultant HTL-SVM, an upper bound of the generalisation error can be obtained. The time complexity of training the HTL-SVM is analysed, and is shown to be comparable to that of training the one-versus-one (1-vs.-1) SVM, particularly on small-scale datasets. Empirical studies show that the proposed HTL-SVM achieves competitive classification accuracy with a reduced number of support vectors compared to the popular 1-vs-1 alternative.

Restricted gene flow in fragmented populations of a wind-pollinated tree

Fragmentation of natural populations can have negative effects at the genetic level, thus threatening their evolutionary potential. Many of the negative genetic impacts of population fragmentation can be ameliorated by gene flow and it has been suggested that in wind-pollinated tree species, high or even increased levels of gene flow are a feature of fragmented populations, although several studies have disputed this. We have used a combination of nuclear microsatellites and allele-specific PCR (AS-PCR) analysis of chloroplast single nucleotide polymorphisms (SNPs) to examine the levels and patterns of genetic diversity and population differentiation in fragmented populations of juniper (Juniperus communis) in Ireland and inform conservation programs for the species. Significant population differentiation was found for both chloroplast and nuclear markers, indicating restricted gene flow, particularly over larger geographic scales. For conservation purposes, the existence of genetically distinct clusters and geographically localised chloroplast haplotypes suggests that the concept of provenance should be taken into account when formulating augmentation or reintroduction strategies. Furthermore, the potential lack of seed dispersal and seedling establishment means that ex-situ approaches to seed and seedling management may have to be considered.

Repertoire of HLA-DR1-restricted CD4 T cell responses to capsular Caf1 antigen of Y. pestis HLA-transgenic mice

Yersinia pestis is the causative agent of plague, a rapidly fatal infectious disease that has not been eradicated worldwide. The capsular Caf1 protein of Y. pestis is a protective antigen under development as a recombinant vaccine. However, little is known about the specificity of human T cell responses for Caf1. We characterized CD4 T cell epitopes of Caf1 in 'humanized'-HLA-DR1 transgenic mice lacking endogenous MHC class II molecules. Mice were immunized with Caf1 or each of a complete set of overlapping synthetic peptides, and CD4 T cell immunity was measured with respect to proliferative and IFNgamma T cell responses and recognition by a panel of T cell hybridomas, as well as direct determination of binding affinities of Caf1 peptides to purified HLA-DR molecules. Although a number of DR1-restricted epitopes were identified following Caf1 immunization, the response was biased towards a single immunodominant epitope near the C-terminus of Caf1. In addition, potential promiscuous epitopes, including the immunodominant epitope, were identified by their ability to bind multiple common HLA alleles, with implications for the generation of multivalent vaccines against plague for use in humans.

Kinematic modeling of Exechon parallel kinematic machine

The studies on PKMs have attracted a great attention to robotics community. By deploying a parallel kinematic structure, a parallel kinematic machine (PKM) is expected to possess the advantages of heavier working load, higher speed, and higher precision. Hundreds of new PKMs have been proposed. However, due to the considerable gaps between the desired and actual performances, the majorities of the developed PKMs were the prototypes in research laboratories and only a few of them have been practically applied for various applications; among the successful PKMs, the Exechon machine tool is recently developed. The Exechon adopts unique over-constrained structure, and it has been improved based on the success of the Tricept parallel kinematic machine. Note that the quantifiable theoretical studies have yet been conducted to validate its superior performances, and its kinematic model is not publically available. In this paper, the kinematic characteristics of this new machine tool is investigated, the concise models of forward and inverse kinematics have been developed. These models can be used to evaluate the performances of an existing Exechon machine tool and to optimize new structures of an Exechon machine to accomplish some specific tasks.