Hierarchical classification of protein function with ensembles of rules and particle swarm optimisation

This paper focuses on hierarchical classification problems where the classes to be predicted are organized in the form of a tree. The standard top-down divide and conquer approach for hierarchical classification consists of building a hierarchy of classifiers where a classifier is built for each internal (non-leaf) node in the class tree. Each classifier discriminates only between its child classes. After the tree of classifiers is built, the system uses them to classify test examples one class level at a time, so that when the example is assigned a class at a given level, only the child classes need to be considered at the next level. This approach has the drawback that, if a test example is misclassified at a certain class level, it will be misclassified at deeper levels too. In this paper we propose hierarchical classification methods to mitigate this drawback. More precisely, we propose a method called hierarchical ensemble of hierarchical rule sets (HEHRS), where different ensembles are built at different levels in the class tree and each ensemble consists of different rule sets built from training examples at different levels of the class tree. We also use a particle swarm optimisation (PSO) algorithm to optimise the rule weights used by HEHRS to combine the predictions of different rules into a class to be assigned to a given test example. In addition, we propose a variant of a method to mitigate the aforementioned drawback of top-down classification. These three types of methods are compared against the standard top-down hierarchical classification method in six challenging bioinformatics datasets, involving the prediction of protein function. Overall HEHRS with the rule weights optimised by the PSO algorithm obtains the best predictive accuracy out of the four types of hierarchical classification method.     

Agent Technologies for Sensor Networks

Wireless sensor networks are increasingly seen as a solution to the problem of performing continuous wide-area monitoring in many environmental, security, and military scenarios. The distributed nature of such networks and the autonomous behavior expected of them present many novel challenges. In this article, the authors argue that a new synthesis of electronic engineering and agent technology is required to address these challenges, and they describe three examples where this synthesis has succeeded. In more detail, they describe how these novel approaches address the need for communication and computationally efficient decentralized algorithms to coordinate the behavior of physically distributed sensors, how they enable the real-world deployment of sensor agent platforms in the field, and finally, how they facilitate the development of intelligent agents that can autonomously acquire data from these networks and perform information processing tasks such as fusion, inference, and prediction.     

Prosthetic valve endocarditis: superiority of surgical valve replacement versus medical therapy only

The objective of our study was to assess the long-term outcome of patients with prosthetic valve endocarditis. We used a multicenter, prospective, observational study design. Six university teaching hospitals with high volume cardiothoracic surgery participated. Seventy-four patients with prosthetic valve endocarditis as defined by explicit, objective criteria were selected for participation. All patients were followed up prospectively for 1 year. Thirty-one percent and 69% had development of endocarditis within 60 days of valve insertion ("early") and after 60 days ("late"), respectively. The most common causes were Staphylococcus epidermidis (40%), Staphylococcus aureus (20%), streptococcal species (18%), and aerobic gram-negative bacilli (11%). Physical signs of endocarditis (new or changing murmur, stigmata, emboli) were seen in 58%. At 6 months and 12 months, mortality was 46% and 47%, respectively. Surgical replacement of the infected valve led to significantly lower mortality (23%) as compared with medical therapy alone (56%), as assessed by both univariate and multivariate analyses (p < 0.05). Improved outcome was seen for the surgical group even when controlling for severity of illness at time of diagnosis. From these findings we conclude that accurate assessment of outcome in prosthetic valve endocarditis requires long-term follow-up of at least 6 months following diagnosis. Surgical therapy warrants greater scrutiny; evaluation in controlled clinical trials is appropriate.     

Edge crossings in drawings of bipartite graphs

Systems engineers have recently shown interest in algorithms for drawing directed graphs so that they are easy to understand and remember. Each of the commonly used methods has a step which aims to adjust the drawing to decrease the number of arc crossings. We show that the most popular strategy involves an NP-complete problem regarding the minimization of the number of arcs in crossings in a bipartite graph. The performance of the commonly employed barycenter heuristic for this problem is analyzed. An alternative method, the median heuristic, is proposed and analyzed. The new method is shown to compare favorably with the old in terms of performance guarantees. As a bonus, we show that the median heuristic performs well with regard to the total length of the arcs in the drawing.     

Rational wavelet decompositions of transfer functions in hardy-sobolev classes

The model approximation of transfer functions using rational wavelets (or molecular decompositions) is considered. By using techniques from Hardy-Sobolev spaces it is shown that Hilbert space methods such as a modified matching-pursuit algorithm and least-squares technique can be employed to obtain good approximations in bothH ² andH norms. Several theoretical results are given on rates of convergence when the methods are applied to delay systems and fractional filters.The research of the first author was supported by E.P.S.R.C.     

In-process monitoring and prediction of droplet quality in droplet-on-demand liquid metal jetting additive manufacturing using machine learning

Journal of Intelligent Manufacturing - In droplet-on-demand liquid metal jetting (DoD-LMJ) additive manufacturing, complex physical interactions govern the droplet characteristics, such as size,...     

The mitochondrial genome integrity gene, MGI1, of Kluyveromyces lactis encodes the beta-subunit of F1-ATPase

In a previous report, we found that mutations at the mitochondrial genome integrity locus, MGI1, can convert Kluyveromyces lactis into a petite-positive yeast. In this report, we describe the isolation of the MGI1 gene and show that it encodes the beta-subunit of the mitochondrial F1-ATPase. The site of mutation in four independently isolated mgi1 alleles is at Arg435, which has changed to Gly in three cases and Ile in the fourth isolate. Disruption of MGI1 does not lead to the production of mitochondrial genome deletion mutants, indicating that an assembled F1 complex is needed for the "gain-of-function" phenotype found in mgi1 point mutants. The location of Arg435 in the beta-subunit, as deduced from the three-dimensional structure of the bovine F1-ATPase, together with mutational sites in the previously identified mgi2 and mgi5 alleles, suggests that interaction of the beta- and alpha- (MGI2) subunits with the gamma-subunit (MGI5) is likely to be affected by the mutations.