Cloud computing-based parallel genetic algorithm for gene selection in cancer classification

Cancer classification is one of the main steps during patient healing process. This fact enforces modern clinical researchers to use advanced bioinformatics methods for cancer classification. Cancer classification is usually performed using gene expression data gained in microarray experiment and advanced machine learning methods. Microarray experiment generates huge amount of data, and its processing via machine learning methods represents a big challenge. In this study, two-step classification paradigm which merges genetic algorithm feature selection and machine learning classifiers is utilized. Genetic algorithm is built in MapReduce programming spirit which makes this algorithm highly scalable for Hadoop cluster. In order to improve the performance of the proposed algorithm, it is extended into a parallel algorithm which process on microarray data in distributed manner using the Hadoop MapReduce framework. In this paper, the algorithm was tested on eleven GEMS data sets (9 tumors, 11 tumors, 14 tumors, brain tumor 1, lung cancer, brain tumor 2, leukemia 1, DLBCL, leukemia 2, SRBCT, and prostate tumor) and its accuracy reached 100% for less than 25 selected features. The proposed cloud computing-based MapReduce parallel genetic algorithm performed well on gene expression data. In addition, the scalability of the suggested algorithm is unlimited because of underlying Hadoop MapReduce platform. The presented results indicate that the proposed method can be effectively implemented for real-world microarray data in the cloud environment. In addition, the Hadoop MapReduce framework demonstrates substantial decrease in the computation time.

     

Parameter-less co-clustering for star-structured heterogeneous data

The availability of data represented with multiple features coming from heterogeneous domains is getting more and more common in real world applications. Such data represent objects of a certain type, connected to other types of data, the features, so that the overall data schema forms a star structure of inter-relationships. Co-clustering these data involves the specification of many parameters, such as the number of clusters for the object dimension and for all the features domains. In this paper we present a novel co-clustering algorithm for heterogeneous star-structured data that is parameter-less. This means that it does not require either the number of row clusters or the number of column clusters for the given feature spaces. Our approach optimizes the Goodman–Kruskal’s τ, a measure for cross-association in contingency tables that evaluates the strength of the relationship between two categorical variables. We extend τ to evaluate co-clustering solutions and in particular we apply it in a higher dimensional setting. We propose the algorithm CoStar which optimizes τ by a local search approach. We assess the performance of CoStar on publicly available datasets from the textual and image domains using objective external criteria. The results show that our approach outperforms state-of-the-art methods for the co-clustering of heterogeneous data, while it remains computationally efficient.     

Preparation,properties, and laser processing of poly(ɛ-caprolactone)/poly(lactic acid) blends with addition of natural fibers as a potential for printing plates application

In this research, biodegradable blend of poly(ɛ-caprolactone) (PCL) and poly(lactic acid) (PLA) is proposed as a new material for the production of a printing plate for embossing process. Printing plates for embossing consist of raised printing elements and recessed nonimage elements. In production of printing plates, laser technology was used in order to form a relief printing plate. The embossing process is based on the principle of the pressure of the relief printing plate into the printing substrate, which causes the controlled deformation of the substrate and three-dimensional (3D) effect. Coir fibers (CFs) were added as a natural filler to PCL/PLA blends to improve and adjust the properties of produced blends. Scanning electron microscopy micrographs, dynamic mechanical analysis analysis, roughness, and hardness were measured on prepared materials, and 2D and 3D microscopy was conducted on laser engraved printing plates. Results have shown that the addition of CFs improved the mechanical properties of produced materials. DMA results indicate the semicrystalline structure of all prepared blends, and that the addition of CFs raises the elasticity of the composites. Laser engraving showed that it is possible to engrave the produced biodegradable materials and to use it as a material for production of printing plates.     

Development of a Patient-Derived Xenograft (PDX) of Breast Cancer Bone Metastasis in a Zebrafish Model

Bone metastasis is a complex process that needs to be better understood in order to help clinicians prevent and treat it. Xenografts using patient-derived material (PDX) rather than cancer cell lines are a novel approach that guarantees more clinically realistic results. A primary culture of bone metastasis derived from a 67-year-old patient with breast cancer was cultured and then injected into zebrafish (ZF) embryos to study its metastatic potential. In vivo behavior and results of gene expression analyses of the primary culture were compared with those of cancer cell lines with different metastatic potential (MCF7 and MDA-MB-231). The MCF7 cell line, which has the same hormonal receptor status as the bone metastasis primary culture, did not survive in the in vivo model. Conversely, MDA-MB-231 disseminated and colonized different parts of the ZF, including caudal hematopoietic tissues (CHT), revealing a migratory phenotype. Primary culture cells disseminated and in later stages extravasated from the vessels, engrafting into ZF tissues and reaching the CHT. Primary cell behavior reflected the clinical course of the patient’s medical history. Our results underline the potential for using PDX models in bone metastasis research and outline new methods for the clinical application of this in vivo model.     

Crosstalk corrections for improved energy resolution with highly segmented HPGe-detectors

Crosstalk effects of 36-fold segmented, large volume AGATA HPGe detectors cause shifts in the γ-ray energy measured by the inner core and outer segments as function of segment multiplicity. The positions of the segment sum energy peaks vary approximately linearly with increasing segment multiplicity. The resolution of these peaks deteriorates also linearly as a function of segment multiplicity. Based on single event treatment, two methods were developed in the AGATA Collaboration to correct for the crosstalk induced effects by employing a linear transformation. The matrix elements are deduced from coincidence measurements of γ-rays of various energies as recorded with digital electronics. A very efficient way to determine the matrix elements is obtained by measuring the base line shifts of untriggered segments using γ-ray detection events in which energy is deposited in a single segment. A second approach is based on measuring segment energy values for γ-ray interaction events in which energy is deposited in only two segments. After performing crosstalk corrections, the investigated detector shows a good fit between the core energy and the segment sum energy at all multiplicities and an improved energy resolution of the segment sum energy peaks. The corrected core energy resolution equals the segment sum energy resolution which is superior at all folds compared to the individual uncorrected energy resolutions. This is achieved by combining the two independent energy measurements with the core contact on the one hand and the segment contacts on the other hand.     

Multidimensional networks: foundations of structural analysis

Complex networks have been receiving increasing attention by the scientific community, thanks also to the increasing availability of real-world network data. So far, network analysis has focused on the characterization and measurement of local and global properties of graphs, such as diameter, degree distribution, centrality, and so on. In the last years, the multidimensional nature of many real world networks has been pointed out, i.e. many networks containing multiple connections between any pair of nodes have been analyzed. Despite the importance of analyzing this kind of networks was recognized by previous works, a complete framework for multidimensional network analysis is still missing. Such a framework would enable the analysts to study different phenomena, that can be either the generalization to the multidimensional setting of what happens in monodimensional networks, or a new class of phenomena induced by the additional degree of complexity that multidimensionality provides in real networks. The aim of this paper is then to give the basis for multidimensional network analysis: we present a solid repertoire of basic concepts and analytical measures, which take into account the general structure of multidimensional networks. We tested our framework on different real world multidimensional networks, showing the validity and the meaningfulness of the measures introduced, that are able to extract important and non-random information about complex phenomena in such networks.     

Large scale tensor regression using kernels and variational inference

Machine Learning - We outline an inherent flaw of tensor factorization models when latent factors are expressed as a function of side information and propose a novel method to mitigate this. We...     

Environmental impact assessment of an automotive component using eco-indicator and CML methodologies

The environmental conscious manufacturing and climate change made automotive manufacturing organisations to recognise sustainability as a vital concept for survival in the global competitive scenario. Many components of automobiles are manufactured using plastics which play imperative role in ensuring sustainability. This research article deals with the application of Life Cycle Assessment on instrument panel of an electric car which is made of ABS (Acrylonitrile–Butadiene–Styrene). The case study has been conducted in an Indian instrument panel manufacturing plant. First, a life cycle model of ABS sheet which is particularly used in instrument panel manufacturing was built, and the environmental impact assessment of the instrument panel was done using Eco-indicator and CML methodologies. In life cycle perspective, the environmental hot spots have been found out. Later the results have been interpreted based on environmental impact assessment factors. The relative comparison of impact factors at different stages in the life cycle enables the generation of improvement options. The sustainability index of instrument panel before and after implementing Environmental Impact Assessment (EIA) has been computed. The results are statistically validated, and the managerial implications have been derived.