An Efficient Feature Subset Selection Algorithm for Classification of Multidimensional Dataset

Dimensionality reduction strategy for Multi-Target Regression paradigm

Multi-Target Regression (MTR) is used to study the relationship between the same set of input variables and multiple continuous target variables simultaneously. A dataset with many input and output variables is the prime issue to address in the MTR, which is computationally complex to build a prediction model. Also, dimensionality reduction from multiple target variables is a challenging and essential task that aims to reduce the size of the dataset to optimize the time complexity of analysis and remove the redundant and irrelevant variables. This paper proposes an efficient feature selection strategy, Multi-Target Feature Subset Selection (MTFSS), for MTR that constructs a unique subset of features by considering multiple targets. On the other hand, two feature evaluators, correlation and ReliefF, support the MTR dataset without discretization. Furthermore, two new score functions, weighted mean aggregation strategy and threshold function, are introduced to identify the significant features. To evaluate the effectiveness of the proposed MTFSS, experiments were carried out on a benchmark dataset. The experimental results demonstrate that the proposed MTFSS can select fewer features and perform better than the original dataset results. Also, the correlation-based feature evaluator performs better than ReliefF with better performance.

Multi-task learning framework for predicting water quality using non-linear machine learning technique

Predicting the quality of water is a very important issue in an ecosystem and it can be used to control the increase of water contamination. Also, water quality prediction is a prominent complex non-linear multi-target learning problem and extracting a relevant subset of features from a large number of features with multiple targets is a challenging task. Existing water quality prediction model not focused on multi-target learning process simultaneously and not identifying the non-linear relationship between the features and target variables. Therefore, this study proposes a multi-task learning method dealing with multi-target regression using non-linear machine learning technique. Finally, experiments are conducted to build a prediction model based on the proposed methods to evaluate accuracy on water quality dataset. The experimental results indicate that our method increases the overall accuracy of the experimental dataset compared with the existing methods with the reduced number of significant features.

Deciphering the signaling mechanisms of the plant cell wall degradation machinery in Aspergillus oryzae

Background

The gene expression and secretion of fungal lignocellulolytic enzymes are tightly controlled at the transcription level using independent mechanisms to respond to distinct inducers from plant biomass. An advanced systems-level understanding of transcriptional regulatory networks is required to rationally engineer filamentous fungi for more efficient bioconversion of different types of biomass.

Results

In this study we focused on ten chemically defined inducers to drive expression of cellulases, hemicellulases and accessory enzymes in the model filamentous fungus Aspergillus oryzae and shed light on the complex network of transcriptional activators required. The chemical diversity analysis of the inducers, based on 186 chemical descriptors calculated from the structure, resulted into three clusters, however, the global, metabolic and extracellular protein transcription of the A. oryzae genome were only partially explained by the chemical similarity of the enzyme inducers. Genes encoding enzymes that have attracted considerable interest such as cellobiose dehydrogenases and copper-dependent polysaccharide mono-oxygenases presented a substrate-specific induction. Several homology-model structures were derived using ab-initio multiple threading alignment in our effort to elucidate the interplay of transcription factors involved in regulating plant-deconstructing enzymes and metabolites. Systematic investigation of metabolite-protein interactions, using the 814 unique reactants involved in 2360 reactions in the genome scale metabolic network of A. oryzae, was performed through a two-step molecular docking against the binding pockets of the transcription factors AoXlnR and AoAmyR. A total of six metabolites viz., sulfite (H₂SO₃), sulfate (SLF), uroporphyrinogen III (UPGIII), ethanolamine phosphate (PETHM), D-glyceraldehyde 3-phosphate (T3P1) and taurine (TAUR) were found as strong binders, whereas the genes involved in the metabolic reactions that these metabolites appear were found to be significantly differentially expressed when comparing the inducers with glucose.

Conclusions

Based on our observations, we believe that specific binding of sulfite to the regulator of the cellulase gene expression, AoXlnR, may be the molecular basis for the connection of sulfur metabolism and cellulase gene expression in filamentous fungi. Further characterization and manipulation of the regulatory network components identified in this study, will enable rational engineering of industrial strains for improved production of the sophisticated set of enzymes necessary to break-down chemically divergent plant biomass.

Electronic supplementary material

The online version of this article (doi:10.1186/s12918-015-0224-5) contains supplementary material, which is available to authorized users.