No abstract

Background: Recent years have witnessed an increasing interest in multi-omics data, because these data allow for better understanding complex diseases such as cancer on a molecular system level. In addition, multi-omics data increase the chance to robustly identify molecular patient sub-groups and hence open the door towards a better personalized treatment of diseases. Several methods have been proposed for unsupervised clustering of multi-omics data. However, a number of challenges remain, such as the magnitude of features and the large difference in dimensionality across different omics data sources. Results: We propose a multi-modal sparse denoising autoencoder framework coupled with sparse non-negative matrix factorization to robustly cluster patients based on multi-omics data. The proposed model specifically leverages pathway information to effectively reduce the dimensionality of omics data into a pathway and patient specific score profile. In consequence, our method allows us to understand, which pathway is a feature of which particular patient cluster. Moreover, recently proposed machine learning techniques allow us to disentangle the specific impact of each individual omics feature on a pathway score. We applied our method to cluster patients in several cancer datasets using gene expression, miRNA expression, DNA methylation and CNVs, demonstrating the possibility to obtain biologically plausible disease subtypes characterized by specific molecular features. Comparison against several competing methods showed a competitive clustering performance. In addition, post-hoc analysis of somatic mutations and clinical data provided supporting evidence and interpretation of the identified clusters. Conclusions: Our suggested multi-modal sparse denoising autoencoder approach allows for an effective and interpretable integration of multi-omics data on pathway level while addressing the high dimensional character of omics data. Patient specific pathway score profiles derived from our model allow for a robust identification of disease subgroups.

No abstract

Abstract:A collective approach to resolve the segmentation problem was proposed. AntClust is a new ant-based algorithm that uses the self-organizing and autonomous brood sorting behavior observed in real ants. Ants and pixels are scatted on a discrete array of cells represented the ants' environment. Using simple local rules and without any central control, ants form homogeneous clusters by moving pixels from the cells of the array according to a local similarity function. The initial knowledge of the number of clusters and initial partition were not needed during the clustering process. Experimental results conducted on synthetic and real images demonstrate that our algorithm AntClust was able to extract the correct number of clusters with good clustering quality compared to the results obtained from a classical clustering algorithm like Kmeans algorithm.

In this paper, we propose a novel method for image segmentation that we call ACS-MRF method. ACS-MRF is a hybrid ant colony system coupled with a local search. We show how a colony of cooperating ants are able to estimate the labels field and minimize the MAP estimate. Cooperation between ants is performed by exchanging information through pheromone updating. The obtained results show the efficiency of the new algorithm, which is able to compete with other stochastic optimization methods like Simulated annealing and Genetic algorithm in terms of solution quality.