RESUMO
The identification of condition-specific gene sets from transcriptomic experiments is important to reveal regulatory and signaling mechanisms associated with a given cellular response. Statistical methods of differential expression analysis, designed to assess individual gene variations, have trouble highlighting modules of small varying genes whose interaction is essential to characterize phenotypic changes. To identify these highly informative gene modules, several methods have been proposed in recent years, but they have many limitations that make them of little use to biologists. Here, we propose an efficient method for identifying these active modules that operates on a data embedding combining gene expressions and interaction data. Applications carried out on real datasets show that our method can identify new groups of genes of high interest corresponding to functions not revealed by traditional approaches. Software is available at https://github.com/claudepasquier/amine.
Assuntos
Biologia Computacional , Software , Biologia Computacional/métodos , Redes Reguladoras de Genes/genética , Perfilação da Expressão Gênica/métodos , Transdução de Sinais/genéticaRESUMO
MicroRNAs, small non-coding elements implied in gene regulation, are very interesting biomarkers for various diseases such as cancers. They represent potential prodigious biotechnologies for early diagnosis and gene therapies. However, experimental verification of microRNA-disease associations are time-consuming and costly, so that computational modeling is a proper solution. Previously, we designed MiRAI, a predictive method based on distributional semantics, to identify new associations between microRNA molecules and human diseases. Our preliminary results showed very good prediction scores compared to other available methods. However, MiRAI performances depend on numerous parameters that cannot be tuned manually. In this study, a parallel evolutionary algorithm is proposed for finding an optimal configuration of our predictive method. The automatically parametrized version of MiRAI achieved excellent performance. It highlighted new miRNA-disease associations, especially the potential implication of mir-188 and mir-795 in various diseases. In addition, our method allowed to detect several putative false associations contained in the reference database.