On the performance of de novo pathway enrichment
Computational biology: Evaluation of network-based pathway enrichment tools De novo pathway enrichment methods are essential to understand disease complexity. They can uncover disease-specific functional modules by integrating molecular interaction networks with expression profiles. However, how sho...
Guardado en:
| Autores principales: | , , , , , , , |
|---|---|
| Formato: | article |
| Lenguaje: | EN |
| Publicado: |
Nature Portfolio
2017
|
| Materias: | |
| Acceso en línea: | https://doaj.org/article/d4aeb2316347474c8564b07783489ae1 |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| Sumario: | Computational biology: Evaluation of network-based pathway enrichment tools De novo pathway enrichment methods are essential to understand disease complexity. They can uncover disease-specific functional modules by integrating molecular interaction networks with expression profiles. However, how should researchers choose one method out of several? In this article, a group of scientists from Denmark and Germany presents the first attempt to quantitatively evaluate existing methods. This framework will help the biomedical community to find the appropriate tool(s) for their data. They created synthetic gold standards and simulated expression profiles to perform a systematic assessment of various tools. They observed that the choice of interaction network, parameter settings, preprocessing of expression data and statistical properties of the expression profiles influence the results to a large extent. The results reveal strengths and limitations of the individual methods and suggest using two or more tools to obtain comprehensive disease-modules. |
|---|