Analysis of normal and pathological behavior of biological systems has
greatly benefited from incorporating networks that capture knowledge about
interactions among genes, proteins or metabolites into explorations of data from highthroughput
techniques such as microarrays, next-generation sequencing, or mass
spectrometry. Signaling, metabolic and regulatory networks can help make results from
statistical data analysis techniques more interpretable to biologists, and can serve as a
regularizing factor narrowing the search space for statistical models that determine
differences between different states of biological systems. However, knowledge about
the underlying biological networks is still incomplete, with false negatives and false
positives, which may affect outcomes of systems biology studies. Thus, results of
network modeling need to be critically evaluated by domain experts, and followed by
additional experimental or computational validation.
Keywords: Biological networks, Computational methods, Differential network
analysis, Network inference.
