Introduction to network inference in genomics

Abstract

The genome is the archetypical complex system: it is a finely tuned whole whose many parts, such as DNA, RNA and proteins, interact at various levels to execute intricate functions, such as repair, replication and adapting to the external environment. One particularly effective way of conceptualizing this complex system is by means of a network, in which the vertices describe the genomic components and the edges describe their physical or functional interactions. With the advent of modern high-throughput genomic measuring devices, such as microarrays, RNA-seq and other next generation sequencing tools, it has become possible to measure the vertices of the genomic system in real time. One central question is whether from these measurements it is possible to reconstruct the edges of the genomic network. This essay describes three modelling and inference strategies to answer this central biological question.