In 2001, a Britain-based team led by Dr. Anthony Monaco found that mutations in a single gene called FOXP2 underlie a rare inherited speech and language disorder inherited as an autosomal-dominant monogenic trait in three generations of a family. The gene encodes a transcription factor protein that is believed to play a central role in the evolution of language in mammals by facilitating the mechanical aspects of sound articulation as suggested by a study published in Nature on 2009 by a team of researchers under supervision of Dr. Daniel Geschwind. In the study Geschwind and collaborators showed that 61 genes increased and 51 genes decreased their expression rate when human FOXP2 was overexpressed in human brains as compared to the same treatment with the homologous protein found in chimps, revealing an interesting role in the regulation of the soft-tissue development-related molecular targets of the affected genes. Since homologous of this transcription factor can be found across the animal kingdom I wanted to visually explore the known interactions of the FOXP2 across vertebrate species in order to find molecular cues of the protein interaction networks that might explain the ability of humans to produce speech.
The dataset was obtained from the GIANT API, an initiative hosted by Dr. Olga Troyanskaya's Laboratory for Functional Genomics and Gene Network Analysis, by querying the database to get known protein interactions of FOXP2 in the human embryo, mice (Mus musculus), and zebrafish (Danio rerio) animal developmental models. The information about molecular function of all the recovered proteins was manually curated using UniProtKB database and six main categories were recovered: Cell signaling, Immunity, Metabolism, Molecular traffic, Structural, Transcription factor, and Unknown. The datasets have the genes as their nodes and place a link between two nodes if there is any line of evidence of interaction, the weight of the link is a metric of the level of evidence of such interaction. Due to the density of the connections in the networks, only the top 2% of links sorted by weight was drawn. In the case of zebrafish the top 5% of the links were drawn due to the sparsity of the network.
To interact with the following visualization please hover with the mouse over the nodes of interest. The color of the nodes indicate their molecular function. The nodes with direct interaction with the selected node appear with their links highlighted, the thickness of the highlighted link is correlated with the level of support of such molecular interaction, the color of the highlighted link indicates the direction of the interaction, and the nodes that do not interact with the selected node (i.e. not any registered interaction at any support level) fade on mouse over.
FOXP2 seems to act in the early developmental processes by acting over a few key master transcriptional regulators. Because of that, mutations over the gene can have enormous implications given that many other genes may not have a direct way of controlling the gene. It is interesting, though, that the interacting proteins do not seem to be conserved in the three species observed, also suggesting extensive remodelling of the gene circuitry.