Fig 2: FoxP2 overexpression is not evident in VSP samples.Using the same primers and reaction conditions as the Area X samples (main text Figure 1C), FoxP2 overexpression is not observed in either virus group relative to GFP.

Fig 3: FoxP2 expression in Area X is elevated for up to 30 days following AAV-FoxP2 injection.Juvenile birds (~30d) were injected with our FoxP2 overexpressing construct, bilaterally into Area X. FoxP2 levels were examined 10 days (n=3), 20 days (n=4), and 30 days (n=3) post-injection. Results were normalized to uninjected birds sacrificed at ~60 days of age (n=3). Area X FoxP2 levels (blue bars), were significantly elevated at 20 days post-injection and remained so at 30 days post-injection. No such increases were observed in outlying VSP (yellow bars).

Fig 4: Gene significance and network position implicate MAPK11 as a molecular entry point to vocal learning mechanisms.(A) The 20 genes with the highest to lowest gene significances to tutor similarity (sorted from top to bottom) are shown. Each column represents a bird and columns are sorted in order of increasing tutor similarity from left to right. Gene expression is scaled such the highest and lowest expression across samples have the brightest shade of red or blue, respectively. (B) Expression of MAPK11 is replotted, here separated by virus group and then sorted by increasing tutor percentage similarity. (C) The FoxP2 binding sequence as annotated by the JASPAR database (top) and a potential binding site found in the MAPK11 ‘promoter’. (D) Amplification of genomic DNA (‘Genomic’) with primers for a region of the MAPK11 ‘promoter’ that contains a putative FoxP2 binding site enrich a fragment of predicted size (red arrowhead) in the pull-down lane (FoxP2) but not the control (IgG) lane. (E) MAPK11 and its 10 closest network neighbors, including green learning-related module members and hub gene ATF2, as defined by topological overlap.10.7554/eLife.30649.031Figure 6—source data 1.The sorted gene expression data as presented in panel 6A.Columns are ordered from left to right by increasing tutor percentage similarity attained by that bird. Rows are sorted by the gene’s significance to learning.

Fig 5: Learning trajectories demonstrate that birds’ songs change over song development.Learning trajectory is calculated by quantifying acoustic similarity between birds’ songs over the course of the experiment and their song on the day of sacrifice (~65d). Upward trajectory indicates the songs for all three virus groups change over time. In agreement with the work of Heston & White, (2015), the FoxP2.FL group’s (black) song changes over time but, as illustrated in main text Figure 3D, not to match the tutor’s song.