Conclusions
We assessed the extent to which adaptation to common environments
targets predictable changes in transcript abundance across independent
evolutionary events. Within lineages, genes with greater expression
plasticity were more likely to diverge in expression between
populations, but in the direction opposite of plastic changes. Despite a
small number of differentially expressed genes shared across drainages,
parallel adaptation to low-predation environments in independent
lineages was associated with divergence in largely non-overlapping
transcripts. While identification of shared genes is generally used as
the starting point for work exploring mechanisms of parallel adaptation
– and our work here indeed suggests that a small number of core genes
may indeed be critical during the earliest stages of adaptation – we
propose that parallel evolutionary transitions are not limited to a
small set of possible transcriptional mechanisms in guppies. Instead,
our results highlight the potential for alternative transcriptional
solutions associated with parallel, adaptive trait evolution even within
a single species. Transcriptional network versatility, in which diverse
alterative network configurations can produce common network outputs and
behavioral phenotypes, may allow underlying networks to simultaneously
accommodate the influences of selection, drift, and genetic background
and thereby facilitate evolution in a species known for rapid adaptation
to novel environments.
Acknowledgements : We thank the members of the Colorado State
University Guppy Group for fish care, especially Sarah E Westrick and
Kimberly E Dolphin for help with tissue collection. We thank Laura R
Stein for assistance with tissue processing and Cameron K Ghalambor for
fruitful discussions and comments on earlier versions of the manuscript.
All high-powered computing was performed on the Odyssey computing
cluster supported by the FAS Science Division Research Computing Group
at Harvard University. This work was supported by the National Science
Foundation DDIG-1311680 (to EKF), RCN IOS-1256839 (to EKF), IOS-1354755
(to KLH), IOS-1354775 (to KAH), IOS-0934451 (to KAH), DEB-0846175 (to
CKG), IIS-1545994 (to WZ), and US Department of Energy DE-SC0018344 (to
WZ).
Data Accessibility : Raw sequencing reads will be made available
through the NCBI SRA repository upon publication. R code for statistical
analyses will be made available on GitHub upon publication.
Author contributions : EKF and KLH conceived of the study; EKF
collected samples and performed molecular work; EKF and KAH performed
gene expression mapping, transcript abundance estimation, and
preliminary differential expression analyses; YS and WZ devised and
performed statistical analyses with input from EKF and KLH; EKF and KLH
wrote the manuscript with contributions from all authors. All authors
gave final approval for publication.