With the genome of green anole available for download, anole researchers are apt to face a dilemma related to both experimental design and the taste and judgment of science consumers. My colleagues and I recently published an article in Genome Biology and Evolution in which the anole genome was downloaded for analyses. Spearheaded by bioinformatics guru Charles Chapus, our experiments were conducted in silico. Through conversations with friends and colleagues, we intuited that we might face resistance to our methods because, although our hypotheses and analyses were novel, our raw data were not. Fortunately, this concern, held by some, did not postpone publication of our work, but it did raise an interesting point for discussion. Where is genomic research going, in relation to generation and/or analysis of sequence data? From the perspective of experimental design, the decision to generate new sequence depends on the research question. For population genetics studies, for example, novel sequence generation is efficacious but for whole genome comparisons, why repeat what is freely available for download?
Have you published work based on pre-existing data? How will you decide whether to amplify or download anole sequences for analyses?
Marc Tollis
Since I study pop-gen of A. carolinensis, my lab work starts in silico with marker development from the database, then I screen across my sample. For phylogeographic analysis of neutral genes in populations, this is not a problem. But for another part of my study, I aim to test if certain transposable element markers are under selection, so the database presents an ascertainment bias (rare alleles are unlikely to be sampled). Therefore, novel sequence generation is necessary.