ABSTRACT
While the availability of various microarray platforms offers Populus researchers multiple options in gene expression analysis, cross-referencing between studies with different microarray systems, including previously developed expressed sequence tag (EST) microarrays, is diffi cult. The situation is further complicated by the constantly evolving nature of genome annotation. Computational annotation was accomplished using a variety of ab initio gene calling algorithms (Tuskan et al. 2006), and only a small number of gene models have been manually curated to date [Joint Genome Institute (JGI) Populus trichocarpa v1.1, https://genome.jgi-psf.org/Poptr1_1; and v2.0, https://www.phytozome.net/poplar.php]. Thus, multiple gene model predictions may be found at any given locus, and their inconsistent use in microarray probe design and annotation, as well as in the literature, represents a major challenge for comparative data analysis.
