Identification and chromosomal mapping

The G. raimondii, A. thaliana, V. vinifera, and T. cacao gene files were downloaded from Phytozome v11.0 (http://www.phytozome.net/). The gene information of G. hirsutum acc. TM-1 was downloaded from http://mascotton.njau.edu.cn . The Hidden Markov Model (HMM) profile of the AP2 domain (PF00847) was obtained from the Pfam website (http://pfam.xfam.org/), and was employed as a query to identify all possible AP2/EREBPs using HMMER (V3.0) software [65]. To validate the HMM search, all candidate sequences were used as queries to search the NCBI non-redundant (nr) protein database with the blastp program, and the results with the best ‘AP2’ hits were retained for further analysis. To predict the GPI-anchor attachment sites, the BGI-PI47 [66] and GPI-SOM48 [67] algorithms were used. The AP2/ERF sequences were confirmed based on the presence of an AP2 domain, and all of the putative AP2/ERF proteins were aligned to Arabidopsis AP2/ERF proteins to classify them into different groups, as described by Feng et al. [20].

Positional information on all of the AP2/EREBPs was parsed from the General Feature Format (GFF) files downloaded from Phytozome v11.0, and the locations of AP2/EREBPs in G. raimondii and G. hirsutum were drafted using MapInspect software (http://mapinspect.software.informer.com/).