Genome-wide Gene Duplication Analysis

Different gene duplication patterns were identified using DupGen_finder (Qiao et al., 2019 (link)). First, an all-vs.-all local BLASTP search was performed using protein sequences to screen for potential homologous gene pairs in each genome (E < 1e-10, top 5 matches, and m8 format output). Then, the A. thaliana protein sequence was selected as the outgroup and the DupGen_finder-unique program was used to identify five gene duplication events (i.e., TD, PD, DSD, TRD, and WGD). The detect_gene_conversion pipeline (Qiao et al., 2019 (link)) was used to identify homologous gene quartets and analyze gene conversions.
The synonymous substitution rate (Ks) and the non-synonymous substitution rate (Ka) for the duplicated gene pairs were calculated using KaKs_calculator 2.0 and the NG model (Wang et al., 2010 (link)) (Supplementary Table S3). The Ks values were converted to divergence times using the formula T = Ks·(2r)⁻¹, where T is the divergence time and r is the neutral substitution rate (6.50 × 10⁻⁹) (Gaut et al., 1996 (link)).