2.4 | Sequence assembly and sequence analysis
After quality control, clean data from complete mtDNA were aligned against a reference genome (GenBank, NC_001788.1) using Bowtie2. Reads on the calibration were reserved and SPAdes v3.13.0 (Parameters: K 127) was used for genome assembly. The splicing result was compared with the closed reference genome using BLASTN, and the assembly result was determined accordingly. tRNAs were recognized and their secondary structures were predicted by tRNAscan-SE v2.0 (Chan & Lowe, 2019). The most variable region was identified by sequence alignment against the reference genome. We derived a circular map of the mitogenome using ORDRAW (Greiner et al., 2019), and base composition skew was calculated using AT-skew = (A − T)/(A + T) and GC-skew = (G − C)/(G + C) (Perna & Kocher, 1995).
The 418 bp section was manually edited and aligned against the reference sequence using CLUSTALX v2.0 (Larkin et al., 2007). DNAsp v6 (Rozas et al., 2017) was used to calculate the nucleotide diversity, polymorphic sites, and haplotype diversity of the most variable region in the HGD mtDNA D-loop region. Genetic relationships among populations were determined using median-joining networks (MJN) using Network v.10.1.0.0 software (Bandelt et al., 1999), the MJN was constructed from 11 haplotypes found in this study and 60 reference sequences found previously (Ozkan Unal et al., 2020; Xia et al., 2019), the 15 sequences including Hap4, Hap6, Hap7, and Hap10, Hap12, Hap16, Hap18, Hap20, Hap22, Hap23, Hap24, Hap25, Hap26, Hap27, Hap29 belonged to Clade Ⅰ. Reference mtDNA D-loop region sequences of wild and domestic donkey were obtained from GenBank and used as comparators of HGDs to build phylogenetic trees and speculate their origin (Han et al., 2014; Ivankovic et al., 2002; Kimura et al., 2011; Oakenfull et al., 2000; Ozkan Unal et al., 2020). The phylogenetic tree was drawn using the Neighbor-joining (NJ) method. Bootstrap values were estimated using 1,000 repetitions (Felsenstein, 1985), based on the Kimura-2-parameter genetic distances (Saitou & Nei, 1987), and reconstructed using MEGA v7.0 (Kumar et al., 2016).