2.6 Pseudomolecule construction by Hi-C
To construct a chromosomal-level assembly of the genome, Hi-C raw data were first trimmed by fastp v. 0.12.6 (Chen et al. 2018). After quality control, the low-quality reads, adapter contamination and ambiguous bases as N’s were removed, whilst duplicates were filtered out. High quality clean paired-end reads were retained. The clean paired-end reads were aligned with the draft assembled genome using Juicer pipeline v. 2.3.2 (Durand et al. 2016). Afterwards, according to the location of DpnII restriction sites, the ratio of Self Circle, Dangling End and Dumped Pairs was identified so as to evaluate the validity of the paired-end reads. The contigs were then clustered, ordered and oriented using the 3D de novo assembly (3d-DNA, v. 170 123) pipeline (Dudchenko et al. 2017). Hi-C contact matrix was visualized using Juicebox v. 1.9.8 (Dudchenko et al. 2018; Robinson et al. 2018). The misassembly and misconnection were manually adjusted based on neighboring interactions. The validated assembly was used to construct pseudomolecules using the finalize-output.sh script from 3d-DNA. Meanwhile, the completeness of the assembly was evaluated using Benchmarking Universal Single-Copy Orthologs (BUSCO) v3.1 (Simão et al. 2015).