**2. Materials and Methods**

#### *2.1. Sample Collection and DNA Extraction*

*L. kasmira* samples were collected from Meiji Dao (MJ) and Zhubi Dao (ZB), which are separated by approximately 200 km (Figure 1). A total of 30 *L. kasmira* samples were collected, comprising 14 samples from MJ and 16 samples from ZB (Table 1). Tissue was collected by clipping the fin strip of the experimental samples. The tissue samples were fixed in 75% anhydrous ethanol and stored at room temperature. A Tissue DNA Extraction CZ Kit (Mobio, Guangzhou, China) was used to extract genomic DNA from the pterygiophore of each fish according to the manufacturer's protocol.

**Figure 1.** Locations of the 30 collected *Lutjanus kasmira*.



#### *2.2. RAD-Seq Library Construction and Sequencing*

RAD-seq libraries were prepared as described by Yangkun Wang and Yan Hu [11]. First, genomic DNA samples were digested by the restriction endonuclease EcoRI. Then, P1 adapter was added to both ends of the digested genome fragment, which contains a complementary sequence that binds to the primer for PCR amplification and a complementary sequence that binds to the primer for Illumina sequencing. The barcode used for sample tracking and the corresponding restriction enzyme cutting site were also part of the P1 adapter. Next, the sequence of the added P1 adapter was interrupted. Through agarose gel detection, the target band was selected in the range of 400~500 bp. The DNA fragment was then attached to the P2 adapter. The multiplexed libraries were sequenced on the Illumina 1.9 platform with an average sequencing depth of 1.5× for each individual.
