Reveal the Population Genetic Characteristics of Bombay Duck (Harpadon nehereus) in Coastal Waters of China with Genotyping-by-Sequencing Technique

YANG Tianyan, HUANG Xinxin, and JIANG Yanlin

Reveal the Population Genetic Characteristics of Bombay Duck () in Coastal Waters of China with Genotyping-by-Sequencing Technique

YANG Tianyan1), 2), *, HUANG Xinxin1), and JIANG Yanlin1)

1),,316022,2),,316022,

Bombay duck () is an economically important species in the estuarine and coastal offshore waters of the Indo-West Pacific. This study aims to reveal the genome-wide genetic characteristics of five populations offrom the coastal areas of China by using the genotyping-by-sequencing technique. Afterstrictfiltering,32088 high-quality single-nucleotide polymorphism markers were detected and analyzed. The average observed heterozygosity and expected heterozygosity ranged from 0.41651 to 0.56725 and from 0.30998 to 0.45531, respectively, indicating that heterozygosity excess occurred inpopula- tions. The nucleotide diversity ranged from 0.63664 to 0.74868, which was larger than those from other marine fishes. No obvious genetic structure was detected amongpopulations, and the genetic variation originated within individuals. Extensive gene exchange caused by longshore currents in the reproductive season might be the primary reason for the weak genetic differentia- tion. Among various environmental factors, water temperature might be the key element affecting the genetic structure of. Due to the destruction and overfishing of spawning grounds, the fishery resources declined in recent years. This study could serve as a reference for the resource protection and rational utilization of

; genotyping-by-sequencing; population genetics; single-nucleotide polymorphisms

1 Introduction

Bombay duck (Hamilton, 1822) be- longs to Synodontidae, and is a commercially importantfish that is widely distributed in the neritic waters (less than 50m) and estuarine areas of the Western Indo-Pacific Ocean (Nelson, 2016). In China,main- ly inhabits from the Yellow Sea to the South China Sea(Chen and Zhang, 2015). As a familiar lizardfish with eco- nomic value in China, Pakistan, and India,is favored by consumers because of its delicate, delicious, and nutrient-rich meat. It is susceptible to spoilage because of the high moisture content (90%) of fresh meat. The fried fish is a famous dish known as ‘Dried bummelo’ in China, and ‘Bombil fry’ in the western coastal areas of India.

Given the increasingly prominentissues of marine en-vironmental degradation and traditional fisheryresourceshortages since the 1980s, the proportion of annual yield for small and medium-sized fishes represented byhas increased consistently. According to FAO fishery statistics, the global landing of Bombay duck reached280000t in 2008. In India, it was the most abundant on the west coast of Maharashtra and Gujarat, and the catch of this fish contributed approximately 9.6% of all marine fish(Jaiswar and Chakraborty, 2016). The average landing catch was 66t at the Karachi Fish Harbor, and the maxi- mum catch was recorded to be 101t in 1996 (Kalhoro., 2013). The estimated biomass ofwas 2125t in the East China Sea, and the potential biomass could surpass5000t or more (Lin, 2009). Nowadays,has be- come an important commercial catch along the southeast coast of China and the Bay of Bengal region and has gra- dually become a key species in the food chain of the coas- tal ecosystem (Kurian and Kurup, 1992; Lin, 2009; Mohan- raj., 2009; Pan, 2013; Yang., 2013). With the increase in fishing pressure of, thefish caught have become smaller and younger in recent years, indicating that the stock ofis overexploited (Chen., 2012; Nugroho., 2015; Jaiswar and Cha- kraborty, 2016).Nowadays,has been added to ‘The IUCN Red List of Threatened Species’ as a near-threatened (NT) species (https://www.iucnredlist.org/spe- cies/75143569/75144431) (Russell., 2019). Previous studies onmainly focused on the areas of stockdynamic and assessment (Khan, 1989; Kurian and Kurup, 1992; Ghosh., 2009; Kalhoro., 2013), fishery bio- logical characteristics (Luo., 2012; Ghosh, 2014;Zhang and Jin, 2014; Hasin, 2016; He., 2019; Taqwa., 2020), morphologicalmeasurement and description (Ku- mar., 2014; Pazhayamadom., 2015; Firdaus., 2018), and meat quality and safety (Rupsankar, 2010; Pra-vakar., 2013; Bhattacharya., 2016). The genetic variation and population structure of the species remained poorly understood.

The genetic diversity of marine fish populations consi- derably affects their dynamics, and understanding the po- pulation genetic structure is crucial to the scientific pro-tection and effective management of marine fishery re- sources (Carvalho, 2004; Johnson., 2016). Scientific investigation on the population genetic background must be conducted to avoid the degeneration of germplasm re- sources and realize the sustainable utilization ofresources. Population genetics focuses on the genetic composition and diversity of populations, including gene- tic drift, mutation, and gene flow (Gillespie, 1998). It is helpful to understand the relationships among speciation, adaptive evolution, population dynamics, and environment.With the development of inshore fishery resources ma- nagement, studies on population structure and genetic va- riation based on genomic DNA have attracted considerable attention in recent decades (Carvalho and Hauser, 1998).Over the past years, several PCR-based DNA molecular markers have been applied to detect the geneticcharacte- ristics of; these markers include simple se- quence repeats (Xu, 2011),sequence-related ampli- fied polymorphisms (Zhu., 2014), and mitochondrial markers (Zhang., 2013; Zhang., 2018; Guo.,2019; Saha., 2019; Jiang., 2020). The above stu- dies undeniably accumulated valuable references for fur- ther investigations of genetic background. However, tradi-tional molecular biotechnology has limited power to revealcomplex genetic information. Meanwhile, the rapid deve- lopment of next-generation sequencing (NGS) holds a greatpromise for studying genetic diversity and evolutionary his- tory at the genome-wide level.

Genotyping-by-sequencing (GBS) is a relatively low- cost, high-throughput genotyping approach for sequenc- ing large genomes in diverse organisms, and it has become one of the most widely used reduced-representation genome sequencing methods. It relies on restriction enzymes to re- duce genome complexity and performs genetic analysis or genotyping based on the single-nucleotide polymorphism (SNP) marker system (Elshire., 2011). As a cost-ef- fective and unique tool for population genetics and geno- mics-assisted breeding in the absence of reference genomedata, the GBS approach has been increasingly used in fish- eries and aquaculture (Li and Wang, 2017). In the present study, we adopted the GBS strategy to analyze the popul- ation diversity and structure ofin the coastal waters of China. This study could serve as a reference for the exploitation and utilization ofresources.

2 Materials and Methods

2.1 Sample Collection and DNA Extraction

Forty-three specimens were collected between July 2018 and September 2019 from north to south along China’s coastline. The sampling information is shown in Table 1 and Fig.1. Fresh muscles were preserved in anhydrous ethanol and stored at ?20℃ for further experiments.

Fig.1 Sample collection locations of H. nehereus. Capital letters in brackets mean the abbreviations of sampling lo- cations.

Table 1 Sampling information of different populations

Genomic DNA was extracted from muscle tissues usinga standard phenol-chloroform method (Sambrook., 1982). The integrity of DNA was first observed by 1.0% agarose gel electrophoresis. Then, DNA purity (OD260/OD280ratio) and concentration were detected on a Nano-Drop 2000 spectrophotometer (Thermo Scientific, Waltham, MA, USA) and a Qubit 2.0 fluorometer (Invitrogen, Carls- bad, CA, USA), respectively. The qualified DNA precipita- tion was dissolved in sterilized ddH2O for subsequent li- brary construction.

2.2 GBS Library Construction and High-Throughput Sequencing

Two restriction endonucleases (and) were selected to digest the genome DNA (0.1–1μg) in compre- hensive consideration of the genomic coverage and base repetition ratio.The P1 adapter with a forward amplifica- tion primer, a sequencing primer, and a barcode was add- ed to enzyme digestion products. Then, a P2 adapter was added to the pooled and sheared DNA fragments. The adap-tor-ligated DNA sequences were amplified using polymerasechain reaction (PCR). The PCR products were obtained using the AxyPrep DNA gel extraction kit (AxyGEN, CA, USA), and the final concentration was adjusted to 1ngμL?1for later use. Agilent 2100 was applied to detect the insert sizes, and qPCR was used to quantify the effective con- centration of the libraries (effective concentration>2nm).

The required fragments were selected for library cons- truction. Individuals from each sampling site were pooled into one library, and the pooled libraries were sequenced based on the Illumina HiseqTM2500 sequencing platform with the 150-paired-end sequencing strategy. GBS librariesand sequencing were carried out by Novogene Co., Ltd. (Beijing, China).

2.3 Quality Control and Data Processing

Paired-end sequencing reads were trimmed of adapter sequences by Cutadapt 1.2.1 (Martin, 2011) and analyzed for quality using FastQC 0.11.7 (http://www.bioinforma-tics.babraham.ac.uk/projects/fastqc/). Meanwhile, the ob-tained sequences were evaluated for contamination by Blast+ (e-value≤1e?10, similarity>90% and coverage>80%).The loci of each sample were identified using the ‘ustacks’program in Stacks 1.0 (Catchen, 2013) and subsequent- ly merged into a catalog using the ‘cstacks’ program. Theloci data were matched against the catalog using ‘sstacks’ pro- gram for determining the allele status in each sample.

High-quality sequencing data were mapped to the re- ference genome by BWA (Burrows-Wheeler Aligner) soft- ware with the settings mem ‘-t 4 -k 32 -M, ’ where -t pre- sents the number of threads, -k presents the minimum seed length, and -M presents the option to flag shorter split hits as secondary alignments (Li and Durbin, 2010). The align- ment reads in SAM formatwere extracted and converted into BAM format and then sorted according to genomic positions by SAMtools with a maximum of 1000 reads at a position per BAM file (Li., 2009).

2.4 SNP Calling and Genome-Wide Genetic Variation Detection

SNP refers to the DNA sequence polymorphism caused by single-nucleotide variation at the genomic level, includ- ing single-base transition/transversion,single-base deletions, and single-base insertions. SNP calling was implemented using the ‘mpileup’ function of SAMtools (Li., 2009). Population genetic parameters, including the expected he- terozygosity (), observed heterozygosity (), nucleo- tide diversity (), and pairwiseST, were calculated for every SNP using the population program in Stacks. Gene- tic variation within and among populations was estimated with an analysis of molecular variance (AMOVA) in Arle- quin 3.5 (Excoffier and Lischer, 2010).

2.5 Assessment of Population Structure, Phylogeny, and Environmental Adaption

Vcftools waspreviously used to convert vcf file to bi- nary ped format, the input file format of Plink (Purcell., 2007), and then the population genetic structure was in- vestigated by Frappe software package (Tang., 2005) with the filtering settings ‘Dp1-miss0.9-maf0’. Principal component analysis (PCA) was carried out using the PCA module in GCTA (http://cnsgenomics.com/software/gcta/ pca.html), and R software was used to draw a PCA scatter plot. The-distance matrix was built by TreeBest (http:// treesoft.sourceforge.net/treebest.shtml).On this basis, the neighbor-joining (NJ) phylogenetic tree of the 43 indivi- duals was constructed using MEGA 6 (Tamura., 2013). The confidence levels at the branch nodes in the tree were calculated from 1000 bootstrap replications.

The interrelation between genetic variation and two ma-jor environmental factors (salinity and temperature) was investigated. The Bayesian approach was applied as im- plemented in Bayenv on all SNPs to remove outliers (Coop., 2010). The salinity and temperature data (from 2005 to now) were retrieved from the National Oceanic and At- mospheric Administration website (https://oceanwatch.pifsc. noaa.gov/doc.html).LFMM 2.0 (Kevin., 2019) was applied to analyze the association between environmental factors and selected loci. TheBonferroni correction at the significance level of 0.05 was performed to screen SNPs withvalues higher than 5(<0.01).

3 Results

3.1 GBS Data and SNP Detecting

High-quality data are the precondition for subsequent analysis. In this study, 24.03Gb clean data were generatedafter strict filtrationand normalization, with an average of 558.75Mb clean data per sample. The mean GC content ranged from 39.16% to 41.84%, and the average Q20 and Q30 were 96.35% and 90.57%, respectively, representing the lower base error ratio and normal distribution of GC content. The number of clean reads mapped to the refe- rence genome was 85521206 with the mapping rate rang- ing from 77.22% to 90.35%.

An average of 2728880 tags per individual was gained with a mean sequencing depth of 10.7× after the BWA alignment to the reference genome. After SNP calling and filtering, 32088 high-quality SNPs were obtained for sub- sequence procession.

3.2 Population Genetic Diversity

Genetic diversity is the basis for evolution in a species, and evaluating the genetic variation of different populations is an important content of genetic resources protection (Hilbish, 1996). The average observed heterozygosity(), expected heterozygosity (), and nucleotide diversity () were estimated by Arlequin soft based on the fil- tered SNP information (Table 2). A relatively high level of genetic diversity was detected in Nantong population, followed by Qingdao and Chaozhou, and the lowest level of genetic diversity was found in the samples from Zhou- shan.

Table 2 Population genetic diversity parameters of H. nehereus

3.3 Genetic Variation and Population Structure

AMOVA was conducted to estimate thedifferent sources and structures of genetic variation of(Table 3)The genetic variation ofwas significantly higher within individuals (99.93%) than among popula- tions (0.36%) (<0.05). PCA was applied to cluster indi- viduals into different subgroups according to their charac- teristics based on the SNP differences at the genome level to investigate the genetic differentiation among populations. Most of the samples were clustered together and no obvi- ous genetic differentiation was found, except for a few de- viant samples that might be due to statistical error, as shown in Fig.2.

Table 3 Analyses of molecular variance (AMOVA) in different populations of H. nehereus

Fig.2 Principal component analyses of the fiveH. nehereuspopulations along the coastal waters of China. Each sample is represented by a triangle.

Population genetic structure refers to the nonrandom dis- tribution of genetic variation, and this study utilized spa- tial autocorrelation analysis to explore variation among ge- notypes in species or populations (Gyllensten, 1985). The genetic structure among all individuals was investigated using Frappe software to understand the degree of admix- ture in the populations. The maximum likelihood method was applied to infer thegenetic ancestor of each indivi- dual. We predefined the number of genetic clusters from=2 to=8to explore the convergence of individuals. Theoptimalvalue was determined according to the coeffi- cient of variation, and the optimal assumption was that the individuals originated from=5 ancestor groups (Fig.3). The cluster results showed no obvious genetic differentia- tion among different geographical populations and random mating of them.

3.4 Population Phylogenetics and Adaptive Evolution

The NJ phylogenetic tree was constructed by 1000 boot- strap permutations on the basis of SNP loci information. The topology displayed individuals from different popu- lations clustered together, indicating close relationships to each other (Fig.4). In our study, the Bayenv and LFMM methods were combined to verify outlier SNP loci. The for- mer was based on population genetic differentiation, and the latter was based on the relationships between SNP al- lele frequency and environmental variables. A total of 149 selected SNPs (143 associated with temperature and 6 as- sociated with salinity) were detected inby both methods.

4 Discussion

4.1 Feasibility of Genetic Analysis Based on GBS

Fish stock is the fundamental unit in fisheries manage- ment and assessment,and understanding the population ge- netic structure of marine fishes is crucial to the sustain- able exploitation of fisheries resources (Reiss., 2010). Althoughis an important commercial species in China and the coastal countries of the Arabian Sea and the Bay of Bengal, its genetic background remains unclear. Since the beginning of the 21st century, NGS haspresent- ed a new approach for population genomics detection. The massive SNPs detected using GBS have been widely used for genetic diversity and adaptive differentiation analysisin several fish species, such as(Larson., 2014),(Glazer., 2015),(Guo., 2016),(Farrell., 2016),(Xu., 2019), and(Yang., 2020). In the present study, we applied GBS to decipher the popula- tion diversity and genetic variation ofalong the Chinese coast for the first time. The mean Q30 value (90.57%) and effective rate (99.99%) indicated high qua- lity and integrity of library construction and low sequen- cing error rate. Given the numbers of effective reads obtain- ed by the NGS platform, we considered that the evenly dis-tributed data of each individual could meet the needs of subsequent analysis.

Fig.3 Individual cluster of H. nehereus with genetic structure analysis. Each sample is represented by a histogram, which is partitioned into different colors. Each color represents a genetic cluster.

Fig.4 Phylogenetic tree of all samples collected along coas- tal waters of China.

4.2 Genetic Diversity Analysis

Genetic diversity is the fundamental element of biodi- versity, which is important for the survival, reproduction, and evolution of species in the world. Parameters, such as nucleotide diversity (), observed heterozygosity (), and expected heterozygosity () based on SNPs, are often used as criteria to evaluate population genetic diversity (John- son., 2016; Yang., 2020). In previous studies, the average nucleotide diversities based on mtDNA Cytandgenes were 0.000371±0.000379 and 0.000427±0.000424, respectively (Guo., 2019; Jiang., 2020), which were obviously lower than the results we obtained (0.63664–0.74868). This finding was related to the limitedmarker loci selected by traditional molecular markers, whichpossibly caused the loss of massive genetic information.Therefore, the genetic diversity ofcan be more accurately evaluated byapplying GBS based on the whole genome sequencing. The genetic diversity was also higher than many other marine organisms, such as(0.0025–0.0027) (Du, 2018),(0.136±0.064) (Xu., 2019),(0.1665–0.1937) (Gao., 2020), and(0.1439–0.1674) (Wang., 2020). The above outcomes demonstrated thatpopulations with high levels of genetic diversity had robust capabilities for adapting to changing environmental conditions.

Xu. (2011) developed and characterized the mi- crosatellite markers forand revealed that the observed and expected heterozygosities ranged from 0.3500 to 0.8421 and from 0.5244 to 0.6244, respectively. Com- pared with our results, the difference betweenthetwo da- tasets might largely be caused by sampling locations and molecular marker differences. In this study, the average(0.41651–0.56725) washigher than the average(0.30998–0.45531) in all populations of. He- terozygosity excess across different populations suggest- ed that demographic change led to a recent bottleneck of. Historical population dynamics analysis based on the Cytgene also confirmed thathad ex- perienced a population expansion event, which occurred about 0.08–0.32 million years ago in the middle and late Quaternary Pleistocene (Guo., 2019).

4.3 Genetic Differentiation and Environmental Adaptation

Molecular variance analysis revealed that 99.93% of the genetic variation resided within individuals, and rela- tively weak genetic variation originated from interpopula- tion differences. The above results agreed with previous analyses using mitochondrial DNA genes (Cytand) (Guo., 2019; Jiang., 2020) and microsatellite markers (Xu., 2011). In the PCA plot, the specimens presented mixed distribution and could not be separated by three eigenvectors,which also proved the lack of sig- nificant genetic difference among individuals in differentgeographic populations. These results were consistent withthe topology of the NJ phylogenetic tree andclustering graph of population structure.has the reproductive strategy of inshore migration and spawning pelagic eggs (Luo, 2012). Furthermore, no obvious geo- graphical barrier existed among different sampling loca- tions. Therefore, thefree-floating eggs and larvae could be easily transported bystrong coastal currentsduring the breeding season. The frequent gene flow and drift caused a low level of genetic differentiation among populations along the coast of China.

Marine fishes inhabiting diverse environments have beensubject to a great deal of natural selection pressure. The genetic variation caused byadaptive evolution reflects their potential to cope with different surroundings (Hoffmann and Willi, 2008). Environmental adaptation analysis using different SNP datasets was implemented in. Temperature and salinity are two of the most important abiotic factors affecting fish distribution. In the present study, more SNP loci were associated with temperature change than with salinity change, implying thatcould better adapt to temperature gradients than to salinity vari- ations. As a wide-temperature-salinity fish,canlive in a wide range of water temperature (9.11–32℃) and salinity (31.36–35.26) (Lin, 2009). In conclusion, the large temperature span caused abundant genetic variations inin different sea areas.

Acknowledgements

This research was supported by the Scientific Research Projects of the Zhejiang Bureau of Education (No. Y201 942611), and the Open Foundation from Marine Sciences in the First-Class Subjects of Zhejiang Province.

Bhattacharya, T., Ghorai, T., Dora K, C., Sarkar, S., and Chowd- hury, S., 2016. Influence of chemical preservatives on the qua- lity and shelf-life of dried Bombay duck ()., 11 (1): 1-8.

Carvalho, G. R., 2004. Population genetics: Principles and appli- cations for fisheries scientists., 5 (1): 94- 95.

Carvalho, G. R., and Hauser, L., 1998. Advances in the mole- cular analysis of fish population structure., 65 (sup1): 21-33.

Catchen, J., Hohenlohe, P. A., Bassham, S., Amores, A., and Cres- ko, W. A., 2013. Stacks: An analysis tool set for population ge- nomics., 22 (11): 3124-3140.

Chen, D. G., and Zhang, M. Z., 2015.. China Ocean University Press, Qingdao, 416-418 (in Chinese with English abstract).

Chen, L., Shui, B. N., and Dong, W. X., 2012. Growth charac- teristics and resources sustainable utilization of., 3: 68-70 (in Chinese with English abstract).

Coop, G., Witonsky, D., Rienzo, A. D., and Pritchard, J. K., 2010.Using environmental correlations to identify loci underlying local adaptation., 185: 1411-1423.

Du, X., 2018. Study on the phylogeography and adaptive diffe- rentiation ofalong the coast of China. Master thesis. Zhejiang Ocean University, 50-51 (in Chinese with Eng- lish abstract).

Elshire, R. J., Glaubitz, J. C., Sun, Q., Poland, J. A., Kawamoto, K., Buckler, E. S.,, 2011. A robust, simple genotyping- by-sequencing (GBS) approach for high diversity species., 6 (5): e19379.

Excoffier, L., and Lischer, H. E. L., 2010. Arlequin suite ver 3.5: A new series of programs to perform population genetics ana- lyses under Linux and Windows., 10 (3): 564-567.

Farrell, E. D., Carlsson, J., and Carlsson, J., 2016. Next Gen Pop Gen: Implementing a high-throughput approach to population genetics in boarfish ().,3: 160651.

Firdaus, M., Lelono, T. D., Saleh, R., Bintoro, G., and Salim, G., 2018. The expression of the body shape in fish species(Hamilton, 1822) in the waters of Juata Laut, Tarakan city, North Kalimantan., 11 (3): 613- 624.

Gao, L., Bao, X. B., Yu, S. M., Li, Z., Li, Y. F., and He, C. B., 2020. Analysis of the genetic characteristics ofpopulations along the Yellow and Bohai Seas using GBS., 27 (2): 204-212 (in Chi- nese with English abstract).

Ghosh, S., 2014. Fishery, reproductive biology and diet charac- teristics of Bombay duckfrom the Sau- rashtra coast., 43 (3): 418- 426.

Ghosh, S., Pillai, N. G. K., and Dhokia, H. K., 2009. Fishery and population dynamics of(Ham.) off the Saurashtra coast., 56 (1): 13-19.

Gillespie, J. H., 1998.. The Johns Hopkins University Press, Baltimore, 27pp.

Glazer, A. M., Killingbeck, E. E., Mitros, T., Rokhsar, D. S., andMiller, C. T., 2015. Genome assembly improvement and map- ping convergently evolved skeletal traits in sticklebacks with genotyping-by-sequencing., 5 (7): 1463-1472.

Guo, B. C., Li, Z. T., and Meril?, J., 2016. Population genomic evidence for adaptive differentiation in the Baltic Sea herring., 25 (12): 2833-2852.

Guo, Y. J., Yang, T. Y., Meng, W., Hang, Z. Q., and Gao, T. X., 2019. The genetic structure of the Bombay duck () based on mitochondrial Cytgene., 43 (5): 945-952 (in Chinese with English abstract).

Gyllensten, U., 1985. The genetic structure of fish: Differences in the intraspecific distribution of biochemical genetic varia- tion between marine, anadromous, and freshwater species., 26 (6): 691-699.

Hasin, T., 2016. Length-weight relationship and condition factor of Bombay duckfrom landings of fishery ghat, Chittagong., 3 (2): 355-360.

He, X., Li, J., Shen, C., Shi, Y., Feng, C., Guo, J. H.,., 2019. Length-weight relationship and population dynamics of Bom- bay duck () in the Min River Estuary, East China Sea., 35: 253-261.

Hilbish, T. J., 1996. Population genetics of marine species: The interaction of natural selection and historically differentiated populations., 200 (1-2): 67-83.

Hoffmann, A. A., and Willi, Y., 2008. Detecting genetic responses to environmental change., 9: 421-432.

Jaiswar, A. K., and Chakraborty, S. K., 2016. A review on fish- ery, biology and stock parameters of Bombay duck(Hamilton, 1822) occurring in India., 43: 67-96.

Jiang, Y. L., Meng, W., Zhu, W. B., and Yang, T. Y., 2020. Gene- tic diversity analysis ofpopulations based on mitochondrial DNAgene., 39 (2): 102-109 (in Chinese with English abstract).

Johnson, D. W., Freiwald, J., and Bernardi, G., 2016. Genetic di- versity affects the strength of population regulation in a ma- rine fish., 97 (3): 627-639.

Kalhoro, M. A., Liu, Q., Memon, K. H., Chang, M. S., and Jatt, A. N., 2013. Estimation of maximum sustainable yield of Bom- bay duck,fishery in Pakistan using the CEDAand ASPIC packages., 45 (6): 1757- 1764.

Kevin, C., Basile, J., Johanna, L., and Olivier, F., 2019. LFMM 2: Fast and accurate inference of gene-environment associations in genome-wide studies., 36 (4): 852-860.

Khan, M. Z., 1989. Population dynamics of the Bombay duck,(Ham.), off Saurashtra coast., 36 (2): 93-101.

Kumar, V. V., Reddy, A. D., Choudhury, S. R., Balakrishna, C. H.,Satyanaryana, Y., Nagesh, T. S.,., 2014. Morphometry and meristic counts of Bombay duck,(Hamil- ton, 1822) along Sunderban region of West Bengal, India., 4 (3): 95-105.

Kurian, A., and Kurup, K. N., 1992. Stock assessment of Bom- bay duck(Ham.) off Maharashtra coast., 39: 243-248.

Larson, W. A., Seeb, J. E., Pascal, C. E., Templin, W. D., and Seeb, L. W., 2014. Single-nucleotide polymorphisms (SNPs) identi- fied through genotyping-by-sequencing improve genetic stock identification of Chinook salmon () from western Alaska., 71 (5): 698-708.

Li, H., and Durbin, R., 2010. Fast and accurate short read align- ment with Burrows-Wheeler transform., 26 (5): 589-595.

Li, H., Handsaker, B., Wysoker, A., Fennell, T., Ruan, J., Ho- mer, N.,., 2009. The sequence alignment/map format and SAMtools., 25 (16): 2078-2079.

Li, Y. H., and Wang, H. P., 2017. Advances of genotyping-by-se- quencing in fisheries and aquaculture., 27: 535-559.

Lin, L. S., 2009. Spatial distribution and environmental charac- teristics ofin the East China Sea region., 18 (1): 66-71 (in Chi- nese with English abstract).

Luo, H. Z., 2012. Study of main biology character and analysis of resource status on theMaster thesis. Zhe-jiang Ocean University, 11-14 (in Chinese with English ab- stract).

Luo, H. Z., Zhang, H. D., Li, P. F., and Zhou, Y. D., 2012. Ana- lysis of the current situation of fishery biology ofin the East China Sea., 31 (3): 202-205 (in Chinese with English abstract).

Martin, M., 2011. Cut adapt removes adapter sequences from high-throughput sequencing reads., 17 (1): 1-10.

Mohanraj, G., Nair, K. V. S., Asokan, P. K., and Ghosh, S., 2009. Status of marine fisheries in Gujarat with strategies for sus- tainable and responsible fisheries., 22: 285-296.

Nelson, J. S., Grande, T. C., and Wilson, M. V. H., 2016.. Wiley, New York, 752pp.

Nugroho, D., Faizah, R., Prasetyo, A. P., and Badrudin, M., 2015. Bio-exploitation status of Bombay duck (Hamilton, 1822) on trawl fishery in Tarakan waters., 21 (1): 53-59.

Pan, G. L., Wang, Z. M., Zhang, H. L., and Zhou, Y. D., 2013. Spatial-temporal distribution of the biomass ofin the coastal area of south Zhejiang during spring., 32 (2): 99- 102 (in Chinese with English abstract).

Pazhayamadom, D. G., Chakraborty, S. K., Jaiswar, A. K., Sud- heesan, D., Sajina, A. M., and Jahageerda, S., 2015. Stock struc- ture analysis of ‘Bombay duck’ (Hamil- ton, 1822) along the Indian coast using truss network morpho- metrics., 31: 37-44.

Pravakar, P., Mansur, M. A., and Asadujjaman, M., 2013. Qua- lity and safety aspects of three sun-dried marine fish species: Chinese pomfret (), Bombay duck () and ribbon fish ()., 8 (4): 381-387.

Purcell, S., Neale, B., Todd-Brown, K., Thomas, L., Ferreira, M. A. R., Bender, D.,., 2007. PLINK: A tool set for whole-ge- nome association and population-based linkage analyses., 81 (3): 559-575.

Reiss, H., Hoarau, G., Dickey-Collas, M., and Wolff, W. J., 2010. Genetic population structure of marine fish: Mismatch between biological and fisheries management units., 10 (4): 61-395.

Rupsankar, C., 2010. Improvement of cooking quality and gel for- mation capacity of Bombay duck () fish meat., 47 (5): 534-540.

Russell, B., Govender, A., and Borsa, P., 2019... T75143569 A75144431. IUCN Red List Committee.

Saha, S., Ferdous, Z., Jahan, H., Khandaker, A. M.,Khandaker, R. M., and Begum, R. A., 2019. Polymorphic loci analysis of 16S ribosomal RNA gene of economically important marine lizardfish Bombay duck ()., 47 (1): 49-57.

Sambrook, J., Fritsch, E. F., and Maniatis, T., 1982.. Cold Spring Harbor Labora- tory Press, New York, 545pp.

Tamura, K., Stecher, G., Peterson, D., Filipski, A., and Kumar, S., 2013. MEGA6: Molecular evolutionary genetics analysis version 6.0., 30 (12): 2725- 2729.

Tang, H., Peng, J., Wang, P., and Risch, N. J., 2005. Estimation of individual admixture: Analytical and study design consi- derations., 28 (4): 289-301.

Taqwa, A., Burhanuddin, A. I., Niartiningsih, A., and Nessa, M. N., 2020. Nomei fish (, Ham. 1822) reproduc- tion biology in Tarakan waters., 473: 012012.

Wang, F. J., Meng, X. H., Fu, Q., Luan, S., and Sui, J., 2020. Ana- lysis of genetic diversity in three generations of breeding po- pulations ofbased on reduced-re- presentation genome sequencing.,41 (4): 68-76 (in Chinese with English abstract).

Xu, S. Y., Yanagimoto, T., Song, N., Cai, S. S., Gao, T. X., and Zhang, X. M., 2019. Population genomics reveals possible ge-netic evidence for parallel evolution ofin the northwestern Pacific Ocean., 9 (9): 190028.

Xu, T. J., Sun, D. Q., Li, H. Y., and Wang, R. X., 2011. Develop- ment and characterization of microsatellite markers for the li- zardfish known as the Bombay duck,(Sy- nodontidae)., 10 (3): 1701- 1706.

Yang, B. Z., Yang, L., Tang, Y. G., Yan, L., and Zhang, P., 2013. Preliminary analysis on catch composition of gillnet fishery forin north part of the South China Sea., 40 (2): 99-102 (in Chinese with English abstract).

Yang, T. Y., Gao, T. X., Meng, W., and Jiang, Y. L., 2020. Ge- nome-wide population structure and genetic diversity of Japa- nese whiting () inferred from genotyping-by- sequencing (GBS): Implications for fisheries management., 225: 105501.

Zhang, B., and Jin, X. S., 2014. Feeding habits and ontogenetic diet shifts of Bombay duck,., 32 (2): 542-548.

Zhang, B., Xu, T. J., Wang, R. X., Jin, X. X., and Sun, Y. N., 2013. Complete mitochondrial genome of the Bombay duck(Aulopiformes, Synodontidae)., 24 (6): 660-662.

Zhang, H., and Xian, W. W., 2018. The complete mitochondrial genome of the larval Bombay duck(Au- lopiformes, Synodontidae) from Yangtze estuary and the phy- logenetic relationship of Synodontidae species., 3 (2): 657-658.

Zhu, Z. H., Li, H. Y., Qin, Y., and Wang, R. X., 2014. Genetic di- versity and population structure inbased on sequence-related amplified polymorphism markers., 13 (3): 5974-5981.

April 16, 2021;

June 29, 2021;

November 16, 2021

? Ocean University of China, Science Press and Springer-Verlag GmbH Germany 2022

. E-mail: hellojelly1130@163.com

(Edited by Qiu Yantao)