A single specimen of the large-scaled mullet, P. grandisquamis, was collected from the muddy seabed along the coast of Cameroon, Africa (3.6322°N, 9.8005°E) ( Figure 1 ). The identification was confirmed based on taxonomic characteristics described in previous studies, as these often overlap with its closest congener, Chelon bandialensis. However, P. grandisquamis can be distinguished by having large size of scales, 25-30 scales in the longitudinal series and 8-10.5 scales in the transverse series (Trape et al., 2012). Additionally, the buccal teeth of P. grandisquamis are slightly smaller, and its dorsal, anal, and caudal fins are less distinctly yellowish compared to those of C. bandialensis. Muscle tissue was collected from the ventral thoracic region for molecular analysis and novel mitogenome generation. To validate the morphology-based species identification, we amplified a partial fragment of the mitochondrial COI gene using previously published primer pairs and protocols (Ward et al., 2005; Kundu et al., 2019). The amplified sequence showed a 99.79% nucleotide similarity in a BLAST search with the GenBank sequence (Accession number JQ060476) generated in the previous study where the genus and species were reassigned (Durand et al., 2012). The specimen was stored in 10% formaldehyde at the Fisheries and Animal Industries (MINEPIA) facility in Yaoundé, Cameroon. The experimental protocol was approved by the host institute Animal Care and Use Committee (PKNUIACUC-2022-72). The global distribution map of P. grandisquamis was obtained from FishBase and generated using the CMAR c-squares mapper, hosted at https://www.obis.org.au ( Figure 1 ).

The total genomic DNA was extracted by using an AccuPrep^® DNA isolation kit with standardized protocol (Bioneer in Daejeon, Republic of Korea). Both the quantity and quality of the DNA were checked through a NanoDrop Microvolume spectrophotometer (Thermo Fisher Scientific D1000, Waltham, MA, USA). The complete mitogenome of P. grandisquamis was sequenced through a next-generation sequencing approach on the NovaSeq platform (Macrogen, Daejeon, Republic of Korea). The TruSeq Nano DNA high-throughput library preparation kit was used to formulate the sequencing library (Illumina, Inc., San Diego, CA, USA). Primarily, 100 ng of genomic DNA was fragmented through an adaptive-focused acoustic tool (Covaris, Woburn, MA, USA). Further, abiding by the end-repair process, DNA fragments were chosen via a bead-based process, adapted by inserting an ‘A’ base, and later ligated through TruSeq DNA UD indexing adapters. To yield the final library, the resulting products endured purification and subsequent PCR enrichment. The quantification of the library was measured using qPCR, ensuing the typical procedure (KAPA Quantification Kits for Illumina Sequencing), and quality judgment was accomplished through the 4200 TapeStation D1000 screentape (Agilent Technologies, Santa Clara, CA, USA). Lastly, the paired-end (2 × 150 bp) sequencing was directed through the NovaSeq platform (Illumina, Inc., San Diego, CA, USA).

More than 20 million raw reads were screened using the Cutadapt tool (http://code.google.com/p/cutadapt/), applying a Phred quality score threshold (Q score > 20). The target mitogenome was assembled using high-quality paired-end reads with the Geneious Prime version 2023.0.1 software. The complete mitogenome of M. cephalus (Accession No. AP002930) served as the reference sequence during assembly, utilizing default mapping algorithms (Miya et al., 2001). Gene boundaries and strand orientations were validated with MitoAnnotator (http://mitofish.aori.u-tokyo.ac.jp/annotation/input/) (Iwasaki et al., 2013) and MITOS Version 1.1.6, integrated with the Galaxy Version 1.1.6 online servers (https://mitos.bioinf.uni-leipzig.de/) (Bernt et al., 2013). Overlapping regions were further scrutinized using MEGA X to confirm the accuracy of the assembled mitogenome (Kumar et al., 2018). The putative amino acid sequences for each protein-coding gene (PCG) were identified using the Open Reading Frame Finder web tool (https://www.ncbi.nlm.nih.gov/orffinder/), employing the vertebrate mitochondrial genetic code.

To obtain the full length of the control region (CR), a new primer pair (5′-CGTCGTAATTCTTACCTGAATTGG-3′ and 5′-GCCTGATACCAGCTCCTTGTC-3′) was designed from the segment between Cytb and 12S rRNA genes. The targeted CR was amplified using a TaKaRa Verity Thermal Cycler with a reaction mixture containing 1 µL of P. grandisquamis DNA template, 1X PCR buffer, 1 U Taq polymerase, 10 pmol of each primer, and 2.5 mM dNTPs. The amplicon was purified using the AccuPrep^® PCR/Gel Purification Kit (Bioneer, Daejeon, Republic of Korea). Subsequently, the purified amplicon was amplified using the BigDye^® Terminator v3.1 Cycle Sequencing Kit (Applied Biosystems, Foster City, CA, USA) and sequenced bidirectionally on an ABI PRISM 3730XL DNA analyzer via the Sanger sequencing method (Macrogen, Daejeon, Republic of Korea). The consensus sequence was screened using SeqScanner version 1.0 (Applied Biosystems Inc., Foster City, CA, USA) to eliminate noisy base pairs, and CR validation was confirmed by aligning overlapping regions with MEGA X. The final mitochondrial genome sequence of P. grandisquamis was deposited in the global GenBank database, where it received an accession number.

In this study, a circular representation of the assembled mitochondrial genome was generated using the MitoAnnotator online tool (http://mitofish.aori.u-tokyo.ac.jp/annotation/input/). Intergenic spacers between adjacent genes and overlapping sections were manually identified. The nucleotide composition of PCGs, ribosomal RNA (rRNA), transfer RNA (tRNA), and CR was determined using MEGA X. Base composition skew was calculated using the formulas AT-skew = [A − T]/[A + T] and GC-skew = [G − C]/[G + C] (Perna and Kocher, 1995). The mitochondrial genetic code of vertebrates was utilized in MEGA X to verify the initiation and termination codons of each PCG. Additionally, DnaSP 6.0 was employed for comparative analysis, including the calculation of relative synonymous codon usage (RSCU) and the assessment of amino acid abundance (Rozas et al., 2017). The boundaries of tRNA and rRNA genes were confirmed using ARWEN 1.2 and tRNAscan-SE Search Server 2.0 (Chan et al., 2021; Laslett and Canbäck, 2008). To investigate critical components such as PCG substitution patterns and the structure of conserved blocks in CRs, comparative analyses were performed with P. grandisquamis and other mugilids ( Supplementary Table S1 ). The DnaSP 6.0 program was also used to estimate pairwise substitution patterns of PCGs, including non-synonymous (Ka) and synonymous (Ks) substitutions. As detailed in previous studies, CLUSTAL X alignment was used to identify structural features within the conserved blocks in CRs (Thompson et al., 1997; Kundu et al., 2024).

To elucidate the evolutionary relationships among species in the Mugilidae family, including P. grandisquamis, we conducted a phylogenetic analysis using the mitochondrial genomes of 21 species representing 15 genera. The Black Mouth Cameroon Tilapia, Coptodon camerunensis (Cichliformes: Cichlidae) was included as an outgroup taxon (Access number OQ696044) ( Supplementary Table S1 ). The phylogenetic tree was constructed using a dataset comprising 13 PCGs, assembled with iTaxoTools 0.1 (Vences et al., 2021). The optimal substitution model, ‘GTR + G + I,’ was identified based on the Bayesian information criterion (BIC) using PartitionFinder 2 and JModelTest v2 (Darriba et al., 2012; Lanfear et al., 2016). Bayesian analysis (BA) was performed with MrBayes 3.1.2, employing a model with nst = 6, one cold and three hot chains of the Metropolis-coupled Markov chain Monte Carlo (MCMC). The analysis ran for 1,000,000 generations, with tree samples collected every 100 generations and 25% of the initial samples discarded as burn-in (Ronquist and Huelsenbeck, 2003). The resulting Bayesian tree was visualized using the iTOL v4 web server (https://itol.embl.de/login.cgi) (Letunic and Bork, 2007).

In this study, we illustrated the mitogenome of P. grandisquamis, revealing a length of 16,859 base pairs (bp) with access numbers in GenBank OR487150 ( Figure 1 ). The mitogenome for P. grandisquamis contains 37 genes, consisting of 13 PCGs, 22 tRNAs, two rRNAs, and AT-rich CR. The positive strand contains 12 PCGs, 14 tRNAs, and two rRNAs, while the negative strand contains a single PCG (ND6) as well as eight tRNAs (tRNA-Gln, tRNA-Ala, tRNA-Asn, tRNA-Cys, tRNA-Tyr, tRNA-Ser, tRNA-Glu, and tRNA-Pro) ( Table 1 ). The overall mitogenome base composition of P. grandisquamis showed an AT bias of 55.08% (A = 28.30%, T = 26.70%, G = 15.42%, and C = 29.43%), resulting an AT-skew of 0.030 and a GC-skew of -0.312 ( Table 2 ). Further investigation on the gene boundaries and arrangement revealed 15 intergenic spacers with a total of 122 bp and five overlapping regions, totaling of 23 bp in the mitogenome of P. grandisquamis. The longest intergenic spacer with 32 bp was identified between tRNA-Thr (T) and tRNA-Pro (P), while the most extensive overlap region of 10 bp was shared between the genes ATP8 and ATP6 ( Table 1 ).

The mitogenomes of P. grandisquamis contain 13 PCGs with a total length of 11,456 bp (68% of the complete mitogenome). Among that, the ND5 gene has the longest length with 1,824 bp, while the shortest was ATP8 with 168 bp ( Table 1 ). Additionally, the PCG of P. grandisquamis has an AT bias of 54.31% with an AT-skew of -0.019 and a GC-skew of -0.388. In the mitogenome of this species, most PCGs start codons beginning with ATG, except in COI which starts with GTG. Notably, five PCGs (ND1, COI, ATP8, ND4L, ND5) ended with TAA, while the remaining PCGs terminated by incomplete stop codon either TA- or T– ( Supplementary Table S1 ).

The analysis of Relative Synonymous Codon Usage (RSCU) revealed that the codons in the 13 PCGs were conserved and specifically translated to amino acids. The total number of codon transcriptions for the 20 amino acids was 3,808 for P. grandisquamis, excluding the termination codons. The amino acid composition of P. grandisquamis PCGs was predominantly characterized by two hydrophobic amino acids, leucine (14.29%) and alanine (5.86%), along with three neutral amino acids, serine (10.87%), proline (10.06%), and threonine (8.46%). In contrast, three hydrophobic amino acids—tryptophan (0.63%), cysteine (1.02%), and methionine (1.08%)—one hydrophilic amino acid, aspartic acid (1.73%), and one neutral amino acid, glycine (2.42%), were found to be less abundant ( Figure 2A , Supplementary Table S2 ). Notably, the RSCU analysis revealed a noteworthy six variation of codon in Arginine, Leucine and Serine, enabling an increasing its frequency. Moreover, several codons showed its augmentation (> 1.5) to translate a specific amino acid: GGC in Glycine, TCC in Serine, CCC in Proline, and GCC in Alanine, indicating as the most frequent compared to the other codons ( Figure 2B , Supplementary Table S3 ).

The calculation of the nonsynonymous (Ka) and synonymous (Ks) substitution proportions demonstrated that PCGs of P. grandisquamis and its relative species within the family Mugilidae are subject to the different experience of selection pressure. The average pairwise Ka/Ks ratios ranged from the lowest of 0.0124 ± 0.0034 for COI to a high of 1.2976 ± 0.1017 for ND5, following the order: COI < COII < COIII < Cytb < ATP6 < ND3 < ND1 < ND4 < ND2 < ATP8 < ND4L < ND6 < ND5. Most of PCGs showed Ka/Ks ratio value less than one, except ND4L, ND5, and ND6 that have value greater than one ( Figure 2C , Supplementary Table S4 ).

The mitogenome in this study has two ribosomal RNA molecules, explicitly 12S rRNA of 956 bp and 16S rRNA of 1,689 bp which collectively contribute 16% of the complete mitogenome. The rRNA gene displayed an AT bias of 53.91% with AT-skew and GC-skew contributing 0.185 and -0.122, respectively ( Table 2 ). The mitogenome of P. grandisquamis also contained 22 tRNA dispersedly in between rRNA and PCGs. It has a cumulative length of 1,556 bp, contributing 9% of total mitogenome. The tRNA genes of this species exhibited an AT bias of 57.20% with AT-skew and GC-skew of 0.106 and -0.114, respectively ( Table 2 ). Furthermore, most of tRNA showed conventional form of cloverleaf secondary structure, except for tRNA-Ser1 due to lack of nucleotide bond in DHU arm. Fifteen tRNA genes (tRNA-Phe, tRNA-Gln, tRNA-Met, tRNA-Trp, tRNA-Ala, tRNA-Asn, tRNA-Cys, tRNA-Tyr, tRNA-Ser2, tRNA-Asp, tRNA-Gly, tRNA-Arg, tRNA-Ser1, tRNA-Glu, and tRNA-Pro) were constructed through a combination of conventional Watson-Crick base pairs (A=T and G≡C) and wobble base pairs (G-T), while the other seven tRNA genes exclusively use Watson-Crick base pairs. Among those 15 tRNA genes, wobble base pairing was observed prominently in tRNA-Ala and tRNA-Glu. Moreover, Wobble base pairings were identified in the DHU stem of six tRNAs, the TψC stem of eight tRNAs, the anticodon stem of five tRNAs, and the acceptor stem of nine tRNAs ( Figure 3 ).

The total length of control region in P. grandisquamis was 1,103 bp in length, contributing 6.5% of the total mitogenome. The control region of this species showed an AT bias of 57.20%, with an AT-skew of 0.106 and a GC-skew of -0.114 ( Table 2 ). The CR of P. grandisquamis consists of four conserved blocks (CSB-D, CSB-1, CSB-2, and CSB-3), as observed in other fish mitogenomes, including other Mugilids (M. cephalus, Accession No. AP002930 and Moolgarda crenilabis, Accession No. JF911707). Among these conserved blocks, CSB-III is the longest at 31 bp, compared to other regions ranging from 18 bp (CSB-D and CSB-II) to 21 bp (CSB-I) ( Figure 4 ). Comparative analysis of the conserved blocks with other mugilids species demonstrated two groups in CSB-D; the first group with most of species including P. grandisquamis, showed a similarity on conserved nucleotides, however two species (Rhinomugil nasutus and Chaenomugil proboscideus) showed nucleotide similarity as second group. In CSB-I, three groups were observed indicating conserve nucleotides: the first group comprising with four species (Planiliza haematocheilus, Aldrichetta forsteri, R. nasutus, and Dajaus monticola), the second group with two species (M. crenilabis and Osteomugil engeli), and the third group with two species (Planiliza carinata and Planiliza lauvergnii). Similarly, the conserved nucleotides were also found in CSB-II, showing two different groups: the first group comprising six species and the other group with four species. While in CSB-III, three species (P. carinata, P. haematocheilus, and P. lauvergnii) were detected have similar conserved nucleotide, while the other 17 species have different nucleotide patterns, which may be diverged due to the deletion of base pairs in their CRs ( Figure 4 ). In addition, several mugilidis species allowed tandem repeats in the extended termination-associated sequences (ETAS) region, including 2.3 copy of 44 bp consensus sequence in Plicomugil labiosus, 6.8 copies of 79 bp consensus sequence in O. labeo, 2.6 copies of 44 bp consensus sequence in C. proboscideus, and 15.6 copies of 22 bp and 17.4 copies of 22 bp consensus sequences in Minimugil cascasia ( Figure 4 ).

A mitogenome-based phylogenetic analysis was performed using 13 PCGs from P. grandisquamis and other related species within the Mugilidae family. The BA phylogenetic tree clearly resolved the species into distinct classifications with high posterior probability support. P. grandisquamis was found to be closely related to species in the genera Chelon and Planiliza ( Figure 5 ). Additionally, the phylogenetic tree confirmed the matrilineal relationships of 10 monotypic species, including P. grandisquamis. The analysis also addressed taxonomic placement of ambiguous species within the Mugilidae family, such as M. cascasia (formerly Sicamugil cascasia), D. monticola (formerly Agonostomus monticola), M. crenilabis (formerly Crenimugil crenilabis), P. labiosus (formerly Oedalechilus labiosus), R. nasutus (formerly Squalomugil nasutus), Pseudomyxus capensis (formerly Myxus capensis), two species of Osteomugil (formerly under the genus Moolgarda), and four species of Planiliza (formerly under the genus Liza) ( Figure 5 ). Notably, the cladistic analysis revealed a non-monophyletic pattern for the genus Moolgarda, with M. crenilabis and M. cunnesius showing distinct lineages, suggesting the need for further investigation.

The mitochondrial genome of P. grandisquamis comprises 37 genes, with 28 genes located on the heavy strand and nine on the light strand. The nucleotide composition of the genome exhibits an A-T bias, which is consistent with the hydrophobic nature of mitochondrial proteins (Naylor et al., 1995). The gene arrangement and strand organization of P. grandisquamis align with those observed in other teleost mitogenomes (Miya et al., 2003, 2005). Our findings suggest that the mitochondrial genome of P. grandisquamis is conserved relative to other species within the Mugilidae family (Miya et al., 2001). Given the general conservation of mitochondrial genomes, gene rearrangements are an important focus in organismal systematics (Gong et al., 2020; Zhang et al., 2020). Therefore, understanding gene arrangement and nucleotide composition across species is essential for refining species classification. The arrangement of mitochondrial genes can influence various aspects of fish physiology, molecular mechanisms, life histories, and genomic evolutionary processes (Montaña-Lozano et al., 2022).

In PCGs, most genes use ATG as the initiation codon, with the exception of COI, which terminates with either complete or incomplete stop codons. This pattern of initiation and termination codon usage in COI is consistent with that observed in other mugilid species (Miya et al., 2001; Shen et al., 2016a, 2016). In this study, amino acids with neutral hydropathy characteristics were found to be more abundant than those with hydrophobic or hydrophilic properties. This observation aligns with findings from previous studies on other teleost species (Wang et al., 2022a). The hydropathy characteristics of amino acids play a crucial role in the evolution of mitochondrial proteins, which are essential for cellular respiration and energy production that are vital for the survival and adaptation of organisms in dynamic environments (Berthelot et al., 2019). Given the frequent fluctuations in salinity, temperature, and oxygen levels experienced by marine fishes, particularly in the Atlantic Ocean (Rutterford et al., 2023), detailed studies on hydropathy-related modifications in mitochondrial proteins could offer valuable insights into their role in optimizing protein conformation and function under such environmental stressors. These adaptations may enhance thermal tolerance and hypoxia resilience, enabling marine species to survive and thrive in highly variable and challenging habitats.

Codon usage was further assessed using RSCU values, revealing that most RSCU values deviated from ‘1’, indicating varying levels of codon bias for different amino acids. Codons for leucine (Leu) and serine (Ser), which have six codon combinations each, were the most frequently used, whereas methionine (Met) and tryptophan (Trp), each represented by a single codon (ATG and TGG, respectively), were less prevalent. Further, the Ka/Ks ratio, a well-established metric for assessing selective pressure and evolutionary associations at the molecular level (Yang and Nielsen, 2000; Zhao et al., 2022). Most PCGs exhibited Ka/Ks ratios less than ‘1’, suggesting strong negative selection in P. grandisquamis and its relatives within the Mugilidae family, with mutations predominantly replaced by synonymous substitutions. This pattern indicates the role of natural selection in reducing deleterious mutations with adverse selective coefficients, positioning with universal patterns detected in other teleosts (Kundu et al., 2024). However, three PCGs (ND4L, ND5, and ND6) showed Ka/Ks ratios greater than ‘1’, indicating potential positive selection. These genes may be more sensitive to genetic mutations, potentially influenced by environmental adaptations or new functional developments (Liao et al., 2024). The Ka/Ks ratio analysis provides valuable insights into how natural selection shapes speciation and evolutionary trajectories in mugilids across diverse aquatic environments. Previous studies have reported that NADH dehydrogenase genes have evolved under positive selection in anadromous salmonids and other deep-sea fishes, suggesting their adaptation to diverse ecosystems influenced by environmental factors such as salinity, temperature, low oxygen levels, darkness, and high-water pressure (Wang et al., 2022b; Shen et al., 2019). However, the specific roles of these genes in mullets remain unclear. The incorporation of this novel mitogenome data of P. grandisquamis into global databases, along with a comparative analysis of substitution patterns with other mugilids, provides baseline information to better understand the environmental adaptations of mullets. This foundational data is crucial for monitoring their protein-level evolutionary dynamics and adaptive strategies.

Additionally, the rRNAs in P. grandisquamis are located on the heavy strand and are separated by tRNA-Val, a pattern consistent with other fish species (Satoh et al., 2006; Liao et al., 2024). Ribosomes, as conserved ribonucleoproteins, play a crucial role in translating genetic information from mRNA into proteins. The structural association of rRNA genes, including their conserved loops, delivers valuable insights into the catalytic developments that are essential for protein synthesis (Lehman, 2004; Satoh et al., 2006). Our analysis also reveals that most tRNAs exhibit a cloverleaf secondary structure, with the exception of tRNA-Ser1. This cloverleaf pattern is common in the mitogenomes of many bony fish, although the absence of the DHU arm in some tRNAs can serve as a distinguishing feature (Liao et al., 2024). The tRNA molecules function as adaptors that translate genetic information into proteins by delivering amino acids during translation. The proper arrangement of tRNAs, particularly in the WANCY region, is crucial for effective mitochondrial gene expression (Cantatore et al., 1987; Ponce et al., 2008). Thus, the analysis of rRNAs and tRNAs in P. grandisquamis, especially through the evaluation of secondary structures, provides important insights into genetic mechanisms and mitochondrial function.

In the mitogenome of P. grandisquamis, the pronounced AT bias in the CR is consistent with observations in other fish species, which typically exhibit a preference for adenine (A) and thymine (T) bases (Miya et al., 2001, 2003, 2005). The CR is particularly significant due to its dynamic nature, as it is the most variable region in the mitochondrial genome. Notably, the repeat-rich ETAS region within the CR is characterized by specific motifs likely forming stable hairpin loops. These structures act as sequence-specific signals to regulate mitochondrial DNA replication termination (Satoh et al., 2016). Understanding the complex mechanisms controlling the CR—such as random loss, dimer formation of mitogenomes, non-random loss, and genomic rearrangement through double replications—is essential for grasping the structural diversity of mitogenomes and the evolutionary dynamics of mitochondrial genomes (Kundu et al., 2024). Thus, our analysis of the CR, including the identification of polymorphic nucleotides in conserved blocks, provides valuable insights for developing distinct molecular markers for species identification and understanding population structure.

The present mitogenome-based phylogenetic assessment provides valuable insights into the evolutionary relationships of P. grandisquamis, particularly in relation to other monotypic species within the Mugilidae family. This analysis supports the revision of classifications and highlights the non-monophyletic clustering of the Moolgarda genus within the Mugilidae family. Molecular phylogenetics offers robust, independent insights into species evolutionary relationships based on maternal lineage (Caldara et al., 1996; Heras et al., 2009; Liu et al., 2010). Integrating multi-dimensional research, including mitogenomic data alongside morphological information, is essential for elucidating ancestral and descendant lineages and resolving evolutionary patterns that have long been debated among ichthyologists due to morphological contradictions (Harrison et al., 2007; Durand et al., 2012; Rajan et al., 2023). Given the species richness and broad distribution of mugilids across diverse aquatic environments, it is crucial to expand mitogenomic research to enhance our understanding of their speciation and adaptation processes.

The large-scale mullet, P. grandisquamis demonstrates considerable versatility and adaptability across various habitats within the coastal ecosystems of the eastern Atlantic region. However, this adaptability may render it susceptible to genetic mutations and increased heterozygosity in response to environmental changes (Martinez et al., 2018). Coastal and riverine habitat alterations often negatively impact fish populations, potentially creating physical barriers to migration, which can lead to population fragmentation and disruptions in gene flow, thereby increasing the risk of inbreeding in restricted ecosystems (Pimentel et al., 2020; Ovidio et al., 2020). This fragmentation can reduce genetic diversity, threaten population viability, and heighten the risk of local extinction (Rourke et al., 2019). Therefore, integrating traditional species knowledge with molecular data is crucial for assessing genetic variation, population structure, and connectivity within specific ecosystems. Mitogenome-based studies are essential for observing adaptive evolution in Parachelon and other mugilids, which are currently underrepresented. The mitogenome characterization provided here serves as a foundational baseline for monitoring the genetic diversity of this migratory fish in the eastern Atlantic. Our findings highlight the vulnerability of P. grandisquamis to positive selection pressures and unique structural variations in its mitochondrial genome. Notably, mitochondrial PCGs such as ND4L, ND5, and ND6 exhibit heightened sensitivity to gene mutations, offering a basis for further research into the adaptive evolution of this endemic and monotypic species. Understanding how specific genes influence physiological adaptation is crucial for evaluating their adaptability to environmental changes and climate variability (Wang et al., 2022b; Shen et al., 2019). Given that P. grandisquamis is typically catadromous, with a tendency to migrate into freshwater systems, this migratory behavior may have a significant impact on mitochondrial gene evolution, as reflected in their Ka/Ks ratios. These insights will be instrumental in developing effective management strategies and conducting population genetic analyses of this species in near future. Additionally, given the restricted distribution of P. grandisquamis in the eastern Atlantic, this mitogenomic data will aid in elucidating lineage diversification patterns through time tree analyses incorporating multiple mitogenomic datasets from the Mugiliformes lineage.