Advances in Functional Genomics for Watermelon and Melon Breeding: Current Progress and Future Perspectives

Abstract

Watermelon (Citrullus lanatus) and melon (Cucumis melo) are globally important cucurbit crops, with China being the largest producer and consumer. Traditional breeding methods face difficulties in significantly improving yield and quality. Smart breeding, which combines genomics, gene editing, and artificial intelligence (AI), holds great promise but fundamentally depends on understanding the molecular mechanisms controlling important agronomic traits. This review summarizes the progress made over recent decades in discovering and understanding the functions of genes that control essential traits in watermelon and melon, focusing on plant architecture, fruit quality, and disease resistance. However, major challenges remain: relatively few genes have been fully validated, the complex gene networks are not fully unraveled, and technical hurdles like low genetic transformation efficiency and difficulties in large-scale trait phenotyping limit progress. To overcome these and enable the development of superior new varieties, future research priorities should focus on the following: (1) systematic discovery of genes using comprehensive genome collections (pan-genomes) and multi-level data analysis (multi-omics); (2) deepening the study of gene functions and interactions using advanced gene editing and epigenetics; (3) faster integration of molecular knowledge into smart breeding systems; (4) solving the problems of genetic transformation and enabling efficient large-scale trait and genetic data collection (high-throughput phenotyping and genotyping).

1. Introduction

Watermelon (Citrullus lanatus) and melon (Cucumis melo) are both annual vine crops of the Cucurbitaceae family and also are globally significant horticultural crops. In 2023, China dominated in global cultivation area, yield, and consumption for watermelon and melon (https://www.fao.org/faostat/en/#home, FAOSTAT, 2023 (accessed on 29 July 2025)). Recent changes in cultivation practices and escalating consumer demand for complex quality traits have highlighted the limitations of conventional breeding, particularly when it comes to genetic homogeneity. At the same time, innovations in crop breeding technology—notably smart breeding, a cornerstone of modern agricultural biotechnology—are fundamentally transforming genetic improvement methodologies. The evolution of breeding strategies has evolved from domestication and crossing through molecular breeding to today’s cutting-edge approach: smart breeding. The core technologies, such as CRISPR/Cas-mediated gene editing, have become powerful tools for improving crops, enabling precise modification of traits such as disease resistance, stress tolerance, developmental processes, and secondary metabolism [1]. Moreover, smart breeding integrates environmental and crop growth models, allowing breeders to assess the field performance of new varieties and optimize strategies, significantly accelerating the breeding process and enhancing the innovation capacity of varieties [2].

However, the efficacy of both molecular breeding and gene-editing precision breeding critically depends on the identification, mining, and integration of genes related to excellent traits, which is the foundation for generating new germplasms and cultivating new varieties with high yield, good quality, and resistance to disease and pest. Therefore, the key to achieving the transformation of breeding technology to the molecular and intelligent level lies in elucidating the functional genomics of key traits in watermelon and melon—specifically for plant architecture, fruit quality, and disease resistance. This article reviews the progress and persistent limitations of the research on the location and function of genes related to the above important traits over recent decades, as well as discussing the challenges and prospects faced by future breeding.

2. Research Progress on Gene Mapping and Functional Characterization of Plant Architecture-Related Traits

Plant architecture—a core agronomic trait that determines crop yield—is precisely regulated by a complex genetic network. Deciphering its molecular regulatory mechanism and identifying key genes are crucial for developing high-yielding varieties [3,4]. As vining annuals, ideal plant architecture in watermelon and melon maximizes light-use efficiency, land/space utilization, and production efficiency and facilitates labor-saving cultivation [5,6]. Current research mainly focuses on key plant architecture traits: plant height, branching patterns, and leaf morphology. Therefore, a comprehensive analysis of the genetic basis, key functional genes, and regulatory mechanisms for these traits will provide valuable breeding targets and establish theoretical support for ideotype breeding in watermelon and melon.

2.1. Short-Internode or Dwarfing Genes

Plant height, a primary architectural trait determined by internode number and length, directly impacts planting density, photosynthetic efficiency, and field management. Dwarf plants have gained particular attention due to the advantages they offer, including improved lodging resistance, tolerance to higher planting density, enhanced light capture, and simplified crop management. Within protected cultivation systems for watermelon and melon, compact (or short-internode) architecture is particularly advantageous for yield optimization and cost reduction.

Genetic research of watermelon dwarfism began in the 1950s. Mohr [7] initially reported a recessive single-gene locus (dw-1) controlling a short-internode, bushy phenotype, revealing that suppressed internode cell division and elongation underlie dwarfism. This research established single-gene recessive inheritance as a key genetic mechanism for internode regulation. Liu et al. [8] identified another non-allelic dwarfing gene, dw-2, with shorter internodes. Subsequently, Dyutin and Afanasieva [9] characterized a short-internode germplasm (“Somali Local”) with a phenotype similar to dw-1 but exhibiting longer hypocotyls, proposing control by a recessive gene (dw-1s) potentially allelic or closely linked to dw-1. Huang et al. [10] and Guner and Wehner [11] analyzed a short-internode male-sterile mutant, revealing a new gene dw-3 controlling internode length, and a single recessive nuclear gene for sterility. Male sterility co-segregated with the short-internode phenotype, and dw-3 expression was epistatic to dw-1 and dw-2.

Recent advances in molecular biology techniques have enabled fine-mapping and functional characterization of dwarfing genes in watermelon. To date, at least six dwarfing-related genes have been identified in watermelon, five of which are involved in gibberellin (GA) biosynthesis or signaling pathways. Dong et al. [12] established that a ClGA20ox (encoding GA 20-oxidase) mutation causes dwarfism in the dsh mutant. Wei et al. [13], Gebremeskel et al. [14], and Sun et al. [15] independently mapped the same short-internode gene Cla015407 (encoding GA 3β-hydroxylase) across distinct populations. Jang et al. [16] identified a neighboring short-internode gene (Cla015405, encoding GA 2-oxidase) on chromosome 9, suggesting a potential GA-related gene cluster with unknown co-regulatory mechanisms. Zhang et al. [17] confirmed that a ClGA3ox (encoding GA 3β-hydroxylase) mutation led to the Cldw mutant, where its phenotype was rescued by exogenous GA application. Liu et al. [18] identified the gibberellin receptor gene ClGID1L2 as the candidate gene for the short-internode dwarf mutant Cladw. Additionally, non-GA pathway regulators have also been identified. Zhu et al. [19] mapped Cldw-1, an ABCB transporter modulating height via the auxin transport pathway. Sun et al. [20] revealed that ClDUF21 regulates watermelon plant height by participating in the ClDWF1-mediated brassinosteroid biosynthesis pathway, revealing novel regulatory mechanisms.

Melon dwarfism research originated with the discovery of the recessive gene si-1 [21]. Paris et al. [21] further identified a non-allelic gene si-2 in another short-internode cultivar “Persia 202”, which affects height, leaf size, and fruit shape. Knavel [22] discovered a new recessive gene si-3 in the phenotypically stable short-internode mutant “Main Dwarf”. Hwang et al. [23] mapped a new recessive gene mdw1 in the short-internode mutant “PNU D-1”, positioned at genetic distances of 1.7 cM, 0.6 cM, and 1.2 cM from the candidate genes CYP735A (cytokinin oxidase), ERECTA-like kinase, and ubiquitin gene (UBI), respectively. This study significantly advanced our understanding of the genetic basis of dwarfism in melon. Zhang et al. [24] confirmed that the short-internode trait in “X090” is controlled by a recessive gene (MD7) on chromosome 7. While these studies have mapped key genetic loci, their candidate genes and regulatory mechanisms are yet to be elucidated. From a breeding perspective, Zink used the short-internode germplasm “Big River Bush” to develop compact lines (<1.5 m vine, e.g., “Grenshaw Bush”, “Honey Dew Bush”, and “Top Mark Bush”) [25]. Similarly, Halsey used the short-internode germplasm “MCa66-18-14” to breed ultra-compact lines (~1 m vines, e.g., “U.FG508”, “6509”, and “GS10”) [26].

A key molecular breakthrough was achieved by Yang et al. [27], who successfully cloned the first short-internode gene Cmsi which encodes an ERECTA-like protein kinase primarily expressed in vascular bundles. Ectopic overexpression of Cmsi in cucumber and Arabidopsis enhanced internode elongation. Importantly, CmSI interacts with the auxin transporter CmPIN2, which links ERECTA signaling to auxin transport during vine development [27]. Despite this progress, only a limited number of dwarfing/short-internode genes have been cloned and functionally characterized in watermelon and melon. Future efforts should focus on identifying additional regulatory genes to elucidate molecular networks, thereby providing essential genetic resources for new novel germplasm development.

2.2. Lateral Branch

As a key architectural component, lateral branches modulate nutrient allocation and light capture in plants [28]. Regulating branch number is therefore essential for enhancing yield, quality, and cultivation efficiency, driving research into developmental mechanisms. Many important regulators have been identified across species, including MOC1/3, LAX1, LAX2, FON1, and SHI1 in rice [29,30,31,32,33]; TaD27-B in wheat [34]; LS, Blind, RAX, ZFP2, and MAPK1 in tomato [35,36,37,38]; and ll and BRC1 in cucumber [39,40]. Representative examples also include the maize tiller control gene TB1 and the rice ideal plant architecture gene IPA1 [41,42]. In addition to genetic factors, branch development is precisely modulated by hormones: auxin suppresses branching, whereas cytokinin (CK) acts downstream to promote it [43]. In tomato, strigolactones (SLs), downstream of auxin, inhibit branching by suppressing CK synthesis, upregulating bud suppression genes [44], and disrupting polar auxin transport [45]. In rice, SLs and abscisic acid (ABA) synergistically regulate tillering—SLs primarily suppress basal tillers, while ABA inhibits upper-node tiller formation [46].

Research on lateral branches in watermelon and melon remains relatively limited, primarily focusing on phenotypic characterization and genetic mapping. Few regulatory genes are reported in watermelon. Yi et al. [47] reported a mutant affecting on branching, tendril, and inflorescence development, controlled by a single recessive gene. Dou et al. [5,6] demonstrated that the reduced branching is governed by a recessive gene ClTFL1, which encodes TERMINAL FLOWER 1 (TFL1), where Cltfl downregulation in axillary and apical buds prevents branch formation beyond the fifth node. Jiang et al. [48] identified the lateral branch-promoting gene ClLAS. In breeding applications, Dou et al. [49] crossed a dwarf inbred line (WM102) and a branchless inbred line (WCZ) to develop a stable dwarf/branchless double mutant (DM). Using DM as a donor, they developed near-isogenic lines (NILs) with two elite traits, providing valuable genetic resources for plant architecture improvement and commercial production in watermelon.

Research on melon branching remains limited despite its cultivation significance. Zalapa et al. [50] reported that high heritability for branch number and branch length significantly influences architecture. Ohara et al. [51] found that short branches are controlled by a recessive or incompletely dominant gene exhibiting genotype-by-environment interaction. Fukino et al. [52] performed QTL mapping using parents with contrasting branch lengths, identifying two QTLs with a major locus on LGXI (PVE = 50.9%) and validating linked markers for selection—establishing the first genetic map for melon architecture. Fang et al. [53] observed that silencing CmSVPc shortened branch length but caused developmental defects. To date, no lateral branch regulatory gene has been cloned via forward genetics in melon.

2.3. Leaf Shape

Leaf shape is also a key plant architecture trait in watermelon, but remains genetically under explored. The entire leaf phenotype—characterized by a wavy margin lacking typical serrations or lobes—is controlled by a recessive gene (sn). Wei et al. [54] fine-mapped sn to a 127.6 kb region on chromosome 4. Subsequent research identified the candidate gene ClLMI1 (encoding an HD-Zip protein), with significantly downregulated expression in mutants [55]. Functional conservation of LMI1 in rapeseed (Brassica napus) and cotton (Gossypium hirsutum) [56,57] underscores its importance, though its regulatory networks remain unclear. In melon, lobed leaves are dominant, while deeply incised leaves are recessive [58]. The recessive gene pll (controlling lobe depth) was mapped to an approximately 14.6 kb interval between markers G69 and 784RS on linkage group 3 [59]. Overall, genetic and regulatory networks for leaf margin morphogenesis in cucurbits remain largely unknown.

2.4. Leaf Color

Leaf color variation stems from carotenoid synthesis, chlorophyll metabolism, photosynthesis, secondary metabolite accumulation, or chloroplast development defects [60]. Comparative analyses of differentially abundant metabolites (DAMs) and differentially expressed genes (DEGs) suggest that photosynthesis, carbohydrate metabolism, and hormone signaling pathways collectively modulate color via pigment synthesis [61]. Given that increased chlorophyll enhances yield and quality [62,63,64], leaf color mutants serve as valuable models for chlorophyll and photosynthesis studies [65], underscoring significant breeding potential for crop improvement.

Reports on watermelon leaf color mutants are scarce. Barham [66] first described “Royal Golden” (golden leaves, stems, and rind), controlled by the recessive gene go. Subsequently, mutants Yl1 [67] and Yl2 [68], exhibiting yellow aerial organs throughout their life cycle, are controlled by incompletely dominant genes, though homozygotes are albino-lethal. Zhu et al. [69] reported a recessive yellow mutant (w-yl), mapping the locus to a 2.217 Mb region on chromosome 2 via bulked segregant analysis (BSA). Sequence analysis revealed deletions in five genes; virus-induced gene silencing (VIGS) preliminarily implicated Cla97C02G036040, Cla97C02G036050, and Cla97C02G036060 as primary candidate genes. Kidanemariam [70] identified the dg mutant (delayed apical greening), implying that ClFtsH encodes a chloroplast outer membrane protease. ClFtsH is highly expressed in normal leaves, with expression gradually decreasing during leaf development. Cloning and functional characterization of yellowing genes remain limited, necessitating mechanistic studies. Melon research is even more limited, with only Han et al. [71] reporting a yellowish-green mutant to date.

3. Advances in Functional Genes of Fruit Quality Traits

Fruit quality is a crucial factor in consumer preferences and the breeding of new varieties. Parameters of external quality such as fruit shape, size, and rind color, as well as internal quality attributes including texture, sugar–acid ratio, flavor, and phytohormone levels, critically influence consumer decision-making. The formation of fruit quality involves a series of physiological and biochemical changes, such as pigment accumulation, fruit softening, and synthesis of aroma and flavor compounds. Recent advances in genomics, phenomics, metabolomics, and molecular biology, augmented by precision gene-editing platforms, have accelerated the discovery and functional validation of key regulatory genes governing quality traits across crops [1,2,72,73]. This part will focus on reviewing functional genes of fruit quality traits in watermelon and melon. Detailed information of the relevant genes is summarized in Tables S1 and S2. We have also drawn a summary model to display the genes related to plant architecture and fruit quality traits in watermelon and melon in Figure 1.

3.1. Rind Color and Stripe Patterns

Recent advances in watermelon rind color and patterning genetics reveal a complex interplay of genetic and epigenetic factors. These factors collectively govern phenotypic diversity. Historically, classical genetic models proposed allelic variation at the g locus [74,75], later refined into a five-allele series (G > gW > gM > gN > g) or alternative three-gene systems (S, D, Dgo) controlling stripe patterns, color depth, and background pigmentation [76,77,78]. Modern genomic technologies (Illumina, PacBio, Hi-C) have enabled high-resolution genome assemblies [79,80], facilitating the identification of key genes. By using natural populations and the genome-wide association study (GWAS) method, the gene Cla97C06G126710, encoding a WD40-repeat protein, has been identified as the likely candidate gene for rind stripe presence [80]; Cla019205 was identified as the candidate gene for dark-green stripes in watermelon [81]. The APRR2 transcription factor gene Clsc controls the colorless stripe phenotype in watermelon and is related to chlorophyll synthesis and chloroplast development [82]. The KNOX transcription factor gene ClSP regulates dark-green stripe formation by activating the expression of the chlorophyll-regulating transcription factor ClAPRR2, revealing the ClSP-ClAPRR2 module as a crucial pathway for stripe patterning [83,84]. Additionally, the gene ClCG08G017810 on chromosome 8 has been recognized as a potential regulator of outer layer color in watermelon [85], and the yellow rind gene was mapped to a 91.42 kb region on chromosome 4 [86].

In melon, the CmAPRR2 was identified as a key regulator of the green/white rind color during ripening, with a single-base mutation causing significant changes in fruit color [87]. The rind traits of melon are regulated by multiple genes. According to previous studies, the striping pattern of the rind is controlled by the st3 gene, while the non-striped phenotype is determined by a recessive allele [88]. Additionally, mt gene controls spotted patterns on the rind, acts as a dominant gene, and exhibits epistatic effects similar to other genes such as Y and st [89]. In addition to the known genes, another allele st-2 affects rind striping, though its genetic relationship remains unclear [90]. Similarly, a variant of mt designated mt-2 was reported, but its exact genetic role has not yet been fully elucidated [91]. Further research suggests that rind color and spot patterning may be controlled by distinct genes, with spk potentially regulating spot formation [90]. Additionally, the netting trait of the rind is modulated by multiple genes including Rn (N), which contributes to netting development, while netting density may be influenced by a single gene [89]. The CmSN gene, controlling melon rind netting via an EamA-like transporter, was fine-mapped to chromosome 2, with functional SNPs and GWAS evidence implicating its domestication role, aiding molecular research and breeding [92].

3.2. Flesh Color

Recent studies have systematically elucidated the genetic architecture underlying the remarkable diversity of watermelon flesh colors, ranging from white, pale-yellow, canary-yellow, salmon-yellow, and orange to various red shades (coral and scarlet). The color spectrum is determined by the composition and concentration of specific carotenoids, where lycopene produces red hues while xanthophylls yield yellow pigmentation. A “two-switch” genetic model [93] involving a CNV in ClREC2 and an SNP in ClLCYB explains 99.7% of this variation by regulating both carotenoid biosynthesis and chromoplast development. Key molecular mechanisms include the following: (1) ClLCYB (lycopene β-cyclase) polymorphisms reduce protein stability, blocking lycopene-to-β-carotene conversion and resulting in red flesh [94,95]; (2) ClPSY1 nucleotide variations are associated with β-carotene which accumulates in golden-yellow flesh [96]; (3) ClPAP-mediated chromoplast formation influences lycopene crystallization; (4) mutations in isomerase genes (ClZISO and ClCRTISO) cause photosensitive yellow and orange flesh phenotypes, respectively [97,98]. Additional loci include Y^scr on chromosome 6 for scarlet red flesh [99], a nearby region controlling white flesh [100], and qfc10.1/QTL regulating pale-green flesh through chloroplast development [101]. These findings collectively reveal a hierarchical regulatory network where structural genes (ClLCYB, ClPSY1), isomerases (ClZISO, ClCRTISO), and plastid-associated proteins (ClPAP, ClREC2) interact with major QTLs to determine carotenoid profiles and chromoplast ultrastructure, providing both molecular markers for breeding and fundamental insights into flesh color diversification. In melon, two major genes controlling flesh color have been identified—the orange flesh gene gf and the white flesh gene wf. CmOr (MELO3C005449), identified as the gf gene, is a homolog of cauliflower BoOr and regulates β-carotene accumulation. Located on chromosome 9, CmOr exhibits epistatic dominance over wf [90,102,103]. The wf gene, associated with white or green flesh, maps to chromosome 8 [104], with two candidate genes: MELO3C003069, encoding a pentatricopeptide repeat (PPR) protein involved in plastid RNA processing (where carotenoid and chlorophyll pigments accumulate), and MELO3C003097, a homolog of Arabidopsis SG1 (essential for chloroplast development and chlorophyll biosynthesis). Notably, MELO3C003097 shows significantly higher expression in green-fleshed melons compared to white-fleshed varieties during fruit development [103,105]. Additionally, in oriental melon, the transcription factors CmWRKY49 and CmNAC34 modulate carotenoid biosynthesis by activating the key pathway genes CmPSY1 and CmLCYB31 [106]. The variation in flesh color is critical for the genetic improvement of watermelon and melon, as it serves as a key indicator for enhancing nutritional quality and market appeal.

3.3. Sugar and Acid Metabolism

Recent studies have significantly elucidated the molecular mechanisms governing sugar and acid metabolism in watermelon and melon, revealing key genetic determinants of fruit quality. In watermelon, the vacuolar sugar transporter ClVST1 was identified as a critical regulator of sucrose accumulation [107]. Subsequent research demonstrated that during domestication, the alkaline α-galactosidase ClAGA2—whose expression is modulated by promoter SNPs and the transcription factor ClNF-YC2—played a pivotal role in raffinose hydrolysis, while the sugar efflux transporter ClSWEET3 mediated apoplastic sugar export and the tonoplast-localized ClTST2 facilitated vacuolar sugar storage, collectively driving the evolution of sweet modern cultivars from their non-sweet progenitors [108]. Further mechanistic insights revealed that the NAC transcription factor ClNAC68 functions as a transcriptional repressor, directly inhibiting the expression of the sucrose-cleaving invertase ClINV and auxin-degrading enzyme ClGH3.6, thereby coordinately enhancing sucrose retention and auxin accumulation to promote fruit quality and seed development [109]. In melon, CmTST2 was also identified as an essential component for sugar accumulation [110]. These findings collectively establish a molecular framework for targeted genetic improvement of fruit quality in cucurbits through precise manipulation of sugar–acid homeostasis. Significant progress has also been made in elucidating the genetic control of fruit morphology in watermelon. ClSUN (IQD family protein) was found to be a major determinant of oval fruit shape [111], while ClRPK2 (receptor-like protein kinase 2) regulates fruit size through auxin signaling [112].

4. Progress in Disease Resistance-Related Genes

Diverse plant pathogens cause significant yield losses and global food security threats. Recent progress in understanding plant–pathogen interactions at the molecular level, combined with biotech advances, provides strong support for genetically breeding disease-resistant crops [113,114]. Watermelon and melon production have been threatened by major diseases worldwide, such as fusarium wilt (FW), powdery mildew (PM), and gummy stem blight (GSB). This part will focus on reviewing the genetic inheritance, fine-mapping, cloning, and characterization of candidate resistance genes in watermelon and melon. The major QTLs of disease resistance identified in watermelon are summarized in

Table 1. Summary of disease-resistance QTL identified in watermelon.

Diseases	Races	Resistance Sources	QTL a	Notes
FW	Fon 1	HMw017	Fo1.1 **, Fo1.2 *, Fo1.3 *, Fo1.4 *, Fo1.5 **, Fo1.6 *, Fo1.7 *
	Fon 1	PI 296341-FR	Fo1.1 **	derived from PI 296341
	Fon 1	USVL246-FR2	Fon1-9 **	derived from PI 482246
	Fon 2	PI 296341-FR	fon2.1 *, fon2.2 *	derived from PI 296341
	Fon 2	USVL252-FR2	Fon2-1 **, Fon2-2, Fon2-5, Fon2-6, Fon2-8.1, Fon2-8.2, Fon2-11	derived from PI 482252
	Fon 2	UGA147	Fon11 **	derived from PI 169233
	Fon 2	USVL246-FR2	Fon2-2 *, Fon2-5, Fon2-8, Fon2-9 **, Fon2-10	derived from PI 482246
	Fon 2	USVL246-FR2	Fon2-1, Fon2-6, Fon2-8, Fon2-9.1, Fon2-9.2 **	derived from PI 482246
PM	1 W	Arka Manik	pmr2.1 **
	2 WF	R23	pm-lox
GSB	Isolate JS002	PI 189225	gsb8.1 **
	Isolate 12178A	PI 482276	gsb3.1 *, 5.1 *, 7.1 **
AR	Race 1	DrHs7250	n/a
ZYMV	FL	PI 595203	zym-FL
CGMMV	n/a	PI 595203	WPRb

^a * PVE (percentage of phenotypic variance explained) = 10–15%; ** PVE > 15%; underlined: this QTL comes from the susceptible parent; n/a: not available.

. Detailed information for each QTL is provided in Tables S3 and S4.

4.1. Fusarium Wilt (FW)

Fusarium wilt (FW) on watermelon is caused by the soil-borne pathogen Fusarium oxysporum f.sp niveum (Fon), among which four Fon races (0–3) have been described [115,116,117,118]. The resistance to Fusarium wilt in watermelon is considered to be a quantitative trait controlled by multiple genes. Based on current research findings, different resistance genes have been identified against different races. Genetic analysis of resistance to Fon race 1 using F₃ families derived from the resistant parent HMw017 and the susceptible parent HMw013 revealed that the resistance is controlled by one major QTL Fo1.1 and several minor QTLs [119]. This major QTL was confirmed by an independent study using an RIL (Recombinant Inbred Line) population from a cross between PI 296341-FR and the cultivar 97103. This QTL was narrowed down to a 364 kb region by aligning its published interval, and three disease-resistance candidate genes were identified: ClG42_01g0002300 (cytochrome b5), ClG42_01g0002600 (major facilitator superfamily domain-containing), and ClG42_01g000440 (pectinesterase inhibitor) [120]. QTL analysis of FW resistance to Fon race 2 has been conducted using different populations. A QTL study conducted by Ren et al. [121] identified two QTLs of Fon race 2 resistance (Qfon2.1 and Qfon2.2) on chromosomes 9 and 10. A major QTL (qFon2-1, PVE = 18.9%) and three minor QTLs (Qfon2-5, Qfon2-6, and Qfon2-8.2, collectively accounting for PVE = 23.8%) were identified using F₂ and F_2:3 families derived from the cross USVL252-FR2 × PI 244019^PRSV-R [122]. Another QTL analysis of FW resistance to Fon race 2 using F₂ and F₃ families from a cross between UGA147 (resistant) and Charleston Gray (susceptible) identified a QTL (Qfon11) on chromosome 11 [123]. Furthermore, the same resistance source can confer resistance to multiple Fon races. Three QTL studies of FW resistance in USVL246-FR2 indicate that the same genomic region (on chromosome 9) confers resistance to both race 1 and race 2 [124,125,126]. FW on melon is caused by the soil-borne pathogen Fusarium oxysporum f.sp melonis (Fom), among which four Fom races (0, 1, 2, and 1.2) have been described. Thus far, four Fusarium wilt resistance genes (Fom-1, Fom-2, Fom-3, and Fom-4) have been identified in melon, each demonstrating distinct resistance patterns against different physiological races [127,128,129,130]. Using a map-based cloning strategy, researchers confirmed that MRGH9 corresponds to the Fom-1 gene [131]. Later, EI Otmani et al. [132] revealed that the Fom-1 gene encodes a TIR-type NBS-LRR resistance (R) protein.

4.2. Powdery Mildew (PM)

Powdery mildew (PM), caused by Podosphaera xanthii, is a globally significant fungal disease affecting watermelon. Due to the pathogen’s high race diversity and the complex nature of resistance sources, the genetic inheritance of resistance varies among different races of PM. Several studies support that PM resistance is predominantly governed by a single dominant gene [133,134,135]. However, Ben-Naim and Cohen [136] observed polygenic control at later watermelon growth stages, suggesting that the genetic control of PM resistance is dependent on the developmental stage. A major QTL pmr2.1 (PVE = 80%) was mapped to chromosome 2 in an F₂ population derived from a cross between the resistant “Arka Manik” and TS34 [137], and its candidate gene is ClaPMR2 which encodes an NBS-LRR protein homologous to Arabidopsis’s RPW8. Another genetic mapping study confirmed a single dominant locus on chromosome 2 [135]. Furthermore, a study identified a QTL (pm-lox) on chromosome 2 conferring PM resistance in “R23” and confirmed its candidate gene as ClLOX (ClG42_02g0161300) through fine-mapping. This study also found that it confers PM resistance by restricting the pathogen’s spread instead of restricting its infection [138]. PM on melon is predominantly caused by the biotrophic fungi Podosphaera xanthii, and numerous races have been identified, such as races 0, 1, and 2F. Extensive research has been conducted to investigate the genetic architecture of PM resistance in melon, which has been nicely reviewed by Cui et al. [139]. More recently, a QTL analysis reported qCmPMR-12 for resistance to PM on chromosome 12 (22–22.9 Mb), where MELO3C002434 gene encoding an ankyrin repeat-containing protein was the most likely candidate gene through RNA-Seq analysis [140]. Another investigation into the genetic basis of PM resistance to P. xanthii race 1 from the resistant accession “PI 164637” identified a QTL on chromosome 6, of which the candidate region was narrowed down to a 63.5 kb region [141].

4.3. Gummy Stem Blight (GSB)

Another major threat to watermelon production is gummy stem blight (GSB), caused by the fungal pathogen Stagonosporopsis citrulli [142,143,144]. Initial studies of GSB resistance in PI 189225 reported that a single gene, db, controls GSB resistance [145]. However, re-evaluation of PI 189225 and studies on PI 482283 and PI 526233 reported that GSB resistance was controlled by multiple genes with minor effects [146,147,148]. Recently, a major QTL underlying GSB resistance in PI 189225 was reported on chromosome 8, which explains 32% of the phenotypic variance [148]. Another QTL study of GSB resistance in an F_2:3 mapping population identified three QTLs, ClGSB3.1, ClGSB5.1, and ClGSB7.1 (PVE = 6.4–21.1%). The genes underlying the ClGSB5.1 locus contain an NBS-LRR gene (ClCG05G019540), which has been previously identified as a candidate gene for GSB resistance. ClCG07G013230, one of the candidate genes for ClGSB7.1, encodes an Avr9/Cf-9 rapidly elicited disease-resistance protein, which carries a non-synonymous mutation that was significantly associated with GSB resistance [149]. GSB on melon is caused by the fungal pathogen Didymella bryoniae. In the past decades, genetic analyses have identified five independent monogenic dominant resistance loci (Gsb-1, Gsb-2, Gsb-3, Gsb-4, Gsb-6, and Gsb-7(t)) and one recessive resistance locus (gsb-5) conferring resistance to GSB [150,151,152,153]. Yang et al. [154] reported two QTLs for GSB resistance in a melon inbred line HS. Furthermore, a single dominant gene (GsbR) was mapped onto a 108 kb region on chromosome 4 [155], and Gsb-7(t) was delimited to a 140 kb interval on chromosome 7 [153].

4.4. Other Diseases

It is commonly believed that anthracnose (AR) resistance in watermelon is controlled by a single dominant gene [156,157]. Furthermore, Jang et al. [157] reported a non-synonymous SNP located in a leucine-rich repeat domain associated with resistance to AR race 1 in watermelon. Virus diseases, such as Cucumber green mottle mosaic virus (CGMMV) and Zucchini yellow mosaic virus (ZYMV), bring a considerable loss in production and quality to the watermelon industry. However, genetic resources are quite limited for resistance against viruses. Ling et al. [158] reported eIF4E as a candidate gene associated with resistance against ZYMV-FL. To elucidate the genetic mechanisms underlying CGMMV resistance in watermelon, a recent study revealed the recessive gene WPRb providing CGMMV resistance [159]. This study also discovered that the editing of WPRb enhanced resistance to CGMMV; WPRb targets plasmodesma (PD) and physically binds the CGMMV movement protein, promoting viral cell-to-cell movement by modulating PD permeability. The distribution of identified genes or QTL loci related to disease resistance, plant architecture, and fruit quality in watermelon and melon on chromosomes is summarized in Figure 2.

Overall, in the past decades, extensive efforts have been made to investigate the genetic basis of disease resistance in watermelon and melon. However, conventional genetic analysis and fine-mapping are labor-intensive and constrained by the restricted genetic background of the available germplasm. Additionally, the genetic basis of disease resistance can be affected by multiple factors, including epigenetic regulation, gene–gene/environment interactions, which makes it more complex to decipher the genetic mechanisms of disease resistance. Subsequent research should focus on improving the disease response evaluation system by incorporating high-throughput phenotyping platforms and accelerating the mining of disease-resistance genes by involving high-throughput sequencing.

5. Future Perspectives

The rapid advancement of cutting-edge technologies—including high-throughput sequencing, gene editing, multi-omics analysis, and artificial intelligence—presents unprecedented opportunities for functional genomics research in watermelon and melon. While these tools offer transformative potential for molecular breeding, several key challenges need to be addressed to fully realize their capabilities.

5.1. Systematic Discovery and Characterization of Superior Trait Genes

Current efforts to identify and clone genes governing key agronomic traits in watermelon and melon remain limited. The narrow genetic diversity within available germplasm resources, combined with insufficient characterization of genes regulating plant architecture, fruit quality, and disease resistance, significantly hinders the precision breeding progress. Although genomic resources, such as high-quality reference genomes, resequencing data, and variation maps are increasingly accessible, systematic large-scale mining of superior trait alleles remains insufficient. Research predominantly focuses on a few major traits including plant architecture, soluble solids content, fruit morphology, and disease resistance, which leaves the genetic basis of complex traits largely unexplored. Future studies should employ larger, more genetically diverse natural and artificial populations, integrating GWAS and quantitative trait locus (QTL) mapping, to improve both the precision of gene localization and the discovery of novel loci. Given that a single reference genome fails to capture the full genetic diversity of a species, the assembly and application of watermelon and melon pan-genomes will be essential to identify rare alleles and structural variations (SVs). Systematic integration of multi-omics data—including genomics, transcriptomics, epigenomics, metabolomics, and proteomics—with advanced bioinformatics and AI-driven approaches to construct comprehensive gene regulatory networks will dramatically improve the efficiency of candidate gene prediction and screening.

5.2. Deepening Functional Analysis and Regulatory Mechanism Elucidation

The number of genes functionally characterized in regulating key traits in watermelon and melon is limited. Understanding their regulatory mechanisms—particularly transcriptional regulation, protein–protein interactions, signal transduction pathways, and gene regulatory networks across diverse genetic backgrounds and environmental conditions—requires further investigation. Future research should implement an integrated strategy: generating precise mutants using CRISPR/Cas9 systems (e.g., base editing, prime editing) and combing physiological, biochemical, cellular, and multi-omics analyses to fully characterize gene function. Techniques like ChIP-seq, ATAC-seq, and DAP-seq should be routinely applied to resolve the transcriptional regulatory networks governing key genes.

5.3. Accelerating Molecular Breeding System Development and Smart Breeding Integration

Conventional breeding methods are often characterized by lengthy breeding cycles, inefficiency, and low predictability, making it difficult to meet the market demands for high-yield, superior-quality, and multi-resistant cultivars. Establishing an efficient molecular breeding system is the core strategy to transform watermelon and melon breeding from “experience-based selection” towards “precise design”. Achieving this demands the systematic integration of multi-omics resources and smart breeding technologies to accelerate the convert of superior gene discoveries into germplasm innovation and cultivar development.

Genomic Resources as the Foundation: High-quality reference genomes underpin effective molecular breeding systems. While multiple high-quality watermelon genomes are available [79,80,160,161,162], melon genomics has also advanced significantly. Zhao et al. [105] constructed the first comprehensive melon variation map. Li et al. [163] assembled a high-quality telomere-to-telomere (T2T) genome of the semi-wild melon “821”, facilitating research on resistance and quality traits. Mo et al. [164] generated the first T2T gap-free melon genome and constructed a pan-nucleotide-binding leucine-rich repeat genome (pan-NLRome), identifying 226 disease-resistance genes. Chen et al. [165] released the genome of wild melon “P84”, identifying 10589 SVs and locating a key fruit acidity gene (CmPH) via GWAS. However, the complex diversity within melon germplasm demands more refined genetic mapping and pan-genome analysis. Recent breakthroughs in high-throughput genotyping, such as SNP liquid-phase chips for watermelon and melon [166,167], provide cost-effective solutions for population genetics, gene mapping, and marker-assisted selection, thereby enhancing breeding efficiency and offering valuable models for other horticultural crops. Future efforts need to prioritize deeper integration of diverse genomic resources (e.g., SVs, epigenomic datasets) and the comprehensive analysis of complex trait regulatory networks to provide precise targets for smart breeding.

Overcoming Transformation Bottlenecks: Despite the critical importance of transgenic and gene-editing technologies for functional genomics and crop improvement, their broad application is constrained by low transformation efficiency. Overcoming the genetic transformation bottleneck in watermelon and melon is therefore essential. Recent progress includes the following: Pan et al. [168] improved watermelon transformation efficiency to 25% using AtGRF5. Feng et al. [169] established a genotype-independent system based on ClGRF4-GIF1 fusion. Cao et al. [170] optimized key parameters for watermelon transformation. Zhao et al. [171] developed an Agrobacterium rhizogenes-mediated hairy root system for rapid sgRNA assessment. Gu et al. [172] created three highly efficient watermelon transformation systems. Li et al. [173] demonstrated that AtGRF5 overexpression enhances transformation efficiency across cucurbits. Li et al. [174] significantly improved melon transformation efficiency via optimizing Agrobacterium tumefaciens protocols. Despite these advances, current efficiencies remain insufficient to support large-scale gene function validation. Sustained efforts are required to establish highly efficient, stable, and genotype-independent transformation systems that will accelerate functional validation and breeding applications of key trait genes.

Furthermore, the precise acquisition and analysis of large-scale phenotypic data remain major bottlenecks constraining both breeding and functional research [2]. Future work needs to prioritize deciphering complex trait regulatory networks, developing high-throughput automated phenotyping platforms to overcome data collection challenges, and building AI-powered decision models to accelerate the transformation of research findings into improved varieties in practical breeding.

5.4. Epigenetics: A Frontier in Molecular Breeding of Watermelon and Melon

Epigenetics is defined as heritable changes in gene expression that are, unlike mutations, not attributable to alterations in the sequence of DNA. Epigenetic changes include various modifications, such as cytosine methylation, non-coding RNAs, RNA modifications, histone modifications, and chromatin remodeling. Epigenetic mechanisms play significant roles in controlling plant growth, development, and adaptation to environmental stimuli [175,176]. Recent advances in high-throughput sequencing and multi-omics approaches have ushered plant epigenetics into a new phase of multi-level, systematic exploration [177]. Understanding of epigenetically mediated gene regulation has significantly advanced in model plants and has been extensively characterized in major crops such as rice, maize, and soybean. Notably, these mechanisms have been comprehensively shown to regulate key agronomic traits, including yield, quality, and stress resistance [178,179]. Emerging concepts such as the identification and exploitation of epialleles, the construction of epigenomic maps, and epigenetic marker-assisted breeding are accelerating the application of epigenetic research in crop improvement [175]. Additionally, single-cell epigenomic analysis and artificial intelligence techniques (e.g., machine learning and deep learning) are propelling epigenetics toward a more systemic, precise, and application-oriented framework beyond the single-gene level [180,181,182]. Thus, epigenetic research enhances our understanding of gene regulatory networks in plants and offers new theoretical frameworks and technological tools for molecular breeding.

As important economic crops, watermelon and melon have achieved significant progress in elucidating the genetic basis and mapping genes associated with traits like plant architecture, fruit quality, and resistance/tolerance to biotic and abiotic stresses. However, compared to staple field crops and some other horticultural plants, epigenetic research and its practical applications in watermelon and melon remain relatively limited. Current studies demonstrate that epigenetic regulatory mechanisms, such as DNA methylation and histone modifications, play crucial roles in fruit development, quality regulation, sex differentiation, and environmental adaptation [183,184]. Moreover, in watermelon and melon, traits like sex determination and rind coloration are intricately linked to epigenetic modifications [185,186,187]. Studies in horticultural crops including tomato, apple, and strawberry have demonstrated that epigenetic regulation offers significant breeding potential in traits like fruit ripening, flavor accumulation, and improved stress tolerance [188,189,190]. Although systematic epigenomic studies are limited in watermelon and melon, existing theoretical frameworks and technological foundations offer a strong basis for advancing research and applications in this field.

In keeping with the advancing field of epigenetics, more comprehensive epigenomic studies are urgently needed in watermelon and melon. Future research should focus on elucidating epigenetic regulatory networks underlying key agronomic traits, construct high-resolution epigenomic maps across developmental stages, tissues, and environmental contexts, and identify stably inherited epigenetic variants associated with phenotypic variation. These efforts will strengthen the theoretical foundation and broaden technological pathways for molecular breeding in watermelon and melon. Meanwhile, the integration of multi-omics data with advanced algorithms will enable the development of efficient trait prediction and molecular breeding models based on epigenetic information. This strategy holds great potential to accelerate the selection and development of high-quality, high-yielding, and stress-tolerant cultivars, thereby advancing varietal innovation and supporting the sustainable development of the watermelon and melon industry. In conclusion, advancing epigenetic research and its application in the molecular breeding of watermelon and melon is essential for uncovering the mechanisms of complex trait formation and represents a crucial step toward future precision breeding.