Channel catfish, Ictalurus punctatus (Rafinesque, 1818), also known as American catfish is an economically important freshwater farmed fish in China and a pillar species in global aquaculture. The species exhibits a male heterogametic sex-determination system (XX/XY) (Zhang et al. 2019), with distinct growth dimorphism between the sexes; males display a clear growth advantage during the later stages of commercial culture, resulting in higher profitability per unit area (Davis et al. 2007). Consequently, producing all-male channel catfish through sex-control technologies has become a key breeding strategy to enhance farming efficiency and economic returns from channel catfish culture (Xu et al. 2024; Zhong et al. 2018).

Sex differentiation in fish is a complex biological process regulated by genetic, hormonal, and environmental factors (Dai et al. 2025). Among these, exogenous steroid hormone intervention is an effective method for achieving sex reversal in fish. The most critical step in the production of all-male channel catfish is the generation of fertile pseudofemales. 17β-Estradiol, a key estrogen, has been widely used to induce pseudofemales in many economically important fish species, including yellow catfish, Tachysurus fulvidraco (Richardson, 1846) (see Wang et al. 2020), northern snakehead, Channa argus (Cantor, 1842) (see Ou et al. 2026), largemouth bass, Micropterus salmoides (Lacepède, 1802) (see Zhu et al. 2025), Chinese hooksnout carp, Opsariichthys bidens Günther, 1873 (see Liu et al. 2024), and channel catfish, Ictalurus punctatus (see Xu et al. 2022).

A stable breeding system for all-male channel catfish was previously established by the present research group (Xu et al. 2022). The complete breeding process requires three generations: first, estrogen treatment of channel catfish larvae to screen for XY pseudofemales; second, crossing these XY pseudofemales with XY males to obtain YY supermales; and third, crossing YY supermales with XX females to produce all-male XY progeny on a large scale. Nevertheless, this system still has the potential for optimization. Specifically, the offspring from crossing XY pseudofemales with XY males contain a proportion of YY supermales, along with XY males and XX females, necessitating subsequent sex identification to obtain YY supermale broodstock for breeding purposes. To further improve the efficiency of all-male breeding, YY supermales can be sex-reversed into YY pseudofemales, and the cross “YY supermale × YY pseudofemale” can directly yield 100% YY supermale offspring, thereby enabling stable and efficient acquisition of YY supermale broodstock for all-male production.

However, studies on YY pseudofemales in channel catfish remain limited, and their morphological traits have not yet been systematically characterized. For use as breeding broodstock, whether YY pseudofemales differ from females in growth performance and body shape directly affects their production value. Morphological measurement is the primary technical approach for studying fish morphological traits and provides a basis for germplasm resource evaluation (Li et al. 2017). Among these measurements, the relationship between standard length and body weight is a core reference indicator in fishery management (Froese 2006). In the selective breeding of aquatic animals, including fish (Kause et al. 2022), shellfish (Ji et al. 2011; Hedgecock 2023), shrimp, and crabs (Pang et al. 2022), correlation analysis, path analysis, and stepwise regression analysis are commonly used to determine the influence of morphological traits on body weight. Using morphological traits to effectively predict body weight can greatly improve breeding efficiency. These analytical methods have been applied to evaluate body weight and morphological traits in species such as dusky grouper, Epinephelus marginatus (Lowe, 1834) (see Tarôco et al. 2024), big-head schizothoracin, Aspiorhynchus laticeps (Day, 1877) (see Lu et al. 2025), largemouth bass, Micropterus salmoides (Day, 1877) (see He et al. 2025), farmed tilapia, Oreochromis niloticus (Linnaeus, 1758) (see He et al. 2017), and silver pomfret, Pampus argenteus (Euphrasen, 1788) (see Zhang et al. 2024). Employing statistical methods to explore relationships among morphological traits in farmed fish has become an indispensable tool in artificial breeding programs (Qi et al. 2025; Wang et al. 2025; Cheng et al. 2026).

Given this background, the presently reported study built on previous work using YY and XY pseudofemales, XX females, XY males, and YY supermales as experimental stocks. The five stocks were systematically compared for differences in morphological traits, and the effect of each trait on body weight was assessed. This study provides a biological basis for the feasibility of using YY pseudofemales in production practices and a theoretical foundation for their application in future efficient all-male breeding systems.

Experimental fish. The channel catfish broodstock used in the experiment were obtained from the National Genetic Breeding Center for Channel Catfish at the Yangzhong Base of the Freshwater Fisheries Research Institute of Jiangsu Province. Sexually mature XY males and 17β-estradiol-induced XY pseudofemales were selected from the G4 generation (after four generations of continuous selective breeding). After one week of temporary rearing, they were mated to produce families. After hatching in dedicated incubation tanks, 1000 fry were randomly sampled from each family and divided into two groups. One group was fed a microdiet supplemented with 17β-estradiol (60 µg · g^–1 of feed), while the other group received a regular microdiet (Shengsuo Feed Technology Co., Ltd., Yantai, China). The stock solution of 17β-estradiol was mixed with 95% ethanol, incorporated into a microparticulate diet, thoroughly stirred, dried in an oven at 60°C, and stored in sealed bags until use. After the yolk sac was completely absorbed, larvae were fed Artemia nauplii for 3 days. Subsequently, larvae in the 17β-estradiol-treated group were continuously fed to satiation with the microdiet containing 17β-estradiol at a dose of 60 µg · g^–1 of diet until 30 days post-hatching. At 30 days post-hatching, the experimental fish were transferred to outdoor concrete tanks (5 m × 5 m × 1 m) for an additional 6 months and fed a commercial diet (Nanjing Aohua Biotechnology Co., Ltd., Nanjing, China). During the rearing period, the dissolved oxygen level in the culture water was maintained above 5 mg/L, the ammonia nitrogen concentration was kept below 0.02 mg · L^–1, and the stocking density was 20 fish per cubic meter. The fish were fed twice daily at 08:00 and 17:00, and the daily feeding rate was 3% of body weight. The preparation of the 17β-estradiol-supplemented diet and the management of the experimental fish followed the methods described in a previous study by the present research group (Xu et al. 2022). Specifically, the 60 µg · g^–1 dosage was selected based on a dose-response gradient (0, 6, 30, and 60 µg · g^–1) established in that study. The results of that study demonstrated that only the 60 µg · g^–1 treatment achieved a 100% sex reversal rate. The regular diet group comprised XX females, XY males, and YY supermales, whereas the 17β-estradiol-treated group comprised XX females, XY pseudofemales, and YY pseudofemales.

Sample collection. After 6 months of rearing, 1000 individuals were randomly selected from each of the two groups. Each fish was implanted with a passive integrated transponder (PIT) tag (1.4 mm × 8 mm) containing a 12-digit identification code (Haixing Instrument Co., Ltd., Qingdao, China) into the abdominal cavity through the pelvic fin, and a caudal fin tissue sample (approximately 3 mm wide from the distal edge of the caudal fin) was clipped. Fin tissue samples were stored in 95% ethanol until DNA extraction.

Genetic sex identification. Genomic DNA was extracted from the fin clips using a DNA Isolation Mini Kit (Vazyme, China). DNA concentration was quantified using a Qubit fluorometer (Thermo Fisher, USA), and DNA integrity was assessed by 1% agarose gel electrophoresis. For DNA samples that met the quality standards, genetic sex was identified using sex-linked molecular markers (Zhang et al. 2019). The genetic sex of all individuals was determined using the sex-linked simple sequence repeat (SSR) marker reported by Zhang et al. (2019). The specific primers were designed based on the flanking sequences of this marker. For this purpose, the specific forward primer, 5'-TGAATGTGAGACTAACAGGAG-3', was labeled with 6-FAM fluorescent dye at the 5' end, while the reverse primer was 5'-ACATCGCTTTGAGAAGCTGCT-3'. PCR amplification was performed, and the products were subsequently genotyped on an ABI PRISM 3730XL DNA Analyzer (Applied Biosystems, USA), with GS500 as the internal size standard. Allele sizes were determined using GeneMarker software. Based on the unique allele peaks, specifically the presence or absence of the 6-bp deletion allele, the sex chromosome composition (XX, XY, or YY) was clearly distinguished. After genetic sex identification and PIT tag assignment, 50 individuals from each of the following groups—XX females, XY males, YY supermales, XY pseudofemales, and YY pseudofemales—were randomly selected for statistical analysis. No phenotypic criteria, such as body weight or morphological traits, were considered during this random selection process, thereby completely avoiding potential selection bias.

Trait measurement. Based on the genetic sex identification results and PIT tag codes, XY pseudofemales and YY pseudofemales were separated from the estrogen-treated group, and XX females, XY males, and YY supermales were separated from the regular diet group. Morphological traits and body weight were measured and recorded for the five types of experimental fish. The measured traits (Fig. 1) included total length (X₁), standard length (X₂), prepectoral fin length (X₃), predorsal fin length (X₄), prepelvic fin length (X₅), preanal fin length (X₆), head depth (X₇), maximum body depth (X₈), body depth at the anus (X₉), caudal peduncle depth (X₁₀), caudal peduncle length (X₁₁), and body weight (Y). All selected pseudofemales were confirmed to be phenotypically female through anatomical examination. To ensure reliability, each measurement was performed three times, and the average was taken. In addition, all measurements for each morphological trait were performed by the same person to maintain consistency.

Analytical methods. After preliminary statistical analysis of morphological traits and body weight, correlation analysis was performed for the five types of channel catfish (XX females, XY males, YY supermales, XY pseudofemales, and YY pseudofemales). Subsequent data analysis was conducted using SPSS 26.0 software (IBM Corp. 2019). The Kolmogorov–Smirnov test was used to assess the normality of the body weight data. After confirming normality, one-way analysis of variance (ANOVA) was employed to test for overall differences in morphological traits among the five genotypes/sexes. When the ANOVA results showed significant differences (P < 0.05), post-hoc pairwise comparisons were conducted using the Bonferroni correction to control the family-wise error rate. Statistical significance was set at α = 0.05. Pearson correlation analysis was conducted. Subsequently, multiple regression analysis was performed to obtain a regression equation linking morphological traits to body weight through significant partial regression coefficients. The goodness-of-fit of the model was also evaluated, followed by path analysis to investigate how the traits included in the regression equation affected body weight and to quantify their coefficients of determination. The key formulas used in this study are as follows:

The formula for calculating the correlation coefficient is as follows:

rxy=∑i=1n(Xi−X¯)(Yi−Y¯)∑i=1n(Xi−X¯)2∑i=1n(Yi−Y¯)2 [1]

Path analysis represents the partial regression coefficients of the standardized variables. The formula for calculating the direct path coefficient is as follows:

Pi=bY,Xi×σXi/σj [2]

In the formula, P_i represents the direct path coefficient. b_Y,Xi represents the regression coefficient of morphological trait X_i. σ_Xi represents the standard deviation of morphological trait X_i. σ_j represents the standard deviation of body weight.

The formula for calculating the indirect path coefficient is as follows:

PXiXj=rijPj(i≠j) [3]

In the formula, P_XiXj represents the indirect path coefficient of morphological trait X_i through morphological trait X_j for body weight. r_ij represents the correlation coefficient between morphological trait X_i and X_j. P_j represents the direct path coefficient of morphological trait X_i for body weight.

The formula for calculating the coefficient of determination is as follows:

The coefficient of determination for a single morphological trait:

di=Pi2 [4]

The coefficient of determination for the two morphological traits:

dij=2rij×Pi×Pj [5]

Statistical methods, including regression modeling and path analysis, were adopted to explore morphological correlations in accordance with standard quantitative genetic approaches applied in aquatic species (Li et al. 2021; Tan et al. 2025; Li et al. 2026).

Genetic sex determination result. Based on the capillary electrophoresis results of the sex-linked SSR marker, which features a specific 6-bp deletion allele (allele G3) on the Y chromosome, the five distinct genotype/sex categories were identified as follows:

XX females (XX♀): The sex chromosome composition was homozygous XX. The electropherogram did not contain the specific 6-bp deletion allele and displayed only the non-deletion or 3-bp deletion alleles.

XY males (XY♂): The sex chromosome composition was heterozygous XY. The electropherogram showed a heterozygous peak profile containing one 6-bp deletion allele (Y-linked) and one non-deletion allele (X-linked).

YY supermales (YY♂): The sex chromosome composition was homozygous YY. The electropherogram showed a homozygous peak profile containing two 6-bp deletion alleles.

XY pseudofemales (XY♀): The genetic sex chromosome composition was XY. Therefore, their electropherogram profile was identical to that of XY males (heterozygous, containing one 6-bp deletion allele); however, anatomical verification confirmed their phenotypic sex to be female.

YY pseudofemales (YY♀): The genetic sex chromosome composition was YY. Consequently, their electropherogram profile was identical to that of YY supermales (homozygous, containing two 6-bp deletion alleles); however, anatomical verification confirmed their phenotypic sex to be female.

For both XY and YY pseudofemales, anatomical examination subsequently confirmed the presence of normal ovaries, thereby validating that their phenotypic sex was female despite their male genetic constitution.

Morphological traits and body weight in different genotypes and sexes. Morphological traits and body weight data for different sexes and genotypes are presented in Fig. 2 and Suppl. material 1. Among the 11 morphological traits analyzed, all traits in male channel catfish were significantly greater than those in female channel catfish. The mean body weights of XX females, XY males, YY supermales, XY pseudofemales, and YY pseudofemales were 184.96 ± 79.93 g, 253.14 ± 122.35 g, 238.04 ± 109.00 g, 200.95 ± 90.42 g, and 171.18 ± 71.64 g, respectively. Total length, predorsal fin length, head depth, maximum body depth, and body depth at the anus exhibited a trend of XY males > YY supermales > XY pseudofemales > XX females > YY pseudofemales, consistent with the rank order of body weight, indicating that morphological traits overall covaried with body weight.

Analysis of morphological trait differences between YY pseudofemales and other sexes and genotypes of channel catfish. The Kolmogorov–Smirnov test was employed to assess the normality of the measured morphological trait data. The null hypothesis (H₀) was that the data for each morphological trait followed a normal distribution, and the alternative hypothesis (H₁) was that the data did not follow a normal distribution. As shown in Table 1, the Kolmogorov–Smirnov test results indicated that the P-values for all morphological traits in the five different sexes and genotypes of channel catfish were > 0.05; therefore, H₀ was not rejected, indicating no significant departure from a normal distribution. Consequently, the 11 morphological trait datasets and body weight data for the five different sexes and genotypes of channel catfish met the normality assumption, making them suitable for subsequent correlation analysis.

Homogeneity of variances was tested using Levene’s test. All traits met the assumption of homogeneity of variance. Therefore, to investigate morphological differences between YY pseudofemales and individuals of other genotypes and sexes (XX females, XY males, YY supermales, and XY pseudofemales), one-way ANOVA was performed on the 11 morphological indicators. The ANOVA results showed that, except for predorsal fin length (X₄, P = 0.616) and caudal peduncle length (X₁₁, P = 0.094), the other nine morphological traits differed significantly among groups (P < 0.05), with caudal peduncle depth (X₁₀) exhibiting significant differences (P < 0.001).

Further post hoc multiple comparisons (Bonferroni) were conducted, as shown in Fig. 2. YY pseudofemales differed significantly from XY males in several morphological indicators: body weight (P = 0.005), total length (P = 0.023), standard length (P = 0.028), prepelvic fin length (P = 0.008), head depth (P < 0.001), and caudal peduncle length (P = 0.001) were all significantly lower in YY pseudofemales. Compared with YY supermales, YY pseudofemales also exhibited significantly lower body weight (P = 0.014), total length (P = 0.046), prepelvic fin length (P = 0.046), and head depth (P = 0.003). YY pseudofemales were significantly smaller than males in several core measures, including body weight, standard length, and head depth, indicating that YY pseudofemales differed distinctly in body shape from males. In contrast, compared with XX females and XY pseudofemales, none of the 11 morphological traits showed statistically significant differences (P > 0.05). To complement the ANOVA results and quantify the magnitude of the observed differences, pairwise effect sizes (Cohen’s d) were calculated between YY pseudofemales and the other four genotypes and sexes (Table 2). Notably, for all pairwise comparisons that reached statistical significance (P < 0.05) in Fig. 2, the corresponding Cohen’s d values consistently fell within the medium-to-large effect range (|d| ≥ 0.5). Additionally, cluster analysis of the morphological traits and body weights of channel catfish of different sexes and genotypes was performed. The results showed that females and males fell into distinct branches (Fig. 3). XY pseudofemales first clustered with XX females, and together, they then clustered with YY pseudofemales. These results indicate that the morphological phenotype of YY pseudofemales is not intermediate between males and females but is highly similar to that of XX females and XY pseudofemales.

Correlation between morphological traits and body weight in different genotypes and sexes. Pearson correlation analysis performed separately for the five different sexes and genotypes of channel catfish (Fig. 4; Suppl. material 2) showed that all groups exhibited a highly consistent and significant correlation pattern. The pairwise correlation coefficients among the 12 morphological indicators were all above 0.68 and were significant at the 0.01 level (P < 0.01). Body weight, as a core trait comprehensively reflecting the overall size of the fish, showed very strong positive correlations with the other 11 indicators (most r > 0.85), with the strongest correlations for total length, standard length, prepelvic fin length, maximum body depth, and body depth at the anus (r > 0.93). In all five sexes and genotypes of channel catfish, the correlation between standard length and total length was the highest (r > 0.98), indicating high consistency between these two measurements. Maximum body depth and body depth at the anus showed an almost perfectly covarying relationship (r values between 0.987 and 0.995), indicating a very stable body profile of the trunk.

Influence of morphological traits on body weight in different genotypes and sexes. Multiple regression models for the five different sexes and genotypes of channel catfish were highly significant (P < 0.001), with adjusted R² ranging from 0.887 to 0.958, with XY pseudofemales and YY supermales having the best-fitting models (adjusted R² > 0.95).

As shown in Fig. 5, Suppl. material 3, 4, the direct path coefficient reflects the net effect of a single trait on body weight after controlling for other traits. The key positive and negative effects in the different sexes and genotypes of channel catfish are summarized in Table 3.

Many traits had small or even negative direct effects but produced large positive indirect effects through other traits (sum of indirect effects > 1.0), such as total length (X₁), standard length (X₂), and predorsal fin length (X₄) in XX females; preanal fin length (X₆) in XY males; standard length (X₂) and prepelvic fin length (X₅) in YY supermales; and prepelvic fin length (X₅) in YY pseudofemales. This indicates that the traits are highly correlated and that the influence of morphological traits on body weight is often achieved indirectly through other traits.

For XX females, XY pseudofemales, and YY pseudofemales, the key positively contributing trait was body depth at the anus (X₉), indicating that the relationship between morphological traits and body weight in pseudofemale channel catfish more closely resembles that of females.

For XY males and YY supermales, the key positively contributing traits were standard length (X₂) and total length (X₁), respectively, suggesting that male channel catfish exhibit a stronger tendency toward increased length.

Therefore, to achieve body weight gain, selection strategies should be tailored to each population: for female and pseudofemale channel catfish, individuals with greater body depth at the anus should be favored; for XY males, individuals with greater standard length should be selected; and for YY supermales, individuals with greater total length and caudal peduncle depth should be prioritized.

Establishment of multiple linear regression equation for body weight of YY pseudofemales using other morphological data. Cook’s distance (Cook’s D) of the residuals was used to determine whether the dataset contained any influential points. The observed maximum Cook’s distance was 0.537, which is less than the threshold of 1. This indicates that the dataset does not contain any highly influential points, allowing the analysis to proceed.

Based on the direct path coefficients and comprehensive determination coefficients, the key morphological traits X₁, X₅, X₆, and X₉ that had a substantial influence on body weight in the YY pseudofemale population were retained. Using these four traits as independent variables and body weight as the dependent variable, multiple linear regression was performed, as shown in Table 4.

Based on the multiple linear regression results, total length, prepelvic fin length, and body depth at the anus were significantly correlated with body weight in the YY pseudofemale population. The stepwise method was used to validate the multiple linear regression equation. As shown in Table 5, as independent variables were introduced, the coefficient of determination (R²) for the YY pseudofemale population increased from 0.879 to 0.914, while the standard error decreased from 25.435 to 21.905. Multicollinearity of the independent variables was assessed using the variance inflation factor (VIF) and tolerance. The results (Table 6) showed that the VIF values of all independent variables in the model were less than 5, indicating that multicollinearity was within an acceptable range and that the regression coefficient estimates were stable and reliable. For the traits significantly correlated with body weight in YY pseudofemales, the partial regression coefficients were calculated, yielding the final multiple linear regression equation:

Y = − 311.163 + 18.815 X 1 − 24.614 X 5 + 54.933 X 9

(R2=0.914) [6]

To further reveal the nonlinear influence patterns of key morphological traits on body weight in YY pseudofemales, multiple curve models were fitted with total length (X₁), prepelvic fin length (X₅), and body depth at the anus (X₉) as independent variables and body weight (Y) as the dependent variable. As shown in Table 7, the best-fitting model for all three traits was the power function, with coefficients of determination (R²) of 0.905, 0.849, and 0.927, respectively, all higher than those of the linear or other models. Based on the fitted parameters in Table 7, the specific regression equations for the three key traits are: Y = 0.01X₁^2.97; Y = 0.01X₅^2.501; and Y = 0.01X₉^2.506.

Morphological traits of YY pseudofemales resemble those of females. Exogenous estrogen-induced sex reversal of genetically male individuals into fertile pseudofemales has been successfully achieved in several economically important fish species. In yellow catfish, northern snakehead, and largemouth bass, treatment with 17β-estradiol effectively induced XY individuals to develop functional ovaries. However, the response to estrogen treatment varies among species. For example, Vizziano-Cantonnet et al. (2025) found that in Siberian sturgeon, Acipenser baerii Brandt, 1869, although estradiol treatment successfully induced complete female transdifferentiation of the gonads in genetic males, the treatment window was extremely narrow. Successful induction required initiating treatment at a specific stage of molecular sex differentiation (3.5 months of age), whereas delayed treatment was not effective. In spotted scat, Scatophagus argus (Linnaeus, 1766), Jiang et al. (2024) observed that estradiol treatment alone resulted in ovarian malformations and that morphologically identifiable pseudofemales could only be produced when estradiol was combined with trilostane. This finding corroborates the molecular mechanism study of pseudofemales in channel catfish by Xu et al. (2024). Through comparative transcriptome analysis of estrogen-induced pseudofemales and females, Xu et al. found that not only were the ovaries of pseudofemales developing normally, but gene pathways related to female reproduction were also active, with progesterone receptor and androgen receptor genes being downregulated in the ovaries of pseudofemales. This suggests that estrogen induction not only reshapes the reproductive organs of pseudofemales but also results in a fully female morphological phenotype through the regulation of hormone receptor gene expression. This finding is of critical importance from the perspective of breeding applications. In the “YY supermale × YY pseudofemale” breeding system, whether the morphological characteristics of YY pseudofemales as the female parent are compatible with those of females directly affects mating success and reproductive efficiency. The presently reported study confirmed that there were no statistically significant differences in key morphological dimensions between YY pseudofemales and other females, thereby phenotypically ruling out potential morphological barriers to their use as “female broodstock.” This fulfills a prerequisite for subsequent fecundity tests and production-scale mating trials, reducing the risk of breeding failure owing to morphological incompatibility.

Relationship between morphological traits and body weight in different sexes and genotypes of channel catfish. Path analysis revealed that the main morphological traits influencing body weight varied among different sexes and genotypes of channel catfish. Pseudofemale channel catfish (XY and YY pseudofemales) exhibited a pattern similar to that of females, in which weight gain primarily depended on increased body depth at the anus, displaying a “body depth-dominant” pattern. In contrast, male channel catfish (XY males and YY supermales) exhibited a “standard length-dominant” or “total length-dominant” growth pattern. This divergence in growth strategy may arise from the different energy allocation strategies between the sexes. Female channel catfish need to reserve substantial energy for reproduction, and abdominal cavity capacity, which is closely related to body depth, directly affects fecundity, thus favoring an increase in body depth. Males, on the other hand, tend to rapidly increase standard length to gain advantages in competing for territories and mating opportunities (Wang et al. 2020). This sex-based divergence in growth strategy is not unique to channel catfish and has also been observed in other fish species. Hao et al. (2025) studied Diptychus maculatus Steindachner, 1866 in the Muzat River and found that the influence of morphological traits on body weight differed significantly between the sexes: traits such as body depth and standard length had the greatest direct effect on female body weight, while traits such as the linear distance from the dorsal fin origin to the pectoral fin origin and total length had a greater effect on male body weight. Similarly, another study specifically examining sex differences in channel catfish growth clearly indicated that standard length is the key trait affecting male body weight, whereas female body weight is influenced by both standard length and body depth (Xie et al. 2025). The presently reported study extends this pattern to artificially induced YY pseudofemale channel catfish, confirming that their growth patterns are highly consistent with those of the female population. In production practice, if the goal is to rapidly increase the body weight of commercial fish, the selection of pseudofemale broodstock should focus on body depth-related traits. For male lines destined for direct grow-out, individuals with high potential for standard length growth should be prioritized.

Predictive value of morphological traits for YY pseudofemales based on a regression equation. The presently reported study found that all morphological traits were strongly correlated with body weight, and a predictive model for body weight in YY pseudofemales was successfully constructed. This model is consistent with findings in Asian seabass, Lates calcarifer (Bloch, 1790) (see Zhao et al. 2017). It provides important technical support for production practice: when selecting high-quality YY pseudofemale broodstock, attention should be paid to key dimensions, such as total length, pre-pelvic fin length, and body depth at the anus. This selection strategy has been validated in other aquatic animals; for example, in the selective breeding of longsnout catfish, Rhinobagrus dumerili Bleeker, 1864, total length is often used as an important indicator for evaluating growth performance (Yu et al. 2025). Through directional selection, the growth performance of the YY pseudofemale line is expected to be gradually optimized, thereby providing better “female broodstock” for the large-scale production of all-male YY supermales and ultimately improving the farming efficiency of commercial fish. In principle, this approach enables early-stage selection using the established regression equation to predict body weight potential at 6 months of age based on morphological data, thereby screening YY pseudofemales with advantageous morphological traits and superior growth performance, shortening generation intervals, accelerating the breeding process, and increasing breeding success (Diakos et al. 2026). Additionally, its robustness across independent populations and different rearing conditions remains to be demonstrated before it can be adopted as a routine selection tool.

In summary, this study confirmed that YY pseudofemales are morphologically similar to females, establishing a foundation for their integration into all-male breeding systems. Furthermore, it elucidated the differential relationships between morphological traits and body weight across different sexes and genotypes of channel catfish and constructed a predictive model for body weight in YY pseudofemales, providing a theoretical basis for the precise selective breeding of this species.