This study included 80 goats from four breeds: Kacang, Local Gorontalo, Etawa crossbreed, and Saanen crossbreed, with 20 animals per breed. All goats had recently kidded. Colostrum was collected during the first three days postpartum to represent the early stages of lactation. The animals were kept on a traditional farm in Gorontalo Province, Indonesia. Feeding protocols adhered to the National Research Council (NRC) guidelines for small ruminants and aligned with standard practices in Indonesian goat production [18]. Colostrum samples were obtained on Days 1, 2, and 3 postpartum by manual milking immediately after parturition and during subsequent morning milkings. Each goat was sampled once to minimize diurnal variation. Between 50 and 100 mL of colostrum was aseptically collected into sterile polypropylene tubes, discarding the first streams to reduce teat canal contamination. Samples were transported in insulated containers at 2–4°C and delivered to the laboratory within one hour. Upon arrival, colostrum was filtered through sterile gauze to remove debris, aliquoted into 15 mL vials, and stored at 4°C for no more than 12 hours before analysis. Surplus aliquots were preserved at − 20°C for further assays.

Colostrum samples were prepared by gentle homogenization at 20 ± 1°C prior to proximate analysis, and all determinations were performed in triplicate. The dry matter (DM) content was measured gravimetrically according to ISO 6731/IDF 21:2010 by oven-drying approximately 5.0 g of sample at 102 ± 2°C until constant weight [19]. Crude protein was determined by the Kjeldahl method following AOAC 991.20 [20], in which total nitrogen was quantified and converted to protein using the standard factor N × 6.38. Colostrum fat was analyzed using the Gerber butyrometric method (ISO 2446/IDF 105:2008) [21], and high-protein colostrum readings were verified on a subset of samples by the gravimetric Röse–Gottlieb method (ISO 1211/IDF 1:2010) [22]. Lactose was measured enzymatically following ISO 22662/IDF 79 − 1:2004 [23], based on β-galactosidase hydrolysis and spectrophotometric detection at 340 nm, and results were cross-validated in selected samples by HPLC with refractive index detection. All analyses were conducted under controlled laboratory conditions with calibration of balances, volumetric glassware, and thermometers according to IDF/ISO quality assurance standards.

The analysis was conducted according to previously described procedures, with minor modifications [24]. Thawed colostrum samples were filtered using a 0.2 µm cellulose nitrate membrane filter. The resulting hydrolysate underwent pre-column derivatization with phenylisothiocyanate. Amino acid analysis was conducted using a reversed-phase JASCO HPLC System equipped with a JASCO Unifinepak C18 column (3.0 mm I.D x 75 mm L, 1.9 µm). The mobile phase consisted of HPLC-grade acetonitrile containing 0.1 M ammonium acetate at pH 6.5 and methanol-water mixture (46:10:44 v/v). The flow rate was set at 1.0 mL/min, and and UV detection was carried out at 254 nm.

Fatty acids were analyzed using Gas Chromatography (GC). Before the hydrolysis and esterification process was carried out, fat extraction of colostrum samples was done using goldfish extraction. After esterification into fatty acid methyl ester (FAME), the sample was analyzed using GC. A standard solution of 1 mg was added to 100 mL of colostrum sample. Then, 0.5 N NaOH-Methanol solution of 1.5 mL was put into the colostrum sample and then vortexed. The N₂ gas was then exhaled to remove O₂ gas, which can oxidize FAME so that the measurement results will be negative. The mix ture solution was put into a water bath with a temperature of 80–100°C for 5 min so that the saponification reaction occurs. The mixture was added with 2 mL of BF₃-Methanol for the esterification reaction, then blown again with N₂ gas and put back into the 80–100°C water bath for 5 min. The FAME mixture obtained was separated by solvent extraction (liquid-liquid extraction) by adding hexane as much as 1.5 mL and vortexed. If there was no separation, then 3 mL of saturated NaCl was added and vortexed. Then, the hexane phase was taken carefully and added anhydrous Na₂SO₄ to bind water. The liquid formed was then taken using a pipette and injected into the GC (7890A GC System, Agilent Technologies, USA) instrument.

Colostrum samples were centrifuged for 10 min at 12,000 × g and − 4°C. The fat layer was removed. The colostrum serum collected beneath the fat layer was used for subsequent protein characterization. Protein concentrations in the colostrum samples were determined using the Lowry method [25]. Protein samples were solubilized in phosphate buffer at defined concentrations at 20°C. The solubilized proteins were then separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) using a Bio-Rad Mini-Protean 3 cell system, according to Laemmli's protocol [26]. Following electrophoresis, the gel was stained with Coomassie blue dye for two hours and then destained in a solution containing 30% methanol and 10% acetic acid. The molecular weight marker 10–200 kDa (#SM0661) was from Fermentas (USA), and other reagents were of analytical grade. A standard curve was generated using bovine serum albumin (BSA) in the range of 2 to 10 micrograms to determine the concentration of target protein bands. Band intensities were normalized against a pooled colostrum reference sample present in each gel. Each biological replicate (n = 3 per breed) underwent SDS-PAGE analysis in triplicate. Each run included a molecular weight marker and an internal technical control.

The Image Lab (Bio-Rad) software was used to examine the documentation's results using the densitometry technique. The software could predict molecular weight and relative density of each band which was relatively compared to bands from protein ladders/markers, so each molecular weight was determined. By calculating the standard protein quantity based on their density, the bands from BSA at varying concentrations were utilized to forecast the amount of each protein component [27].

The zeta potential and average particle size of goat colostrum were measured after diluting a 200 µL sample with 10 mL of 0.1 mol/L Na₂HPO₄-NaH₂PO₄ (pH 7.0) buffer solution. To get rid of any dust and contaminants, the samples were then centrifuged at 1,078 × g for five minutes at room temperature. The protein particle size was then measured using a Horiba Nano Partica SZ-100 (Kyoto, Japan) nanoparticle size analyzer, following the method of Li et al. [27]. Measurements were performed at 25°C with a scattering angle of 90°, particle refractive index of 1.450, particle absorptivity of 0.8872, and water as the dispersant (refractive index 1.330). Zeta potential was determined using a Horiba Nano Partica SZ-100 (Kyoto, Japan) with laser Doppler microelectrophoresis, following the method of O'Brien et al. [28].

Rheological characterization was performed using a Modular Compact Rheometer MCR302 (Anton Paar MCR 302, Graz, Austria) equipped with a cone–plate geometry (50 mm diameter, 1° angle, 0.5 mm gap) under temperature control with a Peltier system maintained at 25 ± 0.1°C. Data acquisition and processing were carried out using Rheo Compass Software v1.25 (Anton Paar, Austria). Amplitude sweep tests were conducted to determine the Linear Viscoelastic Region (LVE) and yield points, by applying shear strain in a logarithmic ramp from 0.001% to 100% at a constant frequency of 1 Hz (6.28 rad/s). The parameters recorded included storage modulus (G′), loss modulus (G″), and the loss factor (tan δ = G″/G′). Apparent viscosity (η) was calculated as the ratio of shear stress (τ) to shear strain (γ), while the LVE limit was identified as the strain value at which G′ deviated more than 5% from its plateau value. The yield point was defined at the intersection where tan δ ≈ 1, indicating the transition from solid-like to liquid-like behavior.

Following Li and Shah's[29] suggested methodology, the colostrum samples' structures were examined using the Olympus FV1200 confocal scanning laser microscope. The confocal scanning laser microscopy was outfitted with a silicon oil objective lens and an inverted microscope (magnification 150×). A tagged image file format is used to acquire digital image files. Ten microliters of Nile red (1 mg/mL) were used to stain one milliliter of goat colostrum samples. in ethanol solution) and 10 µL of fluorescein isothiocyanate (FITC, 1 mg/mL in ethanol solution) for 30 min. Then, 20 µL of the stained material was pipetted onto a glass slide, covered with a coverslip, and put right into the confocal scanning laser microscope. In order to conduct the observations in a dark room, the emission wavelengths for FITC and Nile Red were set at 495 to 559 nm and 534 to 488 nm, respectively, and the excitation wavelengths were set at 500 to 600 nm.

Results are shown as mean ± SD. Data were analyzed by two-way factorial ANOVA with breed, postpartum day, and their interaction as fixed factors. Significant effects (p < 0.05) were followed by Tukey’s HSD test. Superscripts in tables and figures indicate significance: lowercase letters (a, b, c) for breeds, uppercase (A, B, C) for days. Analysis used SPSS version 25 (IBM Corp., Armonk, NY, USA). Principal component analysis (PCA) was used to examine the proximate composition, amino acid and fatty acid profiles, physical properties, and rheological parameters of different breeds and postpartum days, thereby visualizing multivariate differences among them. PCA preprocessing included centering and Varimax rotation. All tests used p < 0.05 as the significance threshold.

The physicochemical composition of colostrum is significantly influenced by the breed and postpartum day of the goat. Distinct differences were observed among the four goat breeds in Gorontalo-Kacang, local, Etawa crossbreed, and Saanen crossbreed (Figs. 1, 2 and 3; Tables 1 and 2).

The dry matter (DM) content exhibited a highly significant influence of both breed and postpartum day (p < 0.05). On Day 1, Saanen crossbreed goats demonstrated the highest DM concentration (24.95 ± 0.32%), followed by Etawa crossbreed (22.45 ± 0.22%), Local Gorontalo (21.89 ± 0.24%), and Kacang (21.45 ± 0.12%). Statistical groupings confirmed clear breed effects, with Saanen consistently classified in the highest subset (ᵃ), Etawa occupying an intermediate position (ᵇ), and Kacang remaining the lowest (ᶜ). Temporal changes were equally pronounced, as DM declined significantly across all breeds from Day 1 to Day 3 (p < 0.05). For instance, Saanen decreased from 24.95% to 22.87%, whereas Kacang declined more sharply from 21.45% to 17.64%. This progressive reduction reflects the well-documented dilution effect of colostrum, whereby protein- and fat-rich secretions are gradually replaced by transitional milk of lower nutrient density but higher lactose concentration [30]. The magnitude of decline was most pronounced in indigenous breeds, indicating a faster compositional shift compared with specialized dairy crossbreeds. These findings highlight the dual role of genetic background and temporal dynamics in determining colostrum quality, and they reinforce the nutritional superiority of crossbreeds such as Saanen in maintaining higher DM during the critical early postpartum period [31].

Protein composition in goat colostrum was strongly affected by both breed and postpartum day (p < 0.05) in Fig. 1. On Day 1, Saanen crossbreeds exhibited the highest crude protein concentration (16.52 ± 0.34%), followed by Etawa, Local Gorontalo, and Kacang, a hierarchy that persisted despite progressive declines through Day 3, reflecting colostrum dilution into transitional milk [32][33]. This trend was mirrored in amino acid profiles (Fig. 2), where Saanen and Etawa contained greater essential amino acids—particularly leucine, lysine, and valine—along with high non-essential amino acids such as glutamic acid (> 1,100 mg/100 mL) and proline (> 800 mg/100 mL), supporting neonatal muscle growth, immunity, and intestinal development. In contrast, indigenous breeds, though lower in crude protein, contributed proportionally higher sulfur amino acids (methionine, cysteine), conferring antioxidant and metabolic benefits. Amino acids declined by 15–25% across breeds by Day 3, consistent with Kráčmar et al. [34], who reported 10–72% reductions within 2–12 h postpartum. Protein fraction analysis corroborated these patterns (Table 1). Saanen colostrum contained the highest β-casein, α-casein, κ-casein, β-lactoglobulin, and α-lactalbumin, all of which declined by Day 3 yet maintained breed-specific rankings. These results align with Levieux et al.[35] and Sánchez-Macías et al. [36] who observed rapid postpartum decreases before stabilization, while Soloshenko et al. [37] emphasized the functional roles of β-casein A2, β-lactoglobulin, and lactoferrin. Collectively, dairy crossbreeds offer superior protein quality for neonatal development, whereas indigenous breeds provide complementary antioxidant signatures, underscoring the dual nutritional and functional value of breed diversity in colostrum.

Fat content showed significant breed-specific variation and temporal changes (p < 0.05) in Fig. 1. On Day 1, Saanen (6.75 ± 1.54%) and Etawa crossbreeds (6.32 ± 1.42%) exhibited higher fat levels than indigenous breeds, Kacang (5.38 ± 0.36%) and Local Gorontalo (5.51 ± 0.41%), with statistical grouping consistently placing dairy crossbreeds at the upper tier. From Day 1 to Day 3, all breeds showed declines, most pronounced in Kacang (− 16.2%), while Saanen maintained higher values with a modest decrease (− 11%), indicating their capacity to sustain greater energy density in early colostrum. These findings are comparable with results in Toggenburg goats, where fat remained relatively stable during the first 24 h, while Anglo-Nubian goats exhibited an increase from 1 h to 24 h postpartum [38]. In Czech White Shorthair goats, fat declined within the first 36 h [39], and Zaharia et al.[40] reported oscillations in Romanian local breeds, peaking at 6 h before decreasing steadily by Day 7. Similarly, Argüello et al.[32] described initial increases in Majorera goats during the first day, followed by gradual declines. Such dynamics are largely attributed to colostrum dilution as lactose synthesis rises and transitional milk production begins [36]. Collectively, crossbreeds provide consistently higher fat reserves for neonatal energy, whereas indigenous breeds exhibit more fluctuations, highlighting the combined influence of genetic background and physiological adaptation on colostrum fat composition.

The fatty acid composition of goat colostrum showed significant effects of both breed and postpartum day (p < 0.05; Fig. 3). Saturated fatty acids (SFA) dominated across all breeds (65–69%), but indigenous goats (Kacang, Local Gorontalo) consistently exhibited higher SFA, whereas Saanen and Etawa crossbreeds contained more unsaturated fatty acids (UFA). Polyunsaturated fatty acids (PUFAs), notably linoleic (C18:2) and α-linolenic acid (C18:3), increased from Day 1 to Day 3, with larger gains in crossbreeds. These PUFAs are essential for membrane stability, immune regulation, and neural development [41]. Conjugated linoleic acid (CLA) was enriched in Saanen and Etawa colostrum, remaining 25–30% higher than indigenous breeds. CLA is widely recognized for its anti-carcinogenic, anti-obesity, and cardioprotective properties [37][42]. Crossbreeds therefore provide colostrum richer in bioactive lipids that enhance neonatal growth and human health potential, while indigenous breeds retain higher SFA, offering oxidative stability valued in traditional dairy contexts. Collectively, these breed-specific lipid patterns highlight the dual role of crossbred and indigenous goats in sustainable dairy development.

Lactose content exhibited an inverse trend to protein and fat, showing a progressive rise from Day 1 to Day 3 (p < 0.05) in Fig. 1. On Day 1, indigenous breeds such as Kacang (3.55 ± 0.32%) and Local Gorontalo (3.72 ± 0.18%) recorded significantly higher lactose compared with crossbreeds, particularly Saanen (2.32 ± 0.62%). By Day 3, lactose levels increased across all breeds, with Saanen rising to 3.21 ± 0.25%, narrowing the gap with indigenous goats. These findings align with previous reports. Moreno-Indias et al. [43] observed marked lactose elevation within the first 10 h postpartum, while Rachman et al. [44] confirmed increases across Indonesian breeds. Mondeshka and Stoycheva [38] similarly reported rises from 2.19% to 3.24% in Toggenburg goats within 24 h, highlighting a universal trend. Physiologically, lactose drives osmotic regulation, promoting milk volume expansion and colostrum dilution. Indigenous goats provide higher lactose for hydration, while crossbreeds, though initially lower, exhibit rapid increases that support high-volume milk production.

Specific gravity demonstrated a clear hierarchy across breeds throughout the postpartum period (Table 2), with mean values highest in Saanen (1.045) and Etawa (1.045), followed by Local Gorontalo (1.037) and Kacang (1.034). Although daily variation was modest, all breeds exhibited a consistent decline over time. Linear regression slopes ranged from − 0.0010 units per day in Kacang to − 0.0035 units per day in Etawa. From Day 1 to Day 3, percentage reductions were − 0.19% in Kacang (1.035 → 1.033), − 0.38% in Local Gorontalo (1.039 → 1.035), − 0.67% in Etawa (1.049 → 1.042), and − 0.38% in Saanen (1.047 → 1.043). Statistical analysis (ANOVA) confirmed significant breed effects (p < 0.05), with Saanen and Etawa consistently classified in the highest group, reflecting higher concentrations of total solids, particularly protein and fat, in early colostrum. These findings are consistent with broader literature. Romero et al. [45] reported a specific gravity of 1.0528 g/mL in Murciano-Granadina goat colostrum immediately after parturition, which declined to 1.0303 g/mL at 156 h postpartum. Similarly, Sánchez-Macías et al. [36] observed a density of 1.0480 g/mL in Majorera goats, decreasing to 1.0280 g/mL by 90 days postpartum. Mondeshka et al. [46] further emphasized that density is tightly linked to concentrations of protein, fat, and total solids, and thus varies significantly with breed and stage of lactation. These external reports corroborate the present results, confirming that dairy crossbreeds, such as Saanen and Etawa, maintain higher density values than indigenous breeds during early lactation. From a functional perspective, higher specific gravity signals greater nutrient density, which benefits neonatal nutrition and also enhances technological properties for dairy processing.

Zeta potential values became progressively more negative during the first three days postpartum across all breeds, reflecting enhanced electrostatic repulsion and colloidal stability (Table 2). Linear regression analysis indicated slopes ranging from − 1.95 mV per day in Etawa crossbreeds to − 2.22 mV per day in Local Gorontalo goats. Substantial relative changes were observed between Day 1 and Day 3: Kacang (− 20.78 to − 24.98 mV; +20.2%), Local Gorontalo (− 20.65 to − 25.09 mV; +21.5%), Etawa (− 19.90 to − 23.79 mV; +19.5%), and Saanen (− 19.97 to − 23.98 mV; +20.1%). Mean values across days showed that Etawa (− 21.37 mV) and Saanen (− 21.48 mV) had less negative potentials compared to Kacang (− 22.36 mV) and Local Gorontalo (− 22.52 mV), suggesting breed-specific differences in colloidal stabilization. ANOVA confirmed significant effects of both breed and day (p < 0.05). These findings are consistent with previous studies. Sun et al. [3] reported that colostrum exhibits less negative zeta potential than mature milk, reflecting compositional differences during digestion, while Temerbayeva et al. [15] identified increasing negative zeta potential as a marker of micellar stabilization. Collectively, the observed patterns demonstrate the physiological transition from a dense, protein-rich colostrum matrix on Day 1 to a more dispersed and colloidally stable system by Day 3.

Particle size analysis revealed a consistent decrease across all goat breeds during the early postpartum period, reflecting micellar restructuring and fat globule dispersion (Table 2). Quantitatively, reductions ranged from − 1.16 nm per day in Saanen to − 1.68 nm per day in both Kacang and Local Gorontalo. By Day 3, particle size decreased by − 2.64% in Kacang (127.08 → 123.72 nm), − 2.62% in Local Gorontalo (127.78 → 124.43 nm), − 2.42% in Etawa (128.43 → 125.32 nm), and − 1.81% in Saanen (128.41 → 126.09 nm). Across breeds, Saanen (127.27 nm) and Etawa (127.10 nm) maintained larger mean micelles compared with Kacang (125.00 nm) and Local Gorontalo (125.81 nm), consistent with their higher initial solids content. These findings align with previous reports showing that casein micelle size in goat colostrum is larger than in mature milk and decreases progressively as lactation advances, reaching approximately 76–82 nm in mature milk [47]. Sun et al. [48] further demonstrated that Xinong Saanen colostrum exhibited larger micelle diameters and slower digestion, supporting neonatal immune protection before shifting to smaller, more digestible micelles in mature milk. Nutritionally, finer micelle dispersion enhances digestibility and bioavailability of amino acids, while technologically, smaller micelles improve stability and functional behavior in dairy applications. Combined with zeta potential results, the data confirm a physiological transition from dense, nutrient-rich colostrum to a more stable and bioavailable milk system, underscoring both genetic and temporal regulation of micelle architecture in goat colostrum.

The rheological parameters of goat colostrum demonstrated significant effects from both breed and postpartum day (p < 0.05). These results reveal clear genotypic differences and temporal shifts during the transition from colostrum to mature milk.

The rheological behavior of goat colostrum was strongly determined by breed and postpartum stage (Table 3). On Day 1, Saanen and Etawa crossbreeds exhibited superior viscoelastic parameters, with higher linear viscoelastic (LVE) limits (0.85 ± 0.06% and 0.78 ± 0.05%) and yield points (1.20 ± 0.08% and 1.09 ± 0.09%) compared with indigenous breeds such as Local Gorontalo (0.57 ± 0.04%; 0.83 ± 0.05%) and Kacang (0.52 ± 0.03%; 0.74 ± 0.05%). Yield stress and apparent viscosity further highlighted this distinction: crossbreeds achieved values exceeding 400 Pa and > 1.7 × 10⁵ Pa·s, whereas indigenous breeds, particularly Kacang, recorded substantially lower values (311.9 Pa; 9.8 × 10⁴ Pa·s). Across all breeds, viscosity and stress declined significantly (p < 0.05) by Day 3, reflecting macromolecular dilution as protein and lipid levels decreased. These results align with broader dairy rheology studies, where casein and whey proteins are recognized as the principal contributors to structural resilience and viscoelastic strength [49]. The higher rheological stability of crossbreeds is consistent with their enriched protein and fat fractions, supporting neonatal nutrition by providing a denser, more stable matrix. In contrast, the lower viscosity observed in indigenous goats may enhance digestibility, offering a complementary functional advantage. The tan δ values confirmed this trend, with crossbreeds exhibiting more balanced elastic–viscous behavior, while indigenous breeds shifted more rapidly toward elastic dominance. Collectively, these findings demonstrate that genetic background governs rheological performance, with implications for neonatal health, dairy processing, and sustainable utilization of goat colostrum.

Figure 4 presents the confocal laser scanning microscopy (CLSM) images of colostrum microstructure across four goat breeds during the first three postpartum days. The visualizations highlight distinct breed-dependent and temporal variations in protein–fat organization, fat globule dispersion, and micellar density. On Day 1, Saanen and Etawa crossbreeds displayed compact, continuous protein matrices and tightly packed micelles, in agreement with their higher crude protein and casein levels (Table 2). Fat globules appeared relatively large and homogeneously distributed, forming an energy-dense network. In contrast, indigenous breeds (Kacang and Local Gorontalo) showed looser, less organized networks with smaller micelles and more heterogeneous fat globule distribution, reflecting their lower solids content. Temporal progression revealed progressive dilution of the protein matrix, reduced micelle density, and greater globule dispersion from Day 1 to Day 3, consistent with the biochemical shift from colostrum to transitional milk. Indigenous breeds exhibited faster matrix weakening, while Saanen retained larger aggregates, supporting its superior rheological resilience (Fig. 4). These findings are consistent with bovine studies, where Zou et al. [50] demonstrated that colostrum contains more sphingomyelin and polar lipids, conferring higher structural heterogeneity and domain formation in the MFGM compared with mature milk. Collectively, the microscopy results support rheological and physicochemical data, confirming that crossbred colostrum exhibits denser, more stable protein–fat networks, while indigenous breeds transition more rapidly to diluted structures. This genotype-driven microstructural variation underscores the dual importance of composition and organization in shaping both neonatal nutrition and processing functionality.

The PCA biplot provides a comprehensive visual representation of how different breeds of goats are distributed based on their physical colostrum composition. It illustrates the relationships among the measured variables and how they contribute to the overall variation in physical colostrum, Fig. 5A displays the plot of the PCA model of chemical colostrum goat composition, while Fig. 5B displays the plot of the PCA model of physical colostrum goat composition.

The principal component analysis (PCA) of chemical colostrum composition across breeds and lactation days revealed two main dimensions explaining most of the variability (Fig. 5A). The first principal component (PC1, 64.3% of variance) was predominantly associated with nutritional density, being strongly influenced by casein fractions (β-, α-, κ-casein), β-lactoglobulin, total protein, fat percentage, and saturated fatty acids. Breeds with positive PC1 scores, particularly the Saanen crossbreed, were positioned on the right side of the PCA quadrant, reflecting nutrient-rich colostrum, whereas breeds with negative PC1 scores, such as the Kacang, showed a simpler composition with lower protein–lipid content. The second principal component (PC2, 21.6% of variance) captured variation in amino acid composition (glutamate, aspartate, serine, leucine) and specific fatty acids (CLA, MUFA, PUFA). Positive PC2 scores indicated colostrum characterized by greater amino acid diversity and higher proportions of beneficial fatty acids, while negative PC2 scores reflected reduced representation of these compounds. Quadrant interpretation further highlighted the breed-specific profiles: Saanen crossbreeds consistently occupied Quadrant I (PC1+, PC2+), denoting colostrum of the highest nutritional density with enriched amino acid–fatty acid composition; Etawa crossbreeds were located in Quadrant IV (PC1+, PC2−), showing strong nutritional content but lower biochemical diversity; Local Gorontalo goats were positioned in Quadrant II (PC1−, PC2+), indicating moderate nutritional value coupled with enriched amino acid diversity; and Kacang goats fell within Quadrant III (PC1−, PC2−), characterized by the lowest nutritional content and overall simpler biochemical composition.

The principal component analysis (PCA) of physical quality parameters across goat colostrum samples (breed × day) revealed two major components explaining 96.4% of the total variance (Fig. 5B). PC1 (80.2%) was strongly associated with the combined influence of specific gravity, zeta potential, and particle size, and therefore represented a general axis of colloidal stability and structural density in colostrum. Observations with high PC1 values were characterized by greater particle size and altered zeta potential, reflecting more heterogeneous colloidal structures, whereas negative PC1 scores were linked to colostrum with relatively higher specific gravity and lower particle size. PC2 (16.3%) provided an orthogonal dimension, capturing residual variation related to subtle differences in zeta potential and particle distribution, and thus representing a secondary stability axis. The quadrant distribution provided clear breed-level differentiation. Saanen crossbreeds were positioned predominantly in the positive PC1–PC2 quadrant, indicating colostrum with larger particle size and relatively less negative zeta potential, consistent with enhanced aggregation tendency and distinct colloidal behavior. Etawa crossbreeds clustered within the positive PC1 but negative PC2 quadrant, reflecting high particle size but lower secondary stability, suggesting more variable colloidal dynamics. Local Gorontalo goats were located in the negative PC1 and positive PC2 quadrant, characterized by higher specific gravity and moderate zeta potential, indicative of denser and more stable colostrum structure. In contrast, the Kacang breed consistently occupied the negative PC1–PC2 quadrant, associated with smaller particle size and higher specific gravity, suggesting simpler and more stable colloidal systems compared to the exotic breeds. Overall, the PCA quadrant and biplot confirm that physical properties can distinctly separate local and crossbreed goats, with Saanen and Etawa crossbreeds showing greater variability and larger particle structures, while Local Gorontalo and Kacang breeds demonstrate denser and physically more stable colostrum. This physical characterization is crucial for understanding breed-specific differences in colostrum quality, with direct implications for nutritional delivery, neonatal immunity, and sustainable utilization of local goat genetic resources.

To improve our understanding of colostrum quality and its market potential, future research should focus on several key areas. First, investigating the impact of dietary composition on the physicochemical properties of colostrum could optimize nutrient intake and enhance product stability. Additionally, assessing the influence of various lactation stages on colostrum composition through Principal Component Analysis (PCA) would provide valuable insights into the dynamic changes in colostrum quality over time. Furthermore, exploring consumer preferences regarding goat colostrum characteristics could help align dairy product development with market demand. Future studies should also examine the combined effects of diet and lactation stage on the chemical composition of colostrum. Adapting processing methods to accommodate breed-specific variations in colostrum properties may further improve both its nutritional value and commercial viability. Moreover, establishing market segmentation strategies based on colostrum composition could optimize product quality and maximize economic benefits in the dairy industry.