A total of 121 samples were collected from 6 Indonesian native chicken breeds randomly, namely MRG (n= 24), PLG (n= 17), KJM (n= 30), STL (n= 16), NNK (n= 14), and GAG (n= 20) from different regions, as shown in Figure 1. The sampling locations and the characteristics of the Indonesian chicken populations used in the current study were previously described [24]. The unequal and relatively small sample sizes among populations reflect practical limitations related to DNA availability and quality from archived samples, rather than intentional sampling bias. Similar sample sizes have been commonly used in previous MHC-B SNP panel studies [22,23] and are sufficient for robust haplotype detection and diversity analyses, given the high resolution of the SNP panel. DNA samples used in this study were previously obtained from the chicken blood collected and preserved as described by Mustofa et al [24]. Briefly, blood samples were obtained from the wing vein, preserved in EDTA tubes at −20°C, and genomic DNA was isolated using the DNA extraction kit (gSYNCTMDNA Extraction Kit, #GS004, GS100, GS300; Geneaid Biotech). The DNA quality and quantity were verified using agarose gel electrophoresis and NanoDrop spectrophotometer (Thermo Fisher Scientific).

A set of 90 SNPs in the high-density MHC-B SNP panel, ranging from SNP marker MHCJ06 to MHC178, was utilized as described by Fulton et al [16,19]. Subsequently, these 90 SNPs were analyzed using a fluorescence-based (FAM and HEX) genotyping method known as KASP. Genotyping quality control of the MHC-B SNP panel included the use of positive controls and negative controls. All SNP assays produced clear and well-separated genotype clusters, with no ambiguous genotype calls detected; therefore, no SNPs or samples were excluded from downstream analyses. However, the MHC138 SNP showed ambiguous genotyping results in a subset of samples. To resolve this issue, locus-specific primers were designed for the MHC138 locus (MHC138F: GGTGTC ACCAGGGAGAGGTTG; MHC138R: CTCCCTCAGGTT GTGAAGCGAAAG), and direct sequencing was performed to accurately determine the corresponding alleles.

Haplotypes were identified using both manual and computational methods. The manual approach followed published procedures [16,19,20]. For the computational process, PHASE haplotype reconstruction was performed using PHASE v2.1. Analyses were run with a burn-in of 1,000 iterations followed by 1,000 sampling iterations with a thinning interval of 1. A fixed random seed (1520) was used to ensure reproducibility. Phasing accuracy was evaluated using posterior probabilities generated by PHASE. Site-specific phase probabilities were examined using the PHASEPROBS output, and the majority of SNPs showed posterior probabilities of 1.00. SNP positions with posterior probabilities ≥0.90 were considered confidently phased. Only a small proportion of sites exhibited posterior probabilities below this threshold, indicating low overall phasing uncertainty and high confidence in the inferred haplotypes. The results from both methods were compared to ensure the absence of any discrepancies.

The haplotypes identified in the Indonesian populations were aligned with the standard MHC-B haplotypes of heritage breeds and elite commercial egg layer lines [16,19] as well as Asian haplotypes from South Korea [17], Sri Lanka [18], Bangladesh [22], and Vietnam [23]. The alignment was performed using Clustal Omega and the phylogenetic trees were generated using the MrBayes approach (model HKY85) in Geneious Prime software (Geneious Prime 2024.0.7). Haplotypes were designated with the prefix “BSNP” to indicate that they were based on MHC-B SNPs and novel haplotypes were further labeled with “IND” to signify their origin in Indonesia. If a haplotype matched the standard haplotype across all 90 SNPs, it was labeled as standard. Otherwise, new names were assigned to novel haplotypes, with the numbering reflecting the sequence in which they were discovered (e.g., BSNP-IND01, BSNP-IND02, etc.) (Supplement 1). Observed (Ho) and expected heterozygosity (He) were calculated for each population using standard population genetic formulas implemented in R.

High genetic diversity was observed using the MHC-B SNP panel (Table 1), with a total of 126 MHC-B haplotypes were identified across six Indonesian native chicken populations. The number of haplotypes ranged from 11 in the NNK population to 38 in the MRG population, resulting in an overall average of 21 haplotypes per population. All haplotypes discovered in this study were unique to the Indonesian native chicken populations, with the highest percentage similarity being 50% against the standard haplotypes reported by Fulton et al [19]. Three haplotypes were shared among populations: BSNP-IND14 was shared by the KJM and STL populations; BSNP-IND26 was shared by the KJM, NNK, and MRG populations; the KJM and MRG Populations shared BSNP-IND31 (Figure 2). Notably, all haplotypes in the GAG and PLG populations were unique. The observed heterozygosity (Ho) values ranged from 0.857 in the NNK population to 1.000 in the PLG population, while the expected heterozygosity (He) values varied from 0.666 in GAG to 0.966 in MRG (Table 1).

No distinctive major clades were observed in the phylogenetic analysis. Nevertheless, the identified Indonesian MHC-B haplotypes can be divided into three subclades (Figure 2). The networks were colour-coded to differentiate between different subclades. All six populations were represented in each subclade. Although the origin of the breeds is from different islands in Indonesia, the mixed clustering shows the possible human-mediated movement during the colonization period or the transmigration people program from one island to another by the Indonesian government. Similarly, hierarchical clustering of principal component analysis by Tribudi et al [25] reported that MRG and NNK were grouped as one subcluster, showing that although the populations originated from different islands that human-mediated movement played a role in the mixing of these populations. Furthermore, genetic relationships based on microsatellite DNA polymorphisms of Kampung, PLG, STL, and KJM revealed that these local chickens have the same ancestor. Kampung and STL were clustered together, followed by KJM and PLG [21], which further confirmed the dispersed nature of the populations despite the different locations of origin. In addition, Ramezani et al [26] demonstrated that MHC-linked microsatellite LEI0258 variability showed that the LEI0258 alleles from Indonesian chickens clustered into two groups: one private cluster comprising STL, GAG, and NNK breeds and another cluster where MRG and PLG grouped together, while KJM was positioned alongside Iranian and Tanzanian ecotypes. Moreover, Sulandari et al [27] studied 16 Indonesian chicken populations, including KJM, MRG, PLG, and STL, consisting of 335 mitochondrial D-loop sequences. Their analysis revealed that the phylogenetic relationships among these populations are not linked to their geographical locations. Furthermore, strong evolutionary forces, particularly pathogen-mediated balancing selection and local adaptation, are likely responsible for maintaining highly distinct MHC-B haplotypes within each chicken population despite potential gene flow among islands. Polymorphism in the MHC-B locus is frequently maintained by pathogen-driven balancing selection, which promotes the retention of multiple alleles to recognize a range of antigens [28]. In addition, distinct and locally adapted haplotypes are probably maintained by variable selection pressures imposed by various ecological conditions and pathogen [8,29]. Such conditions promote the retention of particular MHC-B haplotypes that are best suited to cope with the unique disease challenges of each environment. Consequently, even if human-mediated movement facilitates genetic exchange, directional selection in local contexts may counteract homogenization, allowing populations to remain genetically similar at neutral loci while continuing to exhibit marked divergence at adaptive immune genes such as MHC-B.

Collectively, these studies highlight that Indonesian native chicken populations are characterized by a dispersed genetic structure shaped more by human-mediated movement, admixture, and the pathogen pressure, rather than geographical origin.

Comparison of MHC-B haplotypes in Indonesian native chickens with MHC-B standard haplotypes shows three possible subclades. Interestingly, except for one clade, the others show clustering of Indonesian MHC-B haplotypes with standard haplotypes (Figure 3). Some of these MHC-B standard haplotypes were also associated with certain B blood groups (serotypes) in chickens with known immune responses against diseases [8,10–14,30–33]. Despite this, none of the Indonesian native chickens’ MHC-B haplotypes exactly matched the standard haplotypes, confirming their novelty and suggesting the potential presence of distinct serotypes. Furthermore, this is the first study to reveal the MHC-B diversity in Indonesian native chickens using the MHC-B SNP panel. However, given that only six populations were included in the analysis, it remains possible that a complete match with standard MHC-B haplotypes may be identified in other untested populations. Leroy et al [34] reported that Indonesian native chickens retain some genetic characteristics from commercial chicken breeds and have not entirely replaced their genetic makeup.

Comparison of Indonesian MHC-B haplotypes with other Asian and standard haplotypes reveals that the majority of Indonesian haplotypes cluster with those from Bangladesh (Figure 4). Although the current study focuses on MHC-B SNP haplotype diversity, phylogenetic evidence from mtDNA D-loop sequences [35] reveals that Bangladeshi Red Junglefowls are closely related to Indonesian fowls (G. g. bankiva, distributed over Sumatra, Java, and Bali in Indonesia). Shared ancestry among South and Southeast Asian chicken populations could partly explain the observed patterns of MHC-B haplotype distribution in this study. However, after analyzing the complete mtDNA D-loop region, Oka et al [36] identified seven haplogroups and observed that Chinese and Korean chickens are derived from Southeast Asia, contradicting the findings of this study. Differences in genetic inheritance, evolutionary pressure, and historical demography can explain the above finding.

Moreover, Indonesia and Bangladesh both trace the ancestry of their chickens to Gallus gallus spadiceus, first domesticated in Thailand, Myanmar, and South China. According to Wang et al [37], Indonesian chickens show a mixture of G. g. bankiva and G. g. gallus, while South Asian populations, including Bangladesh, show admixture with G. g. murghi. The above results place Bangladesh and Indonesia as the endpoints of the ancient domesticated diffusion pathway from Southeast to South Asia, indicating a shared ancestral origin with region-specific gene flow. Moreover, it is evident that Southeast Asian maternal haplotypes have dispersed widely [38], maybe due to human-mediated movement; nevertheless, the extent and timing of such dispersal may differ from the gene flow patterns reflected in MHC-B. Compared with standard and global reference haplotypes, the highly divergent MHC-B haplotypes found in native Indonesian chickens indicate both evolutionary divergence and possible functional uniqueness. Private or novel alleles have been found in the BLB2 and LEI0258 loci in Indonesian native chickens, indicating adaptation to local pathogen pressures [24,39]. These genetic profiles may underline immunological traits that confer enhanced resistance to local pathogens. BLB2 correlates with enhanced antibody responses, including elevated IgY concentrations and resistance to diseases such as Newcastle Disease, suggesting functional immune benefits [39].

Several limitations should be considered when interpreting the results. Firstly, the limited number of samples may reduce the ability to detect already identified MHC-B haplotypes. Furthermore, due to a lack of disease challenge data, the functional association between the identified Indonesian MHC-B haplotypes and disease resistance could not be evaluated. Despite these limitations, this study provides valuable baseline information on MHC-B haplotype diversity in Indonesian native chickens.