Genome characteristics and identification of reproduction-related genes in Southwest meat goats

Abstract

Background

Researching the genetic structure, genetic diversity, and the identification of reproduction-related genes in local goat populations is essential for developing effective conservation strategies and enhancing breeding efficiency.

Results

This study detected a total of 22,117,964 single nucleotide polymorphism sites and 5,563,682 insertion/deletion polymorphism sites across 136 samples from seven varieties. Population structure analysis divided the seven goat breeds into three primary genetic clades: Bor constituted an independent cluster; DZ and JC shared a common clade; while QB, MG, CZ, and YS formed a third distinct genetic group. Phylogenetic analysis indicated that indigenous meat goat breeds in Southwest China shared close genetic relationships and exhibited inbreeding. Gene flow results further confirmed the existence of genetic exchange among these southwestern meat goat populations. Genetic diversity analysis indicated that these local southwestern meat goat breeds possessed higher genetic diversity compared to Bor, suggesting potential heterozygote selection advantages. Genomics selection scans detected signatures of positive selection in genes regulating ovarian steroid biosynthesis (3BHSD, LOC102181972, PPP3CA), ovum development (EREG, MLH1), and meiotic division in oocytes (CDK2).

Conclusions

These discoveries have significant implications for enhancing thegenetic diversity of these local breeds, facilitating conservation efforts, and improving the adaptability of goats in Southwest China. Additionally, this research provides a foundation for investigating genomic characteristics in other important local goat breeds.

Introduction

Analyzing population genetic structure is a crucial approach to studying genetic variations among different populations. The primary goal is to elucidate genetic relationships and exchanges among populations, thereby providing insights into species evolution and adaptability [1, 2]. Genetic diversity represents a critical facet of biodiversity and serves as a fundamental tool for studying genetic polymorphism. Enhanced genetic diversity signifies greater evolutionary potential and heightened resilience to environmental shifts, underscoring its crucial role in species evolution and conservation efforts [3, 4]. As a result, advances in high-throughput sequencing have facilitated the identification of numerous candidate genes associated with reproductive performance in goats. However, despite these advances, research into goat reproductive traits has encountered challenges due to their complex nature as multigene-regulated quantitative traits with low heritability. Therefore, our understanding of these traits remains limited [5]. In this regard, whole genome sequencing (WGS) has become an essential approach for analyzing the role of multiple genes and loci in regulating reproductive traits and for identifying relevant genes and loci. Whole genome resequencing stands as one of the most rapid and effective methodologies for advancing animal breeding research. By sequencing the genome of a species with a known reference sequence, researchers can detect genetic variations such as single nucleotide polymorphisms (SNPs), insertion/deletion polymorphisms (InDels), structural variations, and copy number variations. These variations enable comprehensive analyses of genetic diversity, population structure, evolutionary patterns, and the identification of candidate genes associated with key traits.

Literature reports evaluating the genomic copy number variation of Beichuan white goat populations with high and low litter size by the fixation index (Fst) method found that 485 missense mutations in 1,739 genes were positively selected [6]. In Liu’s research on Xiangdong black goats, selective scanning identified genes associated with genomic regions crucial for fundamental biological processes. Key breeding candidate genes such as CCSER1, PDGFRB, IFT88, LRP1B, STAG1, and SDCCAG8 were pinpointed [7]. Lai’s study focused on extreme groups from two low- and high-fertility populations of Laoshan dairy goats. Through meticulous selective scanning and analysis, they identified specific selection genes associated with high fertility, including CCNB2, AR, ADCY1, DNMT3B, SMAD2, AMHR2, ERBB2, FGFR1, MAP3K12, and THEM4. Conversely, in the low-fertility group, distinct selection genes like KDM6A, TENM1, SWI5, and CYM were identified [8].

Recent findings indicate that the distribution of SNPs and genomic copy number variants around gene transcriptional start sites (TSSs) in Laoshan dairy goats may significantly influence their lambing numbers [9, 10]. Through WGS, several candidate genes crucial for ovarian development, including AURKA, ENDOG, SOX2, RORA, GJA10, RXFP2, CDC25C, and NANOS3, were isolated. An association analysis between mutation sites in these candidate genes and the litter size of 400 goats revealed significant findings. Specifically, mutations in the NANOS3 exon were linked to potential protein structure alterations, while variants in CDC25C, ENDOG, and NANOS3 showed notable associations with litter size [11]. In a study using Dazu black goats categorized into high-kid groups (3 − 5 lambs) and low-kidding groups (1 − 2 lambs), researchers identified 96 candidate genes associated with lambing number [12]. These findings represent a significant advancement in reproductive research on the Caprinae, laying a robust theoretical foundation for future investigations into the reproductive traits of goats [13].

Southwest China, characterized by its intricate topography, diverse climate, and rich local goat resources, presents a unique context for genomic studies. Studies suggest that goat populations in this region may represent some of the earliest domesticated goats in China. However, research on the genetic diversity of indigenous meat goats in Southwest China remains relatively limited. Efforts in this area could provide important insights into the genetic underpinnings of desirable traits in local goat populations, which may help guide breeding strategies and conservation efforts tailored to regional environmental conditions and production needs [14]. In this study, we leveraged cutting-edge whole genome resequencing technology to analyze the genetic makeup of two distinct breeds from Southwest China, the Qianbei ma goat and the Maguan hornless goat. Additionally, we incorporated data from the NCBI database for five other goat breeds, including Boer goat, Chuanzhong black goat, Jianchang black goat, Dazu black goat, and Yunshang black goat, to create a comprehensive genomic dataset comprising seven breeds. This approach enabled us to conduct a thorough comparative genomics study, identifying shared and breed-specific selection regions and genes across these populations.

Our analysis focuses on genes within these regions and functional mutations, shedding light on the genetic diversity and germplasm characteristics that define meat goat breeds in Southwest China. Our study reveals important genes and molecular markers linked to breed evolution and reproductive traits, providing essential knowledge for conserving and selectively breeding local meat goat breeds in the region. Furthermore, our findings lay the groundwork for future exploration into genes and molecular markers relevant to breed evolution and desirable breeding traits in goats. Ultimately, this study not only advances our understanding of the genetic basis of meat goat breeds but also contributes valuable insights into the protection and sustainable utilization of local meat goat breeds in Southwest China.

Experimental methods

Experimental materials

For WGS, a total of 30 whole blood samples from Maguan hornless goat(MG) were collected from farmers in Yunnan Province, while 30 whole blood samples from Qianbei ma goats (QB) were collected from the Qianbei ma goats original seed farm of Fuxing Animal Husbandry Co., Ltd. in Xishui County, Guizhou Province. Whole genome data for the Chuanzhong black goats (CZ), the Jianchang black goats (JC), the Yunshang black goats (YS), the Dazu black goats (DZ), and the Boer goats(Boer) were obtained from NCBI and China National Center for Bioinformation (CNCB) databases, and combined with self-generated data to form a population WGS dataset. Detailed information is shown in Table 1.

Table 1.

Sample information

Variety	Abbreviation	Sample size	Place of origin
Maguan hornless goat	MG	30	Maguan County, Yunnan Province
Qianbei ma goat	QB	30	Xishui County, Guizhou Province
Boer goat	Bor	9	South Africa
Chuanzhong black goat	CZ	12	Lezhi County, Sichuan Province
Dazu black goat	DZ	21	Dazu District, Chongqing
Jianchang black goat	JC	14	Huili City, Sichuan Province
Yunshang black goat	YS	20	Yunling Mountains, Yunnan Province

Whole genome resequencing

The DNA was extracted from the collected blood samples using a DNA extraction kit (Beijing Tiangen Biochemical Technology Co., Ltd., Beijing, China). The purity of the extracted DNA was assessed by Nanodrop (OD260/280), and the purity and integrity of the DNA were examined using 1% agarose gel electrophoresis. The Covaris crusher was used to randomly break the qualified DNA samples into 350 bp segments, and then the DNA segments were repaired by adding a ployA tail. Then, the sequencing adapter was added, and the Illumina sequencing primers were connected. The PCR amplification was performed to construct a cDNA library, and the constructed library was entrusted to Beijing Kangpu Shen Bioinformatics Technology Co., Ltd. to sequence the genomes of the MG and the QB on the Illumina HiSeq 2500 using the Illumina HiSeqPE150 sequencing strategy, with a depth of 30X. Finally, the raw data generated were supplied as a fastq file for cleaning and analysis.

Raw data quality control and filtering

The raw reads were filtered to generate clean data using the following criteria: (1) remove adaptor sequences; (2) remove bases containing non-AGCT at the 5’ end; (3) trim the end of the reads with sequencing quality scores below Q20; (4) remove reads containing N > 10%; and (5) discard reads shorter than 25 bp after filtering.

Genomics alignment and variant calling

QC-filtered high-quality reads were aligned to the goat reference baseline set (CHIR1.0) using samtools V1.16.1, and the aligned dataset was used for short variant detection of individual samples using the HaplotypeCaller tool in GATK v4.0 to obtain a gvcf file for storing the variant loci. Then, the SNPs and InDels were filtered using the VariantFiltration tool in GATK v4.0, and the valid SNPs and InDels were annotated by ANNOVAR.

Genetic structure analysis

SNP data from all samples were converted into FASTA format using VCF2PopTree. Subsequently, phylogenetic trees were constructed using the neighbor-joining method in MAGA7. Principal component analysis (PCA) was performed using GCTA, a software package originally developed to estimate the proportion of phenotypic variance in a complex trait that is explained by all genome-wide SNPs, calculating principal component 1 (PC1), principal component 2 (PC2), and principal component 3 (PC3). Scatter plots were generated using the principal components PC1, PC2, and PC3 as the x- and y-coordinates to reveal clustering relationships among samples. Population structure was analyzed using ADMIXTURE software, with K values ranging from 2 to 6. The CV error values were calculated to evaluate the genetic components of the 136 samples [15–17].

Genetic relationship analysis

GCTA software was used to explore genetic relationships among local meat goat populations in Southwest China, generating a genetic relationship matrix between samples.

Gene flow analysis

In order to explore the genetic material exchange between the Southwest meat goat populations, seven populations were analyzed for gene flow using treemix [18].

Linkage disequilibrium analysis

The LD coefficient (r²) between SNPs across pig breeds was calculated using PLINK (v 1.9) software. LD decay curves were plotted using the ggplot2 package in R.

Genetic diversity analysis

Genetic diversity was analyzed using VCFTOOLS, including measures such as population polymorphism index (θ), Tajima ′D test, polymorphism information content (PIC), expected heterozygosity (He), observed heterozygosity (Ho), within-population inbreeding coefficient (Fis), minimum allele frequency (MAF), and expected alleles (Ae).

Genome-wide selection signal analysis

Selective signal analysis was performed based on Fst and genomic nucleic acid diversity (Pi), and the window size for selective elimination analysis was determined based on the density of SNPs on the genome [19]. Using a 50 kb interval as a sliding window and 25 kb as a sliding step, Fst and Pi values ranked in the top 5% of each window were calculated to define selection regions, and genes in the selection regions were identified as candidate genes.

Enrichment analysis

Functional enrichment analysis of candidate genes in the selected region was performed using Goatools [20]. KEGG pathway analysis was conducted using KOBAS [21]. The false discovery rate (FDR) < 0.05 was required to determine the genes that were significantly overrepresented in those categories.

Results and analysis

Sequencing data quality analysis

After strict filtering of the sequencing data and the original data downloaded from the database, high-quality clean data were obtained. The results are shown in Table 2. A total of 10,173 G raw data were generated from 136 samples, and 9,966 G clean data were obtained after quality control. The Q30 after QC filtering was higher than 90%. The clean reads were aligned to the goat reference genome using BWA software, and more than 97.32% of the paired-end reads were aligned to the reference genome. These results met the requirements of resequencing analysis and were suitable for subsequent analyses.

Table 2.

Summary of sequencing data quality

Sampe	Size	Raw base (G)	Clean base (G)	Q30 (%)	CG (%)	Mapped ratio (%)	Dup ratio (%)
Boer	9	31.60	30.69	92.01	42.39	99.92	0.93
CZ	12	34.30	33.94	92.71	43.40	99.92	1.06
DZ	21	7.10	7.56	94.22	46.8	99.92	1.06
JC	14	23.19	22.89	89.72	43.78	99.69	2.04
MG	30	132.45	129.44	93.12	44.47	99.91	0.92
YS	20	61.87	61.04	94.06	43.72	99.92	0.88
QB	30	127.14	126.62	90.20	42.27	99.92	1.06

Sample indicates the name of the population; Raw base (G) indicates the total number of bases in the original sequencing data; Clean base (G) indicates the total number of bases after filtering; Q30 (%) indicates the percentage of bases with a quality value of more than 30; GC (%) indicates the GC content; Mapped ratio (%) indicates the comparison ratio

SNP variation detection and annotation

A total of 22,117,964 SNPs were detected in the 136 samples from seven goat breeds, including 15,444,561 located in the intergenic regions, 6,098,013 in introns, and 158,724 in exons (Fig. 1A). A total of 29,244 SNPs were identified in the seven breeds, of which 53 were unique to the DZ, 5,016 were unique to the CZ, 8,372 were unique to the Boer, 51,827 were unique to the YS, 88,155 were unique to the QB, 33,887 were unique to the MG, and 6,959 were unique to the JC (Fig. 2A). The number of SNPs in each population is summarized in Table 3. Classification of SNP annotation results by mutation function revealed 86,119 synonymous mutations, 1,207 nonsense mutations, and 71,398 missense mutations, which caused changes in amino acid sequences (Fig. 1A).

Fig. 1.

Variation detection and annotation of seven goat populations. A SNP variation detection and annotation. B InDel mutation detection and annotation

Fig. 2.

Variation detection results of seven goat breeds Verne diagram. A SNP detection results. B InDel test results

Table 3.

Statistics on SNP detection and annotation results of seven goat breeds

Population SNP statistics	Bor	CZ	DZ	JC	MG	QB	YS
Total number	6,189,167	14,364,604	44,092	8,309,453	13,188,450	16,087,921	15,393,838
Intergenic region	4,403,292	10,068,179	27,069	5,679,129	9,237,962	11,269,393	10,794,069
Downstream	40,024	100,405	647	63,358	93,278	112,909	109,119
Upstream	39,034	101,969	839	65,118	94,482	114,745	109,347
Intron	1,658,377	3,941,150	13,090	2,396,418	3,620,106	4,417,931	4,216,092
UTR	20,884	57,325	351	36,684	52,954	64,727	61,513
Exon	27,556	95,576	2,096	68,746	89,668	108,216	103,698
Nonsynonymous	12,725	41,624	1,271	29,271	39,072	47,319	45,831
Stopgain	196	635	32	441	574	721	722
Stoploss	31	65	4	45	73	78	80
Synonymous	14,604	53,252	789	38,989	49,949	60,098	57,065

InDel mutation detection and annotation

A total of 5,563,682 InDel sites were detected in the 136 samples from seven goat breeds, with 3,729,581 located in the intergenic regions, 1,667,298 in introns, and 39,965 in exons (Fig. 1B). The seven breeds had a total of 38,967 InDel loci, of which 10,500 were unique to the DZ, 1,090 were unique to the CZ, 224 were unique to the Bor, 3,989 were unique to the YS, 9,859 were unique to the QB, 6,377 were unique to the MG, and 347 were unique to the JC (Fig. 2B). The number of InDels in each population is summarized in Table 4. A total of 2,214,227 InDel loci were detected in QB, of which 1,496,932 were located in the intergenic regions, 9,011 in exons, and 651,749 in introns. In MG, a total of 5,274,342 InDel loci were detected, of which 3,525,885 were located in the intergenic regions, 1,588,499 in introns, and 108,216 in exons. From the 136 samples, InDel function annotations revealed a total of 13,900 frameshift deletions, 12,916 frameshift insertions, 5,727 non-frameshift deletions, and 5,783 non-frameshift insertions (Fig. 1B).

Table 4.

Statistics of indel assay and annotation results of seven goat breeds

Population SNP statistics	Bor	CZ	DZ	JC	MG	QB	YS
Total number	4,277,262	5,132,567	67,460	5,113,109	5,274,342	2,214,227	2,861,296
Intergenic region	2,885,280	3,444,898	38,598	3,410,583	3,525,885	1,496,932	1,933,429
Downstream	34,735	43,740	1,006	44,024	45,700	20,393	26,135
Upstream	32,895	44,838	1,499	44,787	47,037	23,339	26,482
Intron	1,280,987	1,536,798	24,925	1,549,270	1,588,499	651,749	846,351
UTR	20,145	26,741	651	26,865	28,163	12,803	15,679
Exon	23,220	35,552	781	37,580	39,058	9,011	13,220
Frameshift deletion	8,160	12,356	377	13,074	13,582	3,268	4,597
Frameshift insertion	7,584	11,478	176	12,184	12,623	2,480	4,082
Nonframeshift deletion	3,233	5,105	134	5,296	5,572	1,749	2,243
Nonframeshift insertion	3,248	5,148	74	5,440	5,664	1,254	1,811
Termination of translation	972	1,433	19	1,547	1,578	246	472
Termination codon removal	23	32	1	39	39	14	15

Genetic structure analysis

Based on the genome-wide SNP data from the 136 samples, a neighbor joining phylogenetic tree was constructed (Fig. 3A). The samples of each variety were clustered together and separated from other varieties with high reliability. The six Southwest meat goat breeds were divided into two major branches, with the Boer grouped on a separate branch. Geographically, the JC and the CZ were located close together, but the JC and the DZ were clustered together on a branch. The MG and the YS were geographical neighbors, but the Ma Guan hornless goats were clustered together with the QB, while the CZ and the YS were clustered together. These results indicated that there was a genetic exchange between the southwestern and the central goat populations of China.

Fig. 3.

Genetic structure analysis of seven goat populations. A Population phylogenetic tree analysis. B Principal component analysis. C Structure analysis

The PCA results showed that the PC1/PC2 scatter plot clustered Boer separately, while DZ and JC clustered in one group, and QB, MG, CZ, and YS clustered into another group (Fig. 3B). The classification results observed in the PC1/PC3 plot were consistent with that in the PC1/PC2 plot. When combining the results of PC1, PC2, and PC3, the seven goat species in this study were categorized into three groups, which was consistent with the phylogenetic tree results. The proportion of genetic components of each ancestral population in each sample was calculated to assess the population structure. Each column in the figure represents an individual, where the lengths of the different colored segments indicate the proportion of an ancestral component in the genome of that individual (Fig. 3C). When K = 2, Bor and YS were separated from other local goats in the PCA subgroup; and when K = 3, Bor were completely separated. The Southwest local goat breeds also showed mixed ancestral components; and when K = 4, the QB and the MG were separated, but the ancestral components of the CZ, DZ, and JC were mixed. When K = 5, the JCwas isolated (Fig. 3C).

Genetic relationship analysis

As shown in Fig. 4, the seven goat populations appeared as seven distinct squares in the IBS distance matrix, indicating that individuals within each goat population were genetically closer to each other (darker colored squares) than to individuals from other populations. The squares of Bor and DZ were the darkest, suggesting that these two populations may be highly selective and that a certain degree of inbreeding existed within all the goat populations. Thus, there is a need to develop a reasonable selection and mating plan to strengthen the populations in the process of breeding preservation.

Fig. 4.

Visualization results of IBS distance matrix of seven goat populations. The darker the color, the closer the genetic relationship

Gene flow analysis

Gene flow was used to assess the exchange of genetic material among the Southwest meat goat populations, and the results are shown in Fig. 5. Results revealed strong gene exchange between the MG and the JC, as well as between the DZ and the CZ, while weak gene exchange was observed between the other goat populations.

Fig. 5.

Gene flow analysis. The arrows point towards the receiving group and are coloured according to the percentage of ancestry received from the donor

Linkage disequilibrium analysis

The chained imbalance results are shown in Fig. 6. The Southwest local meat goat breeds had the fastest decay rate, whereas the Bor had the slowest decay rate, possibly because the introduced goat population was subjected to high-intensity selective breeding, resulting in a lower rate of LD decay. In contrast, the Southwest local meat goat population had the fastest rate of LD decay due to its high polymorphism arising from random mating.

Fig. 6.

Group LD decay analysis. In the figure, the abscissa represents the distance where chain imbalance occurs, and the ordinate is the correlation coefficient of chain imbalance, which is a display form of LD value

Genetic diversity analysis

The genomic diversity results of the seven goat populations are shown in Table 5. The nucleotide diversity of the Boer goat was higher than that of the Southwest meat goats. The Tajima ‘D values of the QB, the MG, and the YS were greater than, indicating the presence of medium-frequency alleles in this population, which may be caused by a population bottleneck effect or balanced selection. The Tajima ‘D value of the CZ, the DZ, and the JC was 0, indicating a base mutation rate of the population genome equal to the base diversity of the population genome, consistent with the standard neutral evolution model. The Tajima ‘D value of the Bor population was less than 0, which may be caused by negative selection, indicating the presence of many low-frequency alleles.

Table 5.

Genetic diversity analysis results

Group	Bor	CZ	DZ	JC	MG	QB	YS
Tajima ′D	-0.20	0.00	0.00	0.00	0.06	0.06	0.04
θ	0.37	0.29	0.36	0.32	0.23	0.21	0.24
PIC	0.15	0.23	0.18	0.24	0.24	0.22	0.22
He	0.19	0.28	0.23	0.29	0.29	0.27	0.27
Ho	0.23	0.33	0.46	0.38	0.33	0.28	0.29
Fis	-0.21	-0.18	-1.00	-1.02	-0.14	-0.04	-0.07
MAF	0.29	0.24	0.48	0.27	0.24	0.20	0.22

The PIC and expected number of alleles (Ae) of the six Southwest local meat goat populations were higher than those of the Bor population, suggesting that the genetic diversity of the local populations was higher than that of the Boer. The observed heterozygosity (Ho) was higher than the expected heterozygosity (He) of the seven goat populations, indicating a potential heterozygous selective advantage and that the population was distantly bred.

Selection elimination analysis based on Fst and Pi

To better understand the selection signals in the Southwest local meat goat breeds, the Bor was used as a reference group, and the QB, MG, Jianchang black goat, Yunshang black goat, Dazu black goat, and Chuanzhong black goat were used as target groups. Two different selection methods, Fst and Pi, were used to screen the gene loci that were strongly selected during the breeding of Southwest local meat goats. The results were as follows:

Maguan hornless goat vs. Boer goat

The Bor served as the reference group, and the MG was used as the selection population. By calculating the Fst and Pi selection signals, 172 regions were selected within the top 5% of the MG, and 499 genes were annotated (Fig. 7).

Fig. 7.

Genome-wide analysis of Fst and Pi selection signals in Maguan hornless goat

Qianbei ma goat vs. Boer goat

The analysis of Fst and Pi selection signals showed that 193 regions were selected within the top 5% of QB, and a total of 521 genes were annotated (Fig. 8).

Fig. 8.

Genome-wide analysis of Fst and Pi selection signals in Qianbei ma goat

Jianchang black goat vs. Boer goat

The analysis of Fst and Pi selection signals showed that 271 regions were selected within the top 5% of JC, and a total of 688 genes were annotated (Fig. 9).

Fig. 9.

Genome-wide analysis of Fst and Pi selection signals in Jianchang black goat

Yunshang black goat vs. Boer goat

By calculating the Fst and Pi selection signals, 215 regions were selected within the top 5% of YS, and a total of 543 genes were annotated (Fig. 10).

Fig. 10.

Genome-wide analysis of Fst and Pi selection signals in Yunshang black goat

Dazu black goat vs. Boer goat

A total of 21 regions were selected within the top 5% of DZ, and 41 genes were annotated (Fig. 11).

Fig. 11.

Genome-wide Fst and Pi selection signal analysis of Dazu black goat

Chuanzhong black goat vs. Boer goat

A total of 243 regions were selected within the top 5% of the Chuanzhong black goat, and a total of 594 genes were annotated (Fig. 12).

Fig. 12.

Genome-wide analysis of Fst and Pi selection signals in Chuanzhong black goat

Enrichment analysis

To investigate the biological processes associated with the candidate genes, we performed GO and KEGG enrichment analyses using a significance threshold of.

FDR < 0.05. The GO analysis showed that within the biological process category, these candidate genes were mainly involved in reproduction, reproductive process, immune process, metabolic process, biological regulation, and stimulus-response. Cellular component analysis showed distribution in intracellular, protein-containing complexes and cellular anatomical entities. In addition, molecular function analysis showed that these genes were mainly associated with catalysis, adhesion, molecular function regulators, and transcriptional regulator activity (Fig. 13).

Fig. 13.

Candidate genes GO enrichment analysis. The vertical axis represents enriched GO terms, while the horizontal axis indicates the number of candidate genes within each term

The KEGG database annotation of the candidate genes showed that a total of 298 pathways were enriched. These pathways primarily included the metabolic pathway, olfactory transduction, secondary metabolite biosynthesis, spliceosome, ribosome, PI3K-Akt signaling pathway, neuroactive ligand-receptor interaction, cytokine-cytokine receptor interaction, and MAPK signaling pathways (Fig. 14).

Fig. 14.

Candidate genes KEGG analysis. The vertical axis represents KEGG pathways, while the horizontal axis indicates the number of candidate genes enriched in each pathway

Identification of reproductive candidate genes

The main reproductive pathways enriched for these genes were ovarian steroidogenesis including the 3BHSD, BMP15, BMP6, IGF1R, FSHR, LHR, CYP11A1, HSD17B8, L0C102181972, CYP17A1, STAR, and CYP19A1 genes; oogenesis such as the EREG, RXFP2, MLH1, SPO11, and PIWIL2 genes; and oxytocin signaling pathway including the GNAI2, CACNA2D2, RYR1, NRAS, CAMK4, MAPK7, NFATC3, NFATC4, PRKAA2, EEF2K, PPP3CA, and OXT genes. Genes enriched in oocyte meiosis were ANAPC1, CDK2, and ENSCHIG00000020164.

In the comparative analysis of Bor vs. MG and Bor vs. QB, we found that MLH1 and L0C102181972 on chromosome 22, 3BHSD on chromosome 3, and PPP3CA on chromosome 6 were under strong selection. These four genes are related to reproductive traits and may be related to the reproductive performance of Southwest meat goats (Fig. 15).

Fig. 15.

MLH1, L0C102181972, 3BHSD, and PPP3CA gene selection signal analysis

Discussions

In this study, 30 samples of each MGand QB were subjected to whole-genome resequencing with a sequencing depth of 30×. These data were combined with 14 WGS samples of JC, 20 samples of YS, 21 samples of DZ, 12 samples of CZ, and nine samples of Bor downloaded from the NCBI database to form a dataset of 136 samples. These 136 samples yielded a total of 9,966.47 G of clean data with a Q30 quality higher than 90%, which were compared with the reference genome of goats (ARS1) [22]. The comparison rate exceeded 98.41%, meeting the requirements for detecting genomic variation and enriching the genetic background of endemic Southwest meat goats.

There are many types of variation in biological genomes, and SNPs, which involve mutations in only a single base, are the simplest and most common type of mutation, accounting for approximately 90% or more of all genetic variation [23]. Since the high density of SNPs in the genome and their ease of detection, they are widely used in studies of population-level genetic variation [24]. In this study, a total of 22,117,964 SNPs were detected from seven goat breeds. These SNPs were mainly distributed in intronic regions, which may affect promoter activity, messenger RNA (mRNA) conformation (stability), and subcellular localization of mRNAs and proteins, potentially leading to disease [25]. Intergenic variants can also influence gene expression and post-transcriptional modifications. Annotation of these SNP loci revealed 86,119 synonymous mutations, 1,207 nonsense mutations, and 71,398 nonsynonymous mutations. Non-synonymous mutations alter the coding amino acids, affecting the structure, stability, and function of the resultant protein, whereas synonymous mutations may affect transcription and translation process of gene expression [11, 26, 27]. It was found that non-synonymous mutations in the AMH gene increased lambing rate in goats. Therefore, both nonsynonymous and synonymous mutations are important for production [28].

Whole genome resequencing is one of the most important tools to study the genetic structure of a population, which can help elucidate the formation and affinity of different varieties and subspecies [29]. Phylogenetic tree, PCA, and STRUCTURE analyses are commonly used analytical methods to study the genetic structure of populations, where phylogenetic analysis can help to understand the relationships between species and major transitions in the evolutionary process [30]. PCA analysis is often used to explore kinship among different populations, facilitating the detection and resolution of the genetic variation, while STRUCTURE analysis can detect genetic differentiation among different populations or individuals, cluster samples, and calculate the genetic components of each individual in different populations [31, 32]. Here, 136 samples from five Southwest local meat goat populations along with YS and the Bor, were analyzed for population genetic structure. All seven breeds could be effectively distinguished. Phylogenetic analysis revealed that the seven goat populations were divided into three major clades, with the six local goat breeds forming two major clades and the Boer grouped separately. This may be due to the small sample size of Boer, which affects the objectivity of the results. These findings agreed with the PCA analysis, which showed that Boer clustered separately into one group, the DZ and the JC were clustered into one group, while the QB, MG, CZ, and YS were clustered into one group. A previous study using whole genome resequencing to study the genetic structure of wild and domestic sheep worldwide found a close relationship was found between wild and domestic sheep through phylogenetic, PCA, and STRUCTURE analyses [33]. The results of structure analysis in this study showed that with the increase in K value, the degree of differentiation between Boer and Southwest local meat goat populations became more obvious, and varying degrees of differentiation were observed among Southwest local breeds.

The IBS distance matrix showed that although each population was genetically unique to some extent, the Bor and DZ populations were unique, suggesting that these two populations may be highly selected populations with potential inbreeding. The results of gene flow analysis further confirmed that the Southwest meat goat populations are closely related. In a previous study, the mitochondrial DNA control region was used to investigate the maternal population structure of 10 Chinese native goat breeds in Yunnan, revealing a non-significant phylogeographic structure, which is consistent with the results of this study [34]. These observations suggested that a reasonable selection plan should be formulated and the current breed conservation efforts should be strengthened.

Generally, LD decays faster in wild populations than in domesticated populations because, during domestication, artificial selection may cause certain beneficial gene fragments or haplotypes to become fixed, thereby slowing the rate of decline [35]. Whole genome resequencing showed that the average LD of Angus cattle (highly domesticated) was higher than the average LD of Yunling cattle (local breed) [36]. In this study, the LD decay rate of the Southwest local meat goat breeds was the fastest, whereas the decay rate of the introduced Boer was the slowest, possibly because introduced varieties were subjected to high-intensity selection and breeding, resulting in the slowest LD decay rate. Due to random mating and high levels of polymorphism, the Southwest local meat goat varieties had the fastest LD decay rate. These results revealed the genetic diversity, gene evolution, and population history of the Southwest meat goat and Bor populations, and provided a reference for further resource protection and utilization.

The evaluation of genetic diversity in species guides conservation strategies and reveals population genetic relationships [37, 38]. China has abundant genetic goat breed resources and is one of the countries with the most abundant goat populations in the world [39, 40]. However, due to the impact of introduced varieties and limited awareness of conservation, many local goat breeds are facing continuous population reduction and a decline in genetic diversity. In the past, microsatellite or mitochondrial DNA data were used to evaluate the genetic diversity of goat breeds. Currently, due to the maturity of WGS and the significant reduction in false positives, the accuracy of population genetic diversity evaluation has improved [41–43]. Due to the long-term influence of artificial selection, it is expected that the effective population size and genetic diversity of domestic goat population are reduced compared with the wild goat population. The Tajima ′D test is associated with population genetic diversity, and positive selection effects are detected by comparing two estimates of the population mutation rate, θw (reflecting the base mutation rate of the population genome) and θπ (reflecting the base diversity of the population genome) [44]. Under neutral evolutionary conditions, the values of θw and θπ are approximately equal, and the theoretical value of Tajima ′D is zero. If Tajima ′D > 0, it indicates the presence of a high frequency of alleles, which may be due to population bottleneck effects, population structure, or balancing selection. If Tajima ′D < 0, it may be due to negative selection, indicating the presence of many low-frequency allelic sites.

The results of the genomic diversity analysis of the seven goat populations in this study showed that Tajima ′D > 0 for the QB, the MG, and the YS, indicating the presence of high-frequency alleles and possibly a population bottleneck or balancing selection. According to the 2023 third census of genetic resources, it was found that the population size of the MG was only 472 individuals. Measures should be taken promptly to establish breed protection, and genes of other lineages should be introduced appropriately to rapidly expand their genetic basis and preserve the breed. Under the standard neutral evolution model, the Bor population had a Tajima ′D < 0, which may be caused by negative selection and indicated a large number of low-frequency allele loci, which was consistent with the results of Ferrando [45]. The nucleotide diversity of the Southwest meat goat population was lower than that of the Bor population, indicating that conservation efforts for local breeds should focus on improving genetic diversity. Moreover, the PIC and the expected number of alleles in the Southwest meat goats were higher than those of the Boer, suggesting a higher genetic diversity in the local population. Since Ho > He in the seven goat populations, there may be a heterozygous selective advantage.

Population selection pressure analysis found differences in allele frequency in different populations by comparing the genome sequences to screen candidate genes and identify regions under selection pressure [46]. This method is widely utilized to examine how populations adapt and the genetic structure involved in species evolution [47]. The comparison of Hainan black sheep with six other breeds from different geographic regions using selective elimination analysis identified a variety of functionally relevant candidate genes, including immune resistance to disease (TNFAIP2 and EXOC3L4), melanin biosynthesis (CDH15, ASIP, and PARD3), and photosensitivity (CNGB3 and CNBD1) [48]. Yang et al. analyzed selective pressures in populations of Relict Gull and found that olfactory signaling pathway genes (OR14C36, OR14J1, OR14I1, and OR14A16), inner ear development-related genes (PTN, PTPN11, GATA2, ATP8B1, and MYO15A), and genes associated with high-altitude breeding and hypoxic adaptation are closely linked to alterations in the life-history traits of this species [49]. Chai et al. conducted a selective sweep analysis on domestic yaks, wild yaks, and Tibetan cattle, identifying 11 differentiated genomic regions in domestic yaks, genes under potential positive selection in each group, as well as gene flow among them [50]. In this study, the Boer were used as the reference group, and the MG, QB, JC, YS, DZ, and CZ were used as the target groups. We screened selective regions and genes associated with growth, reproduction, adaptation, and disease resistance in a local southwestern meat goat population using Fst and Pi methods. These selected regions may be due to different genetic drift, gene exchange, environmental adaptation, and natural selection pressures, which could help us to understand the evolutionary history, ecological adaptation, and artificial selection among the Southwest meat goat populations.

The completion of biological processes in organisms usually requires the co-regulation of multiple pathways and genes. Therefore, using gene function enrichment analysis to identify signaling pathways and genes that play a key regulatory role in biological processes is of great significance for revealing and understanding the potential molecular mechanisms of biological processes. Identifying candidate genes for phenotypic variation in reproductive traits and their mutations can help breeders find genetic markers for reproductive traits [51]. Wang et al. searched for candidate genes associated with litter traits in goats by selective scanning and identified KIT, KCNH7, KMT2E, PAK1, PRKAA1, and SMAD9. These genes were enriched in reproductive-related pathways, such as steroid metabolism and cell response to hormone stimulation [52]. Also, the MLLT10, SPIRE2, TCF25, ZNF276, and FANCA genes were associated with low fecundity by comparing Jining Grey goats with Chinese domestic goats. The functional enrichment analysis of these genes showed that they were involved in biological processes such as mammary gland morphogenesis, uterine development, gastrulation, mesoderm morphogenesis and formation, and vascular development [53]. In this experiment, the main enrichment and reproduction-related signaling pathways were ovarian steroidogenesis, steroid hormone biosynthesis, oogenesis, and the oxytocin signaling pathway. Comparative analyses of Boer vs. MG and Boer vs. QB showed that the reproductive candidate genes, MLH1 and L0C102181972 located on chromosome 22, 3BHSD located on chromosome 3, and PPP3CA located on chromosome 6, were under strong selection in the Southwest goat populations. These genes may have undergone natural or artificial selection during the formation of reproductive traits in these goats. Therefore, these results may aid in the selection of reproductive traits during goat domestication, thus contributing to the improvement of goat breeding practices in China.

Conclusion

In this novel study, WGS data were utilized for the first time to investigate SNPs among Southwest local meat goat breeds including the QB, MG, JC, CZ, YS, DZ, and the Bor. The findings revealed a close genetic relationship among the local meat goat breeds in Southwest China, indicating significant gene exchange. Moreover, the genetic diversity observed in these local meat goat breeds exceeded that of the Bor, suggesting a potential heterozygous selection advantage. Furthermore, the study identified promising candidate genes and mutation loci for breeding. These discoveries have significant implications for enhancing the genetic diversity of these local breeds, facilitating conservation efforts, and improving the adaptability of goats in Southwest China. Additionally, this research provides a foundation for investigating genomic characteristics in other important local goat breeds.

Authors’ contributions

Yuan Zhang: Writing–review & editing, Writing–original draft, Data curation, Conceptualization. Yonghong Ju: Data curation, Conceptualization. Ruiyang Li: Methodology. Wen Tang: Software, Data curation. Jiajing Chen: Data curation. Jiafu Zhao: Conceptualization. Guangbin Zhou: Resources, Data curation. Xiang Chen: Resources, Data curation, Software.

Funding

This research was supported by the National Key Research and Development Program (grant no. 2021YFD1200403), the Guizhou High-level Innovative Talents Project (Qiankehe Platform Talents [2022] 021–1), and the Guizhou Provincial Science and Technology Program (Qiankehe support [2022] key 033).

Data availability

In this study, the genomic data of QB and MG have been uploaded to the NCBI database (accession number: PRJNA1262351). The genomic data of JC, CZ, YS, DZ, and Bor goat have been obtained from public databases, NCBI SRA database, and China National Center for Bioinformation (CNCB), with accession numbers in the order of PRJNA608539, PRJCA004405, PRJNA611688, PRJNA479946, and PRJNA548681.

Declarations

Ethics approval and consent to participate

The permissions for conducting the study in the goat farm in Guizhou, China was obtained.All procedures involving animals were approved and authorized by Guizhou University. The laboratory animal castration protocol for this study was approved by the Laboratory Animal Ethics of Guizhou University (No. EAE-GZU-2026-E003). This study has obtained the informed consent from the owner.

Competing interests

The authors declare no competing interests.