Identification of Candidate Genes Associated with Meat Production of Aberdeen Angus Cattle

Simple Summary

This article identifies candidate genes associated with economically useful carcass traits in a population of Aberdeen Angus cattle using a GWAS. The study revealed a number of significant SNPs associated with economically useful traits (slaughter weight, meat yield, meat mar-bling). Functional annotation detected 33 genes associated with biological processes (GO BP) and 5 genes registered in the Cattle QTL database. For the OSBPL3 and RBFOX3 genes, within which a number of economically useful trait SNPs Hapmap39351-BTA-71190 and ARS-BFGL-NGS-11938 are localized, related with slaughter weight and meat marbling, we recommend developing tests for their inclusion in the genomic selection of beef cattle. Of interest is the OSBPL3 gene, within which the SNPs are located, so that tests that are unique and breed-specific can be developed. Our results provide useful information for understanding the genetic architecture and its further use in breeding.

Abstract

The use of GWASs in agriculture allows associations between SNPs and quantitative or qualitative traits in cattle to be estimated. This study investigated the relationships among a number of economically useful carcass traits (slaughter weight, meat yield, and meat marbling) with a group of SNPs that can act as DNA markers. Blood samples from Aberdeen Angus bulls (n = 260) were used as material for SNP genotyping. Genetic architecture for the kill yield trait showed the presence of 31 SNPs. Three SNPs were found to be the most reliable: BTB-00197584 (p = 7.20 × 10⁻⁵), Hapmap46735-BTA-86653 (p = 5.05 × 10⁻⁵), and BTB-00676077 (p = 7.02 × 10⁻⁵). There were 10 SNPs associated with meat yield. For meat yield, 7 SNPs were identified, four of them exceeding the threshold of established validity: ARS-BFGL-NGS-30557 (p = 1.28 × 10⁻¹²), ARS-BFGL-NGS-68920, ARS-BFGL-NGS-30466 on chromosome 8 (p = 1.20 × 10⁻⁶ and 3.03 × 10⁻⁶, respectively) and ARS-BFGL-NGS-40640 (p = 7.10 × 10⁻⁶). For marbling, 11 SNPs were identified on 6 of 29 chromosomes. These SNPs can act as potential markers for productivity assessment in cattle.

1. Introduction

Sequencing technologies have made it possible to study the genomes of cattle of different breeds and to analyze and compare their genetic structure [1]. The discovery of quantitative trait loci (QTL) has led to advances in the identification and comprehension of genetic modifications associated with economically significant phenotypic traits. Genome-wide association studies (GWASs), which use powerful techniques for the genetic analysis of single nucleotide polymorphism (SNP), are a relatively new approach to cattle genetic studies, which have identified different groups of candidate genes responsible for economically useful traits. This is especially important for research on cattle breeds whose genetic potential has not yet been fully realized. This is particularly true of the Aberdeen Angus breed.

The Aberdeen Angus breed of cattle is one of the most popular in the production of marbled meat. Animals of this breed are characterized by calm and malleable temperament, which makes them easy to handle and convenient for keeping on farms [2]. They also adapt easily to different housing and climatic conditions, making them suitable for different regions and farming systems. In terms of economic efficiency, the Aberdeen Angus breed has a good feed-to-meat conversion, which enables farmers to obtain higher yields at lower costs, while good fertility and high calf survival make them profitable for breeding. Earlier marker-assisted selection on Angus bulls in Hungary revealed the influence of a number of SNPs (DGAT1, TG, and Lep) on meat marbling [3]. There are relatively high frequencies of desirable alleles of Lep and CAPN1 genes in Russian beef cattle populations and a positive effect of allele T of the Arg4Cys LEP polymorphism on the fatness of animals from birth to weaning [4,5]. CAPN1_316 polymorphisms were found to affect the average weight gain in Aberdeen Angus cattle [4].

Although there are studies in Europe and the US, there is very little data on the genetic structure of Aberdeen Angus cattle and their meat productivity in Russia, which makes it impossible to provide breeders with reliable information on the breed for the conservation and protection of genetic diversity and use in commercial livestock breeding.

This study identifies candidate genes associated with economically useful traits in a population of Aberdeen Angus cattle using GWASs.

2. Materials and Methods

2.1. Ethical Statement

All activities related to animal care, the collection of biological material, and phenotype characterization were carried out in exact accordance with the regulations of the Federal State Budgetary Scientific Institution “Federal Research Center for Biological Systems and Agrotechnologies of the Russian Academy of Sciences”. Copies of the minutes of the 30 April 2024, Laboratory Animal Control Commission Meeting No. 2/1, were received on 6 May 2024. Data and samples were collected, and no animals were disturbed during data collection.

2.2. Animals and Phenotypes

The research work was carried out on a specialized territory for fattening animals located in the Kostroma region, Russia. Young Aberdeen Angus cattle (n = 260) were selected as the object of study. The area required to accommodate one bull in an outdoor feedlot was estimated at about 10 square meters. All individuals had identical sex and age characteristics and were under the same conditions of nutrition and care. At the time of the experiment, the age of the studied individuals was 532.0 ± 1 days. After 43 days, their weight just before slaughter was 617.6 ± 1.68 kg.

Their diet consisted of a mixture of corn silage and high-moisture corn feed. The animals also had free access to wheat straw located in the center of the feedlot. The feeders were cleaned before morning feeding. The feeding and maintenance conditions were the same for the whole group. The feed contained 53.5% dry matter, 11.75% protein, 10.3% fiber, 2.35% fat, and 5.4% ash. The animals had free access to water.

At the end of the fattening period, the cattle were sent to slaughter at a commercial meat processing facility. Animals were identified with an RFID chip. Carcasses were identified with tags and placed in a refrigerator 24–48 h after slaughtering. After cooling, the carcass was cut into marketable pieces.

2.3. DNA Samples Collection, Genotyping and Quality Control

For the genome-wide association search, 260 blood samples were obtained for genotypic analysis. Blood samples were used to extract genomic DNA. DNA extraction and genotype determination were performed using DNA-Extran (Syntol LLC, Moscow, Russia). Genomic DNA concentration was determined using Qubit (Thermofisher, Waltham, MA, USA), while DNA abundance was measured using Nanodrop ND-1000 (Thermofisher, Waltham, MA, USA).

2.4. Genome-Wide Association Studies

The genotyping of animals was performed using a high-density microarray BovineSNP50 (Illumina Inc., San Diego, CA 92122, USA) containing 53 218 SNPs according to the manufacturer’s standard methodology.

Unmapped SNPs and SNPs located on X and Y chromosomes and in mitochondrial DNA were excluded. SNPs located on autosomes were investigated. Plink 1.90 [6] was used to check the quality of the genotyped breeds based on the following criteria: the call-rate for the SNPs for each individual sample is at least 90% (—mind); the call-rate for each SNP across all genotyped samples is at least 90% (—geno); the minor allele frequency (MAF) is more than 0.05 (—maf); the deviation of SNP genotypes from the Hardy-Weinberg equilibrium in the samples had a significance level of p < 10⁻⁶ (—hwe): --assoc and --linear, to realize the methodology of multiple linear regression of the trait depending on the minor allele coding (0—homozygote for the most frequent allele, 1—heterozygote, 2—homozygote for the minor allele). Thus, the final equation had the general form of the following:

Y = μ + Xb + Kw + Sc + Za + e,

(1)

where Y is the vector of phenotypes; μ is the overall mean; b is the vector of fixed effects, including piggyback and year effects; w is the vector of the live weight of individuals when measurement is complete, treated as a covariance; c is the vector of SNP effects; a is the vector of random additive genetic effects with e ~ N (Gσα2), where G is the matrix of genomic ratios calculated from full genomic genotyping data using medium- or high-density DNA chips, and σα2 is the variance of polygenetic additives; K is the regression coefficient of the live weight of individuals when the measurement is completed; e is the vector of residual errors with e ~ N (Iσe2), where I is the unit matrix and σe2 is the residual variance; and X, S, and Z are the incident matrices for b, c, and a, respectively.

A heterozygosity test was performed to identify those samples that differed from the mean by more than three standard deviations in order to exclude them from the study. A total of 34 289 SNPs and 258 samples underwent quality control procedures and were retained for further study.

The p value (p < 1.45 × 10⁻⁵) of significance for single nucleotide polymorphisms was determined based on the Bonferroni correction method (0.05/N) [7], where N is the total number of SNPs remaining after quality control.

Data visualization was performed in the GWAS rMVP package using the R programming language [8].

2.5. Gene Identification, Function and Pathway Enrichment and Network Analysis

The region boundaries indicated according to the ARS-UCD2.0 genome assembly were applied to recognize cattle genes using the Ensembl 103 databases through the online resource [9]. The Database for Annotation, Visualization, and Integrated Discovery (DAVID), which provides researchers with a comprehensive set of functional annotation tools to understand the biological meaning of large gene lists [10], and the Cattle QTL database (https://www.animalgenome.org/cgi-bin/QTLdb/BT/index accessed on 27 May 2024), was used for functional gene annotation. Bioinformational data processing and work with graphs were carried out using R [8].

3. Results and Discussion

3.1. Information of SNPs

Quality control of SNPs was performed using PLINK 1.9. After filtering, the number of SNPs was 34 289. The filtered SNPs were distributed unevenly across all 29 chromosomes (see Figure 1). The GWAS rMVP package was used to visualize SNP density across chromosomes in the R program.

Figure 1. SNP density plot for the studied traits. The higher the SNP density, the redder the color.

The data revealed that some of the SNPs are associated with a number of phenotypic parameters that may be important economically. The data are presented in Table 1, showing the number and chromosome distribution of SNPs associated with slaughter weight, meat yield, and meat marbling.

Table 1. Number and chromosome distribution of SNPs associated with the indicators.

Indicators	nSNP *	Chromosome Distribution
Slaughter weight, kg	31	1, 3, 4, 7, 8, 10, 11, 14, 15, 16, 17, 18, 21, 25, 26, 29
Meat yield on bone, %	10	2, 5, 8, 9, 16, 21
Marbling, cat.	11	5, 6, 9, 11, 19, 25

* nSNP is the number of reliably detected single nucleotide polymorphisms.

3.2. Association Analysis and the Identification of Candidate Genes Related to Carcass Yield

The genetic architecture showed the presence of 31 SNPs located on 16 chromosomes but significantly associated with the trait. No SNPs exceeding the established threshold of the full genomic study were identified (Figure 2, Table 1).

Figure 2. (A) Distribution of SNPs across cattle chromosomes in relation to confidence level (−log10 (p)) by probability summation (blue line, p < 0.00001) and Bonferroni criterion (red line, p < 1.20 × 10⁻⁶) for slaughter weight. (B) Quartile of the probability distribution of expected and observed deviations from normal distribution for confidence values.

Three SNPs were found to be the most reliable: BTB-00197584 on chromosome 4 (p = 7.20 × 10⁻⁵), Hapmap46735-BTA-86653 on chromosome 14 (p = 5.05 × 10⁻⁵), and BTB-00676077 on chromosome 17 (p = 7.02 × 10⁻⁵) (Table 2).

Table 2. List of SNPs significantly associated with slaughter weight.

Chromosome Number	SNP *	Position	p
1	ARS-BFGL-NGS-116207	36 239 709	1.75 × 10−5
	ARS-BFGL-NGS-103594	149 043 397	3.94 × 10−5
	Hapmap43997-BTA-106314	149 067 642	4.58 × 10−5
3	BTB-00132025	64 044 549	3.22 × 10−5
	BTB-00148451	106 510 754	7.05 × 10−5
	ARS-BFGL-NGS-27698	106 543 574	6.61 × 10−5
4	Hapmap39351-BTA-71190	71 031 890	9.03 × 10−5
	BTB-00197584	74 579 013	7.20 × 10−5
7	BTA-117735-no-rs	86 169 121	2.03 × 10−5
8	ARS-BFGL-NGS-21043	494 866	7.05 × 10−5
	Hapmap60592-rs29025478	79 660 034	3.18 × 10−5
	ARS-BFGL-NGS-2393	105 195 851	2.25 × 10−5
10	ARS-BFGL-NGS-87984	59 731 691	5.22 × 10−5
11	ARS-BFGL-NGS-6860	26 162 970	6.10 × 10−5
14	Hapmap40063-BTA-34263	22 983 665	8.82 × 10−5
	Hapmap46735-BTA-86653	25 401 722	5.05 × 10−5
	ARS-BFGL-NGS-44032	25 425 357	6.69 × 10−6
	UA-IFASA-8451	68 026 718	5.66 × 10−5
	ARS-BFGL-NGS-115921	76 472 657	2.93 × 10−5
	ARS-BFGL-BAC-23729	76 803 498	4.65 × 10−5
	ARS-BFGL-NGS-113293	77 274 386	3.78 × 10−5
15	ARS-BFGL-NGS-24733	29 271 879	4.01 × 10−5
16	Hapmap58079-rs29010441	21 212 680	8.13 × 10−5
17	BTB-00676077	29 057 538	7.02 × 10−5
18	ARS-BFGL-NGS-21074	18 772 590	4.94 × 10−5
21	ARS-BFGL-NGS-31918	14 915 708	5.25 × 10−5
	ARS-BFGL-NGS-23752	60 065 494	2.98 × 10−5
25	ARS-BFGL-NGS-17307	24 092 836	4.38 × 10−5
26	UA-IFASA-5698	42 673 967	7.93 × 10−5
	ARS-BFGL-NGS-32465	47 288 628	4.24 × 10−5
29	ARS-BFGL-NGS-9388	34 746 857	9.60 × 10−5

Note: p is the reliability of the identified SNP. * SNPs exceeding the upper threshold of reliability (p < 1.20 × 10⁻⁵) are marked in bold.

3.3. Association Analysis and Identification of Candidate Genes Related to Meat Yield in Cattle

Ten SNPs on chromosomes 2, 5, 8, 9, 16, and 21 are associated with the trait of meat yield (Figure 3, Table 1).

Figure 3. (A) Distribution of single nucleotide mutations on cattle chromosomes in relation to the level of confidence (−log10 (p)) by probability summation (blue line, p < 0.00001) and Bonferroni criterion (red line, p < 1.20 × 10⁻⁶) for meat yield. (B) Quartile of probability distribution of the expected and observed deviations from the normal distribution for confidence values.

For the meat yield, 7 SNPs were identified, four of them exceeding the threshold of established reliability: ARS-BFGL-NGS-30557 on chromosome 5 (p = 1.28 × 10⁻¹²), ARS-BFGL-NGS-68920, ARS-BFGL-NGS-30466 on chromosome 8 (p = 1.20 × 10⁻⁶ and 3.03 × 10⁻⁶, respectively) and ARS-BFGL-NGS-40640 on chromosome 16 (p = 7.10 × 10⁻⁶) (Table 3).

Table 3. List of SNPs significantly associated with meat yield per bone.

Chromosome Number	SNP *	Position	p
2	ARS-BFGL-NGS-92825	4 201 525	7.58 × 10−5
5	ARS-BFGL-NGS-15506	88 334 676	2.52 × 10−5
	ARS-BFGL-NGS-30557	104 972 270	1.28 × 10−12
8	ARS-BFGL-NGS-68920	10 600 564	1.20 × 10−6
	ARS-BFGL-NGS-30466	85 786 900	3.03 × 10−6
9	Hapmap43790-BTA-121886	8 425 284	6.59 × 10−5
	ARS-BFGL-NGS-170	43 855 787	6.61 × 10−5
16	ARS-BFGL-NGS-40640	49 730 423	7.10 × 10−6
21	ARS-BFGL-NGS-27377	14 127 028	2.59 × 10−5
22	ARS-BFGL-NGS-119471	53 981 070	5.79 × 10−5

Note: p is the reliability of the identified SNP. * SNPs exceeding the upper threshold of reliability (p < 1.20 × 10⁻⁵) are marked in bold.

3.4. Association Analysis and the Identification of Candidate Genes Related to Meat Marbling

The meat marbling category identified 11 SNPs on 6 of 29 chromosomes (Figure 4, Table 1).

Figure 4. (A) Distribution of single nucleotide mutations on cattle chromosomes in relation to the level of confidence (−log10 (p)) by probability summation (blue line, p < 0.00001) and Bonferroni criterion (red line, p < 1.20 × 10⁻⁶) for meat marbling. (B) Quartile of probability distribution of the expected and observed deviations from normal distribution for confidence values.

Meat marbling—the presence of intramuscular fat—is of great economic importance for beef cattle. The Aberdeen Angus breed is considered the best, and the breed purity and origin of animals are important, as only specially bred cattle have the genetic predisposition for accumulating intramuscular fat. The genetic architecture of meat marbling and the identification of genes responsible for the formation of the best grading category by full genome association study on Russian populations of the Aberdeen Angus breed has not been previously conducted.

In our studies, 6 SNPs were identified, of which 3 exceeded the threshold of established validity: ARS-BFGL-NGS-30557 located on chromosome 5 (p = 4.83 × 10⁻⁸), BovineHD0900002742 on chromosome 9 (p = 7.61 × 10⁻⁶) and ARS-BFGL-NGS-36573 on chromosome 19 (p = 8.08 × 10⁻⁶) (see Table 4).

Table 4. List of SNPs significantly associated with meat marbling.

Chromosome number	SNP *	Position	p
5	ARS-BFGL-NGS-30557	104 972 270	4.83 × 10−8
6	Hapmap42761-BTA-05111	31 442 288	2.15 × 10−5
9	BovineHD0900002742	11 030 945	7.61 × 10−6
	BovineHD0900002745	11 034 233	8.24 × 10−5
	BovineHD0900002749	11 045 602	3.50 × 10−5
	BTA-28590-no-rs	11 178 996	3.37 × 10−5
	BTA-84628-no-rs	86 231 860	8.79 × 10−5
11	ARS-BFGL-NGS-44412	91 578 841	4.39 × 10−5
19	ARS-BFGL-NGS-11938	53 239 903	4.67 × 10−5
	ARS-BFGL-NGS-36573	53 831 725	8.08 × 10−6
25	Hapmap43799-BTA-17809	19 123 438	9.52 × 10−5

Note: p is the reliability of the identified SNP. * SNPs exceeding the upper threshold of reliability (p < 1.20 × 10⁻⁵) are marked in bold.

3.5. Structural Annotation of Genes Localized Within and/or in Close Proximity to Selected Regions of the Animal Genome

The structural annotation describes the exact location of various elements in the genome, such as open reading frames (ORFs), coding sequences (CDSs), exons, introns, repeats, splicing sites, regulatory motifs, start and stop codons, and promoters.

Table 5 presents the structural annotation of the genes identified for the studied traits. Structural annotation revealed the presence of 175 genes on 23 of 29 chromosomes.

Table 5. Structural annotation of genes within which and/or in close proximity to SNPs.

Item Number	Chromosome Number	SNP	Position	p	Sign	Gene *
1	1	ARS-BFGL-NGS-116207	36 239 709	1.75 × 10−5	SW	CGGBP1, CSNKA2IP
2		ARS-BFGL-NGS-103594	149 043 397	3.94 × 10−5		HLCS, RIPPLY3, SIM2
3		Hapmap43997-BTA-106314	149 067 642	4.58 × 10−5
4	2	ARS-BFGL-NGS-92825	4 201 525	7.58 × 10−5	MY	HS6ST1, UGGT1
5
12	3	BTB-00148451	106 510 754	7.05 × 10−5	SW	MACF1, BMP8A, KIAA0754
13		ARS-BFGL-NGS-27698	106 543 574	6.61 × 10−5
15		Hapmap39351-BTA-71190	71 031 890	9.03 × 10−5	SW	OSBPL3, GSDME, PALS2
16		BTB-00197584	74 579 013	7.20 × 10−5	SW	CFAP69, GTPBP10, FAM237B, HUS1, TNS3
20		ARS-BFGL-NGS-15506	88 334 676	2.52 × 10−5	MY	CMAS
21		ARS-BFGL-NGS-30557	104 972 270	4.83 × 10−8	Grade	NTF3
23	6	Hapmap42761-BTA-05111	31 442 288	2.15 × 10−5	Grade	GRID2
24	7	BTA-117735-no-rs	86 169 121	2.03 × 10−5	SW	COX7C
25	8	ARS-BFGL-NGS-21043	494 866	7.05 × 10−5	SW	MFSD14B, ANXA10, DDX60
26		ARS-BFGL-NGS-68920	10 600 564	1.20 × 10−6	MY	ELP3, PNOC, NUGGC, SCARA5
27		Hapmap60592-rs29025478	79 660 034	3.18 × 10−5	SW	TUT7
28		ARS-BFGL-NGS-30466	85 786 900	3.03 × 10−6	MY	SPTLC1, ROR2
29		ARS-BFGL-NGS-2393	105 195 851	2.25 × 10−5	SW	PAPPA
30	9	Hapmap43790-BTA-121886	8 425 284	6.59 × 10−5	MY	ADGRB3, LMBRD1
31		ARS-BFGL-NGS-170	43 855 787	6.61 × 10−5	MY	PRDM1, ATG5
32		BTA-84628-no-rs	86 231 860	8.79 × 10−5	Grade	UST
34		ARS-BFGL-NGS-87984	59 731 691	5.22 × 10−5	SW	SPPL2A, USP8
35	11	ARS-BFGL-NGS-6860	26 162 970	6.10 × 10−5	SW	LRPPRC, DYNC2LI1, PLEKHH2, ABCG5, ABCG8, PPM1B
37	14	Hapmap40063-BTA-34263	22 983 665	8.82 × 10−5	SW	XKR4, TMEM68, TGS1, LYN
38		Hapmap46735-BTA-86653	25 401 722	5.05 × 10−5	SW	TOX
39		UA-IFASA-8451	68 026 718	5.66 × 10−5	SW	PTDSS1, MTERF3, UQCRB, GDF6
40		ARS-BFGL-NGS-115921	76 472 657	2.93 × 10−5	SW	WWP1, ATP6V0D2, CA13, LRRCC1, PSKH2, CA1, RBIS, CA2, CA3, E2F5
41		ARS-BFGL-BAC-23729	76 803 498	4.65 × 10−5	SW
42		ARS-BFGL-NGS-113293	77 274 386	3.78 × 10−5	SW
43	15	ARS-BFGL-NGS-24733	29 271 879	4.01 × 10−5	SW	PHLDB1, UBE4A, ATP5MG, KMT2A, ARCN1, TMEM25, TTC36, TREH, DDX6
44	16	Hapmap58079-rs29010441	21 212 680	8.13 × 10−5	SW	SPATA17, GPATCH2
46		ARS-BFGL-NGS-40640	49 730 423	7.10 × 10−6	MY	PRDM16, ARHGEF16, MEGF6
47	17	BTB-00676077	29 057 538	7.02 × 10−5	SW	SCLT1, JADE1
48	18	ARS-BFGL-NGS-21074	18 772 590	4.94 × 10−5	SW	TENT4B, ADCY7, BRD7, NKD1
49	19	ARS-BFGL-NGS-11938	53 239 903	4.67 × 10−5	Grade	RBFOX3, ENGASE, CANT1, LGALS3BP, CYTH1, PGS1, THA1, TMC6, TMC8, BIRC5, TIMP2, C1QTNF1, CEP295NL, USP36, DNAH17, SOCS3, AFMID, TK1, TMEM235, SYNGR2
50		ARS-BFGL-NGS-36573	53 831 725	8.08 × 10−6	Grade
52	21	ARS-BFGL-NGS-27377	14 127 028	2.59 × 10−5	MY	CHD2, RGMA
53		ARS-BFGL-NGS-31918	14 915 708	5.25 × 10−5	SW	SLCO3A1
54		ARS-BFGL-NGS-23752	60 065 494	2.98 × 10−5	SW	CLMN, DICER1
58		ARS-BFGL-NGS-119471	53 981 070	5.79 × 10−5	MY	TMEM158, CLEC3B
60		Hapmap43799-BTA-17809	19 123 438	9.52 × 10−5	Grade	ABCA14, CRYM, ANKS4B, ZP2, LDAF1, DNAH3
61	26	UA-IFASA-5698	42 673 967	7.93 × 10−5	SW	DMBT1, SPADH2, SPADH1, FAM24A, PSTK, ACADSB, IKZF5
62		ARS-BFGL-NGS-32465	47 288 628	4.24 × 10−5	SW	CLRN3, FOXI2, PTPRE
63	29	ARS-BFGL-NGS-9388	34 746 857	9.60 × 10−5	SW	NTM

Note: SW—slaughter weight; MY—meat yield; Grade—marbling category. * Genes within which the identified SNPs are localized are shown in bold.

3.6. Functional Annotation of Genes Localized Within and/or in Close Proximity to Selected Regions of the Animal Genome

Functional annotation analysis involved annotating genes with GO terms and pathway information (see Table 6).

Table 6. Functional annotation of identified genes using the web-based DAVID program.

Chromosome Number	Gene *	DAVID
1	SIM2	Development of the nervous system and lungs
2	HS6ST1	Development of lung alveoli, development of neurons, development of blood vessels of the labyrinthine layer
	UGGT1	Catabolic process of misfolded protein
4	OSBPL3	Lipid transport
	GSDME	Positive regulation of the immune response to tumor cells
	CFAP69	Positive regulation of sperm flagellar motility, reaction to odor
8	DDX60	Defensive reaction to the virus
	ELP3	Central nervous system development
	SCARA5	Cellular iron ion homeostasis
	SPTLC1	Regulation of fat cell apoptosis process
9	ATG5	Immune system process
11	ABCG5	Cholesterol efflux, cholesterol homeostasis
	ABCG8	Cholesterol efflux, cholesterol homeostasis, negative regulation of intestinal cholesterol absorption
14	TMEM68	Lipid metabolism
	LYN	Adaptive immune response, innate immune response
	GDF6	Fat cells differentiation
	CA2	One-carbon metabolism
	CA1	One-carbon metabolism
	CA3	One-carbon metabolism
	CA13	One-carbon metabolism
15	TTC36	Tyrosine metabolic process, development of central nervous system neurons
16	SPATA17	Mitotic cell cycle, spindle organization, establishment of meiotic spindle localization
18	ADCY7	Regulation of adaptive immune response
19	RBFOX3	Nervous system development
	THA1	Metabolic process of cellular amino acids, process of glycine biosynthesis, catabolic process of threonine
	C1QTNF1	Glucose metabolism regulation
22	CLEC3B	Ossification, positive regulation of plasminogen activation, bone mineralization
25	ABCA14	Lipid transport
	CRYM	Thyroid hormone metabolic process, thyroid hormone transport, thyroid hormone transport
	ZP2	Binding of spermatozoa to the zona pellucida, prevention of polyspermia
26	SPADH2	Single fertilization
	SPADH1	Single fertilization
	ACADSB	Fatty acid metabolism

* Genes within which the identified SNPs are localized are shown in bold.

The HS6ST1 gene is associated with muscularity in Marchigiana and Chianina breeds and has significant additive associations with carcass weight in crossbred cattle slaughtered between 12 and 36 months of age with reported carcass phenotypes [11]. In our studies, this gene is correlated with meat yield by biological processes associated with alveolar development in the lungs (GO:0048286), neuronal development (GO:0048666), and labyrinthine layer blood vessel development (GO:0060716). The SNP ARS-BFGL-NGS-92825 was found within the HS6ST1 gene, and therefore, it can be recommended for further test development.

The UGGT1 gene is associated with characteristics including carcass weight and carcass fatness in Italian cattle breeds [12]. The OSBPL3 gene has previously been associated with the regulation of feed intake and energy expenditure in Nellore cattle, and it acts as a lipid transporter or sensor at membrane contact sites, affecting lipid metabolism [11]. In the analysis of our bull population, this gene is correlated with slaughter weight, and the SNP Hapmap39351-BTA-71190 is located where the single nucleotide substitution occurs. The GSDME gene is associated with feed conversion and feed intake, and the CFAP69 gene is associated with mammary immune system mechanisms [11].

The DDX60 gene was previously identified as predicting buffalo semen quality and fertility [13]. The ELP3 gene is significantly associated with predicted residual feed intake in commercial beef cattle littermates [14]. In our studies, the SNP ARS-BFGL-NGS-68920 was found within this gene and is correlated (p = 1.20 × 10⁻⁶) with meat yield. The SCARA5 gene is involved in iron delivery and is located on the centromere end of chromosome 8 of the bovine genome, which is essential for the genetic predisposition of muscle mineral composition in beef cattle [15].

The BTA9:43.3–43.6 Mb region, associated with meat yield, includes the ATG5 (autophagy 5) gene, which has an autophagy function mainly in preimplantation ovaries [16]. This autophagy function was previously found in significant amounts in ovaries under heat stress in pigs [17]. This gene has also been associated with embryo development after four- and eight-cell stages in mice [18]. An ATP-binding cassette, subfamily G (WHITE), member 5 (ABCG5) was identified on chromosome 11 in the 27 Mb region. ABCG family members are associated with cellular lipid transport into macrophages and hepatocytes in Nelore cattle [19].

The TMEM68 gene has been linked to feed intake and growth phenotypes in cattle [20]. In our study, it was significantly associated with the slaughter weight of animals. The TTC36 gene has been identified in cattle of the Braford breed as resistant to skin mite infestation [21].

The SPATA17 gene was previously identified in beef cattle as influencing reproduction and muscle formation [22]. In this study on Aberdeen Angus cattle, it is correlated with slaughter weight, and SNP Hapmap58079-rs29010441 was detected within it.

The RBFOX3 gene is a candidate regulator of mineral content in the muscle of Nelore cattle [23]. In our study, this gene is associated with meat marbling, and within it, SNP ARS-BFGL-NGS-11938 is localized on chromosome 19. The expression of the C1QTNF1 gene was detected in the adipose tissue of lactating Holstein cows [24]. The CLEC3B gene is involved in calcium ion binding and encodes a protein that is important for bone mineralization, cellular response to transforming growth factor beta stimulation, the positive regulation of plasminogen activation, and the development of the body’s skeletal system [25].

The ABCA14 gene is associated with the marbling category in our analysis, it was detected on chromosome 25, and within it, the SNP Hapmap43799-BTA-17809 was identified. Previously, this gene was found in a population of dairy cattle and correlated with male fertility [26]. The ZP2 gene, discovered by DNA extraction from semen straws of Holstein bulls, is responsible for oocyst maturation at different stages [27]. The analysis of commercial frozen semen from six Holstein Friesian bulls aged 2 and 4 years shows that two genes, SPADH1 and SPADH2, play a role in single fertilization [28].

By analyzing the functional annotation of the genes for the three traits studied, we can expand the understanding of the genetic architecture and select genes for further molecular validation.

Gene set enrichment analysis and gene set correlation were performed in the WebGestalt web-based program. To avoid false positives, it is necessary to designate enrichment categories using multiple tools (five enrichment tools for GO biological process, six for GO molecular function, and three for KEGG pathways) that are considered consistently enriched. Using these selection criteria, the general categories of biological, molecular, and cellular GO processes can be identified (see Figure 5).

Figure 5. Ratio of 44 functionally identified genes across the following library groups: biological, cellular, and molecular.

The results showed the highest reliability (p = 3.91 × 10⁻⁴) for the GO:0007338 libraries responsible for the biological process, namely, predisposition to single fertilization (high fertility). It included the genes SPADH1 (spermadhesin 1), SPADH2 (spermadhesin 2), and ZP2 (zona pellucida glycoprotein 2), and the enrichment ratio was 20,57. The following annotation of the identified genes was performed in the Cattle QTL database, and the results are presented in Table 7.

Table 7. Functional annotation of genes registered in the Cattle QTL database.

Chromosome Number	Gene *	Cattle QTL
4	OSBPL3	QTL 56147
11	ABCG5	QTL:31769
25	CRYM	QTL:168108
25	ZP2	QTL:57176; QTL:57209; QTL:57136
26	ACADSB	QTL:160374

* Genes within which the identified SNPs are localized are shown in bold.

The highest number of QTLs was identified for the RBFOX3 gene (15 QTLs). This gene and the OSBPL3 gene, the results for which are presented in Table 6 and registered 1 QTL, are significant for the further development of tests for them and their introduction into genomic selection programs of beef cattle.

4. Conclusions

The results provide useful information for understanding genetic architecture and provide some foundations for molecular breeding. The overall reliability of animal genotyping was 99.5%. Functional annotation performed using the web-based program DAVID detected 33 genes associated with biological processes of the organism (GO BP) and 5 genes registered in the Cattle QTL database. For the OSBPL3 and RBFOX3 genes, within which the SNPs Hapmap39351-BTA-71190 and ARS-BFGL-NGS-11938 are localized, interrelated with slaughter weight and meat marbling, we recommend developing tests for their inclusion in the genomic selection of beef cattle. Of special interest is the OSBPL3 gene, within which the identified SNPs are located, due to which it is possible to develop unique and breed-specific tests in the future.

The study revealed a number of significant SNPs associated with economically useful traits (slaughter weight, meat yield, meat marbling) in Aberdeen Angus cattle. Genes containing SNPs (OSBPL3, ABCG5, CRYM, ZP2, and ACADSB) were identified mainly for their involvement in biological quality and metabolism.