Population structure and selection signatures over generations in the composite Montana tropical ®

doi:10.21203/rs.3.rs-2374029/v1

Download PDF

Research Article

Population structure and selection signatures over generations in the composite Montana tropical ®

https://doi.org/10.21203/rs.3.rs-2374029/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Knowledge of population structure and breed composition is important for managing breed information within a breeding program. Composite Montana Tropical® was developed by crossing animals of four different biological types or racial groups. By using genomic data it is possible to conduct several studies of population structure and selection signatures. Pedigree and genotype data (51962 SNP) from 3215 Montana Tropical® animals were used to (1) predict the composition of biological groups in the pedigree and how this composition relates to generations within the composite breed; (2) characterize the distribution of runs of homozygosity (ROH) in relation to generations within the composite breed; (3) estimate and compare the inbreeding coefficient (F) based on information from genomic markers (FROH) and pedigree (FPED); (4) identify genomic regions of islands of ROH; and (5) identify related traits in each significant genomic region under selection. The population structure (PCA and ADMIXTURE) was not related to the racial composition of the pedigree and has a strong relationship with the number of generations within the breed. There was an increase in FROH as the number of equivalent generations increased. Estimates of FROH in the Montana Tropical® population also showed that from the equivalent generation five onwards, there was a reduced number of animals with FROH equal to zero. It is believed that using PCA results in a breed’s genetic improvement program to correct for population structure is an interesting alternative to replace the correction by NABC proportions based on pedigree. The population study showed selection signatures in regions of chromosomes 4, 6, 20 and 26. The frequency with which markers in homozygosity were observed increased with the generations within the breed, demonstrating that it is really a selection process that is occurring after the formation of the composite population. Most of these selection signature regions present carcass quality QTLs demonstrating that the objective of developing a breed of cattle for pasture production systems is being achieved at the genomic level over the generations of the breed.

Adaptation

pedigree

genomics

Knowledge of population structure and breed composition is important to manage breed information within a genetic improvement program and make more assertive and accurate decisions in the conservation of diversity, selection of traits of interest and to evaluate the adaptability of composite breeds to a given environment of production (Reist-Marti et al., 2003, Kuehn et al., 2011).

The estimation of breed composition based on pedigree is a conventionally used and accessible method, however it is subject to annotation errors, incomplete or missing data, which impairs the reliability of the information (Ron et al., 1996). Advances in genomic technologies have allowed for innovations and implementations of methods for estimating the racial composition of domestic animals (Huang et al., 2014). Methods that used whole-genome data have been used satisfactorily in species such as cattle, pigs and sheep for prediction and estimation of general composition and genomic racial proportion with regression coefficients (Dodds et al., 2014, Funkhouser et al., 2017, Gobena et al.,2018).

By using genomic data, estimation of breed composition is less subject to input errors and the effect of Mendelian sampling biases during gametogenesis that affect estimates based on pedigree (Kuehn et al., 2011). The disadvantage of using this technology to predict breed composition is that it costs more to obtain information (Gurgul et al., 2014). This cost can be mitigated due to the diversity of information that can be obtained when using the same genomic data, such as parentage testing, estimating genomic genetic values and tracking diseases (Gobena et al., 2018).

An additional information that can be obtained with genotypic data is the identification of ROHs which are continuous homozygous segments inherited from common ancestors (Ferencakovic et al., 2011, Purfield et al., 2012). The formation of ROHs can be influenced by inbreeding, genetic drift, population bottlenecks, as well as natural and artificial selection, that is, their study can characterize the history of the population structure (Curik et al., 2014).

According to the size of the ROH, it is possible to calculate the genomic inbreeding coefficient and classify inbreeding, with long segments related to recent inbreeding and short ones to older inbreeding (McQuillan et al. 2008). These regions, depending on the frequency they appear in the population, can be considered selection signatures and allow identification of characteristics of productive importance that are under selection throughout the development of a breed (Peripolli et al., 2018)

Composite Montana Tropical® was developed by crossing animals from four different biological types or breed groups: 1) Zebu breeds (Bos taurus indicus; N), 2) Adapted taurine breeds (Bos taurus taurus; A), 3) British breeds (Bos taurus taurus; B), and 4) continental European breeds (Bos taurus taurus; C) (Ferraz et al., 1999). Differences between biological groups can be used to adjust the genotype of animals to the environment and increase production efficiency, allowing better performance in environments that would be more challenging for pure breeds (Silva et al., 2007).

The composite Montana Tropical® has already been studied in previous works and has shown to have signature traits of recent selection (Peripolli et al., 2020). Using the Montana Tropical® composite population, we sought in this work to (1) for the first time predict the composition of the biological groups of the pedigree with genomic data and how this composition relates to the generations within the composite breed; (2) characterize the distribution of ROH in relation to generations within the composite breed; (3) estimate and compare the inbreeding coefficient (F) based on information from genomic markers (FROH) and pedigree (FPED) ; (4) identify genomic regions of ROH islands and (5) identify related traits in each significant genomic region under selection.

Population structure

According to the PCA, the first principal component (PC1) explained 19.66% of the proportion of variation between the animals of the composite Montana Tropical®, with the hierarchical structure related to the equivalent generation estimated by the pedigree (Fig. 1). The second principal component (PC2) explained 15.86% of the variation between animals.

Correlations between all principal components and breed proportion of animals derived from the pedigree were calculated. The first principal component was positively correlated with breed group A (r = 0.45; p < 0.001) and negatively correlated with breed group B (r=-0.44; p < 0.001). Next, PC3 showed a positive correlation with breed group N (r = 0.37; p < 0.001) and a negative correlation with breed group B (r=-0.43; p < 0.001). After this, PC4 showed a negative correlation with the breed group N (r=-0.23; p < 0.001) and a positive correlation with the breed group B (r = 0.25; p < 0.001).

Correlation analyses revealed significant positive correlations between the biological components of pedigree A (r = 0.33; p < 0.001) and C (r = 0.16; p < 0.001) with the estimated equivalent generations in the study population. Negative and significant correlations were observed between the biological components N (r=-0.19; p < 0.001) and B (r=-0.29; p < 0.001) and the equivalent generations estimated by the pedigree. The regression analysis confirmed the trend of a 4% increase in the composition of adapted taurine (A) in the pedigree of the animals at each generation and a 3% decrease in the proportion of British breeds (B) in the pedigree for each generation within the composite breed.

A regression analysis was performed to determine the value of the proportion of the animal's biological group through the PCA values. The regression equations with the principal component values were able to predict the NABC components in the pedigree, however these showed low regression coefficients that were inadequate to employ a prediction model for the biological components of the pedigree for composite Montana Tropical®. This indicates that the principal component structure represents a population structure different from the breed composition of the pedigree.

The analysis in the ADMIXTURE software demonstrated that at a value of K = 2 the animals with an equivalent generation (equiGen) < 2.5 presented a uniform composition. Additionally, clusters for animals 2.5 < equiGen|<4.5 were more definitively grouped with a smaller proportion of C2 compared to other animals. In equiGen > 4.5, a group of more differentiated animals with higher proportion of C2 could also be observed. At K = 3, the grouping pattern within generational classes remained similar to K = 2 with a greater presence of C2 in animals with equiGens > 4.5. The results of K = 4, on the other hand, clearly separate a cluster (C1) for equiGens > 4.5 and revealed a large composition variation for animals 2.5 < equiGens|<4.5 (Fig. 2).

The clusters showed the following correlation results with the pedigree components. First, C1 showed a significant and positive correlation with breed group A (r = 0.42; p < 0.001) and a significant and negative correlation with breed group B (r=-0.45; p < 0.001). Second, C3 showed a positive and significant correlation with breed group N (r = 0.43; p < 0.001). Finally, C4 showed a significant and positive correlation with breed group B (r = 0.65; p < 0.001) and a significant and negative correlation with A (r= -0.44, p < 0.001).

The analysis of the ADMIXTURE clusters revealed a positive correlation between C1(r = 0.58, p < 0.001) and C2(r = 0.23, p < 0.001) with the equiGens from the pedigree. Clusters C3(r=-0.45, p < 0.001) and C4(r=-0.42, p < 0.001) showed a negative correlation with the equivalent generations calculated by the pedigree. The regression analysis showed a trend of increase of 12% of C1, 4% of C2 and reduction of 9% of C3 and reduction of 0.7% of C4 in each generation.

It is not possible to state that C1 represents biological group A, as its intercept is negative (-0.22) indicating that this cluster arises within Tropical Montana. C2 apparently also does not represent the biological group A because its intercept values (0.03) do not match the use of A in the breed formation strategy. The representation of C3 by Nellore and C4 by Britânico seem to be consistent with the use of these biological groups over generations.

The result of the two methods of estimating the NABC composition of the pedigree were similar, that is, despite being statistically significant with a low regression coefficient. Component B was the one that obtained the best estimate both by the PCA and by the ADMIXTURE (r2 = 0.45). The regression coefficients of groups N (PCA r2 = 0.22 ADMIXTURE R2 = 0.24) and A (PCA r2 = 0.27 ADMIXTURE r2 = 0.25) were similar. Component C had the same r2 = 0.04 in the equation with PCA and ADMIXTURE being the lowest value among the breed groups. These results demonstrate that the genomic structure of the composite Montana Tropical® is different from the structure demonstrated by the pedigree.

The principal component analysis involving the results of the first 4 principal components of the genotypes, the 4 ADMIXTURE clusters and the NABC components of the pedigree (Fig. 3) demonstrated that the PC1 variation components of the genotypes, equivalent generation, C1 and C2 and the breed component A show similar direction. This result is consistent with the presented correlations of the population structure, suggesting that C1 and C2 together may be representing the use behavior of breed group A, and the emergence of a new cluster within Montana Tropical®, but not being able to individualize it. them between the clusters. C3, PC3 and N in the pedigree are also in the same direction. C4 and B also in the same direction. Therefore, the ADMIXTURE clusters indicate to be associated as follows: C1 with A and number of generations within the breed, C2 with number of generations within the breed, C3 with N and C4 with B.

Runs of Homozygosity

Of the 3215 evaluated animals, 3140 had some ROH segment in the genome, totaling 2349 blocks of haplotypes. The mean ROH of the Montana animals was 7.48 ± 0.08 Mb, ranging from 0 to 39 (Table 1). The average sum of all ROH segments per animal was 61.21 resulting in an average length of 8.18 MB.

Table 1

Statistical Homozygosity Runs (ROH) per animal
Total Sample	Média	SE	SD	Min	Max
SROH	61,21	1,10	62,89	0	845,53
NROH	7,48	0,08	4,97	0	39
LROH	8,18	0,05	7,81	1,01	109,11
FROH	0,02	0,0004	0,02	0	0,31

Standard error (SE), mean (Mb), standard deviation (SD), min (Mb), max (Mb), SROH (mean length of genome covered by ROH), NROH (mean number of ROH), LROH (mean length of ROH in Mb) and FROH (inbreeding coefficient)

The genomic inbreeding coefficient (FROH) showed a positive correlation with the inbreeding values calculated by the pedigree. The overall FROH of Composite Montana Tropical® calculated from different length categories was 2% and pedigree inbreeding 0.8%.

There was an increase in FROH as the number of equivalent generations increased, as seen in Fig. 3. FROH estimates in the Montana population also demonstrate that from equivalent generation 5 onwards, the number of animals with FROH = 0 is reduced.

Of the animals studied, 95.89% had some ROH sequence of size 1Mb < ROH < 6.5Mb, 33.81% of size 1Mb < ROH < 6.5Mb, 34% of size 8.7Mb < ROH < 15.7Mb. The largest sequences of ROH > 15.6 Mb (with a common ancestor in the previous 2.5 generations) were identified in 39.45% of the composite Montana Tropical® animals, indicating the emergence of recent inbreeding.

Equivalent generations > 4.5 showed significantly higher values (p < 0.0001) of inbreeding coefficient (FROH) than all generations, especially in runs of homozygosity > 15.6 Mb.

In each class of equivalent generation, 5 regions of higher frequency in overlapping homozygosis were located (Fig. 4), BTA 4 (position: 62316125–73792670), BTA 6 (positions: 16674075–24813978), BTA 20 (position: 23007509–51003225) and BTA 26 (position: 36413577–42757206).

The ROH islands found in BTA4, BTA 6, BTA 20 and BTA 26 showed their frequency in homozygosity growing as the number of equivalent generations increased. Thus, this indicates that these regions are actually being selected within the composite race.

In the regions identified as ROH islands, 1272 genes were found (Table 2) and some QTLs already reported in the literature related to traits such as carcass quality, disease resistance and milk production and quality, summarized in Table 2.

Table 2

Regions of homozygosity observed in Montana Tropical® Composite animals and identification of the genes underlying the QTL in each region
Chr	Start regions (Mb)	End regions (Mb)	n Genes	Genes	QTL traits
4	62.31	73.79	1	ROH	Myristic acid content of milk, percentage of milk fat, marbling score, shear force, white line disease, amount of connective tissue, digital dermatitis, acetone content of milk
6	16.67	24.81	38	RPL34, LEF1, HADH, CYP2U1, SGMS2, TRNAC-GCA, TRNAG-CCC, PAPSS1, DKK2, GIMD1, AIMP1, TBCK, NPNT, GSTCD, INTS12, ARHGEF38, PPA2, TET2, CXXC4, TACR3, CENPE, BDH2, SLC9B2, NHEDC1, CISD2, UBE2D3, MANBA, NFKB1, SLC39A8, BANK1, PPP3CA, EMCN, DDIT4L, H2AFZ, DNAJB14, LAMTOR3, TRNAS-GGA, DAPP1	Marbling Score, Yield Away from Milk, Milk Protein Percentage, Milk Protein Yield, Shear Force, Connective Tissue Amount, Milk Yield
20	23	51	114	ANKRD55, IL6ST, MIR2359, IL31RA, DDX4, SLC38A9, PLPP1, MTREX, DHX29, CCNO, MCIDAS, CDC20B, MIR449C, MIR449B, MIR449A, GPX8, GZMA, GZMK, ESM1, SNX18, HSPB3, ARL15, NDUFS4, FST, MOCS2, ITGA2, ITGA1, PELO, ISL1, PARP8, EMB, HCN1, MRPS30, FGF10, NNT, PAIP1, C20H5orf34, TMEM267, CCL28, HMGCS1, TRH, NIM1K, ZNF131, SELENOP, CCDC152, GHR, FBXO4, C20H5orf51, OXCT1, TRNAY-GUA, PLCXD3, C6, MROH2B, C7, CARD6, RPL37, PRKAA1, TTC33, PTGER4, TRNAE-CUC, DAB2, C9, FYB1, RICTOR, OSMR, LIFR, EGFLAM, GDNF, WDR70, NUP155, MIR2360, CPLANE1, NIPBL, SLC1A3, RANBP3L, NADK2, SKP2, LMBRD2, UGT3A2, CAPSL, IL7R, SPEF2, PRLR, AGXT2, DNAJC21, BRIX1, RAD1, TTC23L, RAI14, C1QTNF3, AMACR, SLC45A2, RXFP3, ADAMTS12, TRNAT-UGU, TARS, NPR3, SUB1, TRNAG-CCC, ZFR, MTMR12, TRNAW-CCA, GOLPH3, PDZD2, C20H5orf22, DROSHA, CDH6, TRNAC-GCA, CDH9, TRNAC-ACA, TRNAS-GGA, TRNAF-AAA, CDH10	Milk production, milk fat percentage, milk protein percentage
26	36.41	42.75	50	GFRA1, CCDC172, PNLIPRP3, PNLIP, PNLIPRP2, C26H10orf82, HSPA12A, MIR2395, ENO4, SHTN1, VAX1, KCNK18, SLC18A2, PDZD8, EMX2, TRNAC-AC, RAB11FIP2, FAM204A, PRLHR, CACUL1, TRNAG-CCC, NANOS1, EIF3A, FAM45A, SFXN4, PRDX3, GRK5, TRNAC-GCA, RGS10, TIAL1, BAG3, INPP5F, MCMBP, SEC23IP, PLPP4, WDR11, FGFR2, ATE1, NSMCE4A, TACC2, MIR2396, BTBD16, PLEKHA1, HTRA1, DMBT1, SPADH2, SPADH1, C26H10orf120, FAM24A, C26H10orf88	Tetracosanoic acid content of milk, amount of connective tissue, susceptibility to bovine respiratory diseases, marbling score, amount of connective tissue

The population structure of the composite Montana Tropical® demonstrated in the analysis of principal components a genomic distance from the animals of more advanced generations. The results of the admixture analysis demonstrated that animals of equivalent generation greater than 4.5 presented a more homogeneous genomic composition profile when compared to the initial generations.

After the development of a composite race, at least 5 generations are required to form a new genomic profile in a composite of two races (Paim et al., 2020, Hay et al., 2022). The genomic structure of the Montana breed is much more complex due to the use of multiple breeds in its formation, and even so we observed this greater homogeneity and the formation of a specific cluster of the population close to the fifth generation.

In the composite Montana Tropical®, it was found that the prediction equations using PCA and ADMIXTURE results were not able to predict the components of the pedigree of the population, demonstrating that there is a difference in the genetic structure of the pedigree and genomic structure in the compound. The low ability to predict the racial composition of the pedigree in Montana Tropical® and the discrepancy found between the genomic and pedigree composition in other works with a composite of two breeds, may be due to the pedigree being incomplete or incorrect, or due to the effect of Mendelian sampling during recombination in meiosis (Kuehn et al., 2011, Gobena et al., 2018).

The adapted taurine component presents a much higher proportion in the more advanced generations in Montana, represented in the pedigree and by C1 and C2 in the ADMIXTURE analyzes that may be explaining the genomic composition of this biological group. And the Nellore and British components show a reduction in the pedigree proportion and in the genomic composition of C3 and C4. This is in line with the results described by Grigoletto et al., 2020, where they demonstrate a switch from the use of the Nelore zebu component to the adapted component, explained by a preference for meat quality characteristics that is provided by the use of Adapted taurine.

For individuals whose ancestors originate in different populations, their genetic makeup exhibits multiple ancestries associated with multiple genetic groupings or different populations that can be described by admixture models (Pritchard et al., 2000, Alexander et al., 2009). This has been successfully proven in identifying the racial composition in crossbreeding composite breeds with number of founders with 2 and 3 breeds, and the tools capable of estimating the differentiation of clusters according to the proportion of founders through regression (Akanno et al., 2017, Gobena et al., 2018, Hay et al., 2022), however Montana Tropical® presents a history of use with a large number of breeds in its formation, represented by the NABC biological groups, making the determination of the racial group that the cluster represents more complex.

The use of relationship matrices based on SNPs and population structure in the estimation of genomic genetic values makes it possible to improve the results of the evaluations, as they manage to capture the differences of Mendelian sampling and the relationships that are more distant than the pedigree (Hayes et al., 2009). Thus, considering the use of population structure information using analyzes such as PCA can provide improvements in genetic value estimates in the composite Montana Tropical®.

The results of the levels of homozygosity found in the Montana Tropical® population showed a slightly higher ROH value than those reported in another ROH study in a composite Montana Tropical® (Peripolli et al., 2020).

However, the levels found in the current study are still considered low (Purfield et al., 2012). As expected, FPED (0.8%) underestimated the inbreeding results of the population when compared to FROH (2%), this is probably due to flaws in the use of information from the pedigree, confirming the importance of using genomic data to better understand the real population inbreeding scenario (Caivio-Nasner et al., 2021). The FROH is a Genomic method that provides one of the most reliable measures of the true value of inbreeding in animal populations as it considers the past and recent relationship of individuals being a good source of information, and is indicated for calculating inbreeding in composite breed populations (Stella et al., 2010, Reverter et al., 2017).

The Montana Tropical selection index is composed of 70% of growth traits, with 20% for weaning weight (Grigoletto et al., 2019). Selection for weaning weight may have unintended consequences. Selection for weaning weight can have unintended consequences. Although cow milk production is phenotypically significantly correlated with calf weaning weight at different weaning ages (MacNeil & Mott, 2006), genetic between both traits can be negative (McHugh et al., 2014). Despite this possibility, Montana Tropical® presented selection signatures with QTLs for milk production.

The selection signature regions have large amounts of genes, which can be explored more deeply regarding their physiological and structural functions. Like, for example, the GHR gene found in the signature region of chromosome 20. This gene has a pleiotropic effect that affects protein production and milk quantity in dairy cattle and growth in beef cattle (Yurchenko, et al., 2018).

The genomic population structure of Montana Tropical ® bears little relation to racial composition based on the NABC system. The levels of inbreeding within the breed are low, but demonstrate an evolution over the generations. The detections of selection signatures with QTLs related to meat production demonstrate that the objective of producing a composite breed animal focused on the production of beef on pasture beef is being achieved.

Animals

Pedigree data and imputed genotypes (51962 SNP) of 3,215 animals from the composite Montana Tropical® were used. The samples came from farms in Brazil (Midwest, South and Southeast) and Uruguay and were supplied by the Animal Improvement and Biotechnology Group of the Faculty of Animal Science and Food Engineering (Pirassununga, São Paulo, Brazil).

Quality control of genotypes was performed using Plink software version 1.09 (https://www.cog-genomics.org/plink/1.9/) (Purcell et al., 2007). After quality control with 0.5 filtering for GC score (Gen Call), sample call rate (> 90%), SNP call rate (> 90%), MAF (> 1%) and LD pruning (-- indep-pairwise 50 5 0.5) the set retained for data analysis was 40,634 SNP on 29 autosomal chromosomes.

Pedigree Rating

The pedigree file was evaluated using the optiSel package (Wellmann, 2017) in the R 3.4.2 software (R Core Team, 2017). Animals that presented racial composition 1 for any biological group on the N:A:B:C scale were considered ancestors.

The number of equivalent generations was calculated using the equation $g=\sum (1/2)\text{n}$ where g is the number of equivalent generations and n the number of generations separating the individual from each known ancestor Welsh et al (2010).

PCA

Plink 1.9 software was used to perform Principal Component Analysis (PCA) based on the standardized variance relationship matrix to verify stratification and genetic distance between individuals in the population (Purcell et al., 2007). Analyzes were performed with up to 10 principal components. The individual eigenvectors were plotted along the different axes of a graph. The total variance explained by the main principal components was also estimated.

Admixture

Cluster analysis based on genotyping data was performed using the unsupervised maximum likelihood method implemented in the ADMIXTURE program version 1.3.0 (Alexander et al., 2009). K values (parameter that describes the number of subpopulations that make up the total population in a dataset provided as input) of 2, 3, and 4 were used for the dataset to establish the clustering patterns in the population (Ahmad et al., 2020). When trying to find the ideal k value for the population, K from 2 to 20 was used, but a point was not found where the CV was adequate to determine it, as the decay was continuous without a point increasing.

The outputs from K = 2 to 4 were plotted using the packages ggplot2 (Wickham, 2016), forcats (Wickham, 2021), ggthemes (Arnold, 2021) and patchwork (Pedersen, 2020) in the R software and the data organized by classes of equivalent generations (< 2.5, 2.5 < equiGen < 4.5 and > 4.5).

Correlation, regression and graph analyzes between PCA results, Equivalent Generations, ADMIXTURE and NABC composition of the pedigree were performed using the R software packages Hmisc (Frank & Harrell, 2022), MASS (Venables & Ripley, 2002), ggplot2 (Wickham, 2016) and corrplot (Wei & Simko, 2019) respectively.

Runs of homozygosity

ROH islands were identified using the Plink 1.9 software from the -- homozyg function, obeying the following criteria (Saravanan et al., 2020):

a sliding window of 50 SNPs across the genome (--homozyg-windows-snp);

a homozygous window overlap ratio of 0.05(--homozyg-windows-threshold);

a minimum length to consider a homozygous segment of 1000 kb (--homozyg-kb);

a maximum gap between consecutive homozygous SNPs of 1000 kb (--homozyg-gap);

the density of a SNP per 100 kb (--homozyg-density)

SAW. requirement to contain at least 20 SNPs (--homozyg-snp);

VII. requirement not to allow missing genotypes and not to allow heterozygotes Caivio-Nasner et al., 2021);

The estimate of genomic inbreeding was calculated as follows:

$$FROH=\sum \frac{LROH}{LAUTO}$$

where FROH is the measure of homozygosity per animal, LROH is the sum of ROH per animal and LAUTO the total length (2715853792) of the genome with reference ARS 1.2.

The incidence of common ROH was transformed to generation class frequency by dividing by the number of animals from each generation class in the analysis. Normality tests were performed and the frequency threshold that defines the top 1% of observations for each equivalent generation class was determined. Homozygous regions above the frequency threshold of each pedigree class of equivalent generation (5% for all generations, 4.3% for equiGen < 2.5, 4.9% for 2.5 < equiGen < 4.5 and 12% for equiGen > 4.5) were considered as regions selected.

According to the length of the ROH, it is possible to estimate the number of generations traced back to the common ancestor, which generates homozygosity in that region. ROH were classified into 4 categories (1 = more than 6 generations, 2 = between 2.5 and 4.5, 3 = between 4.5 and 2.5 and 4 = less than 2.5 generations) using the equation proposed by Curik et al. (2014): E (LIBD − H | gcA ) = 100/(2 gcA ), where E (LIBD–H | gcA) is the expected length of a haplotype identical per offspring (IBD) (in centiMorgans – cM), and gcA is the number of generations of the common ancestor. The conversion of the recombination rate metric to physical distance (from cM to Mb) was performed using the average of the results from Arias et al. (2009) and Weng et al. (2014). Based on Curik et al. (2014), for example, a ROH greater than 15.6 Mb probably originated from a common ancestor less than 2.5 generations ago.

Genes in each selected region were searched in Genome Data Viewer (NCBI platform) using ARS-UCD1.3 (GCF_002263795.2). To identify traits related to genes located in each significant genomic region, a search was performed on the Animal Genome QTL Database (htt ps://www.animalgenome.org/cgi-bin/QTLdb/EC/index).

A Adapted taurine breeds (Bos taurus taurus)

B British breeds (Bos taurus taurus)

C Mainland European breeds (Bos taurus taurus)

F Inbreeding coefficient

IBD Identical by descent

LD Linkage desequilibrium

MAF Minor allele frequencie

N Zebu breeds (Bos taurus indicus)

PCA Principal components analysis

QTL Quantitative trait loci

ROH Runs of homozygosity

SNP Single nucleotide polymorphismis

Ethics approval and consent to participate

Not applicable

Consent for publication

Not applicable

Availability of data and materials

Datasets generated and/or analyzed during the current study are not publicly available due to belongs to the Montana Tropical Breeders Association, but are available from the corresponding author upon reasonable request.

Competing interests

The authors declare that they have no competing interests.

Funding

CNPq and CAPES

Acknowledgements

Animal Improvement and Biotechnology Group of the Faculty of Animal Science and Food Engineering (Pirassununga, São Paulo, Brazil)

Authors’ contribuition

Camila Alves dos Santos - performed the analyzes, interpretation of the results, wrote the article

José Bento Sterman Ferraz - provision of data, review of results

Tiago Paim – Design of the work, interpretation of the results, reviewer of the article

Rafael Espigolan - provision of data, review of results

Elisangela Chicaroni de Mattos Oliveira - provision of data

Joanir Pereira Eler - provision of data

Gabriela Giacomini - provision of data

Ahmad SF, Panigrahi M, Chhotaray S, Pal D, Parida S, Bhushan B, …, Singh RK. 2020. Revelation of genomic breed composition in a crossbred cattle of India with the help of Bovine50K BeadChip.Genomics,112(2),1531–1535. https://doi.org/10.1016/j.ygeno.2019.08.025
Alexander DH, Novembre J, Lange K. 2009. Fast model-based estimation of ancestry in unrelated individuals. Genome Research 19: 1655–64. http://doi10.1101/gr.094052.109
Arias JA, Keehan M, Fisher P, Coppieters W, Spelman R. A high density linkage map of the bovine genome. BMC Genet. 2009;10(1):1–12. https://doi.org/10.1186/1471-2156-10-18.
Caivio-Nasner S, López‐Herrera A, González‐Herrera LG, Rincón JC. Diversity analysis, runs of homozygosity and genomic inbreeding reveal recent selection in Blanco Orejinegro cattle. J Anim Breed Genet. 2021;138(5):613–27.
Curik I, Ferenčaković M, Sölkner J. 2014. Inbreeding and runs of homozygosity: a possible solution to an old problem.Livestock Science,166,26–34. https://doi.org/10.1016/j.livsci.2014.05.034
Dodds KG, Auvray B, Newman SAN, McEwan JC. Genomic breed prediction in New Zealand sheep. BMC Genet. 2014;15(1):1–15. https://doi.org/10.1186/s12863-014-0092-9.
Ferencakovic M, Hamzic E, Gredler B, Curik I, Sölkner J. Runs of homozygosity reveal genome-wide autozygosity in the Austrian Fleckvieh cattle. Agriculturae Conspectus Scientificus. 2011;76(4):325–9.
Frank & Harrell Jr. 2022. Hmisc: Harrell Miscellaneous. R package version 4.7-0
Funkhouser SA, Bates RO, Ernst CW, Newcom D, Steibel JP. Estimation of genome-wide and locus-specific breed composition in pigs. Translational Anim Sci. 2017;1(1):36–44. https://doi.org/10.2527/tas2016.0003.
Gurgul A, Semik E, Pawlina K, Szmatoła T, Jasielczuk I, Bugno-Poniewierska M. The application of genome-wide SNP genotyping methods in studies on livestock genomes. J Appl Genet. 2014;55(2):197–208. https://doi.org/10.1007/s13353-014-0202-4.
Hay EH, Toghiani S, Roberts AJ, Paim T, Kuehn LA, Blackburn HD. Genetic architecture of a composite beef cattle population. J Anim Sci. 2022;100(9):kac230. https://doi.org/10.1093/jas/skac230.
Hayes BJ, Chamberlain AJ, Maceachern S, Savin K, McPartlan H, MacLeod I, Sethuraman L, Goddard ME. A genome map of divergent artificial selection between Bos taurus dairy cattle and Bos taurus beef cattle. Anim Genet. 2009;40(2):176–84. https://doi.org/10.1111/j.1365-2052.2008.01815.x.
Huang Y, Bates RO, Ernst CW, Fix JS, Steibel JP. Estimation of US Yorkshire breed composition using genomic data. J Anim Sci. 2014;92(4):1395–404. https://doi.org/10.2527/jas.2013-6907.
Kuehn LA, Keele JW, Bennett GL, McDaneld TG, Smith TPL, Snelling WM, …, Thallman RM. Predicting breed composition using breed frequencies of 50,000 markers from the US Meat Animal Research Center 2,000 Bull Project. J Anim Sci. 2011;89(6):1742–50. https://doi.org/10.2527/jas.2010-3530.
McQuillan R, et al. Runs of Homozygosity in European Populations. Am J Hum Genet. 2008;83(3):359–72. https://doi.org/10.1016/j.ajhg.2008.08.007. http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=2556426&. tool=pmcentrez&r endertype = abstract
Paim TDP, Hay EHA, Wilson C, Thomas MG, Kuehn LA, Paiva SR, …, Blackburn H. (2020). Genomic breed composition of selection signatures in Brangus beef cattle. Frontiers in genetics,11, 710.
Pedersen TL. 2020. patchwork: The Composer of Plots. R package version 1.1.1
Peripolli E, Stafuzza NB, Amorim ST, de Lemos MVA, Grigoletto L, Kluska S, …, Baldi F. Genome-wide scan for runs of homozygosity in the composite Montana Tropical® beef cattle. J Anim Breed Genet. 2020;137(2):155–65. https://doi.org/10.1111/jbg.12428.
Purfield DC, Berry DP, MCparland S, Bradley DG. Runs of homozygosity and population history in cattle. BMC Genet. 2012;13:70. https://doi.org/10.1186/1471-2156-13-70.
Reist-Marti SB, Simianer H, Gibson J, Hanotte O, Rege JEO. Weitzman's approach and conservation of breed diversity: an application to African cattle breeds. Conserv Biol. 2003;17(5):1299–311. https://doi.org/10.1046/j.1523-1739.2003.01587.x.
Reverter A, Porto-Neto LR, Fortes MRS, Kasarapu P, De Cara MAR, Burrow HM, Lehnert SA. Genomic inbreeding depression for climatic adaptation of tropical beef cattle. J Anim Sci. 2017;95(9):3809–21. https://doi.org/10.2527/jas.2017.1643.
Ron M, Blanc Y, Band M, Ezra E, Weller JI. Misidentification rate in the Israeli dairy cattle population and its implications for genetic improvement. J Dairy Sci. 1996;79(4):676–81. https://doi.org/10.3168/jds.S0022-0302(96)76413-5.
Saravanan KA, Panigrahi M, Kumar H, Parida S, Bhushan B, Gaur GK, …, Singh RK. Genome-wide assessment of genetic diversity, linkage disequilibrium and haplotype block structure in Tharparkar cattle breed of India. Animal Biotechnol. 2022;33(2):297–311. doi.org/10.1080/10495398.2020.1796696.
Silva AMD, Alencar MMD, Regitano LCDA, Oliveira MC. D. S., & Barioni Júnior, W. 2007 Artificial infestation of Boophilus microplus in beef cattle heifers of four genetic groups.Genetics and Molecular Biology, 30(4),1150–1155. https://doi.org/10.1590/S1415-47572007000600020
Stella A, Ajmone-Marsan P, Lazzari B, Boettcher P. Identification of selection signatures in cattle breeds selected for dairy production. Genetics. 2010;185(4):1451–61. https://doi.org/10.1534/genet ics.110.116111.
Venables WN, Ripley BD. 2002. Modern Applied Statistics with S. Fourth Edition. Springer, New York. ISBN 0-387-95457-0
Wei T, Simko V. 2021. R package 'corrplot':Visualization of a Correlation Matrix
Welsh CS, Stewart TS, Schwab C, Blackburn HD. Pedigree analysis of 5 swine breeds in the United States and the implications for genetic conservation. J Anim Sci. 2010;88(5):1610–8. https://doi.org/10.2527/jas.2009-2537.
Wickham H. ggplot2: Elegant Graphics for Data Analysis. New York: Springer-Verlag; 2016.
Wickham H. 2021. forcats: Tools for Working with Categorical Variables (Factors). R package version 0.5.1.

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

Population structure and selection signatures over generations in the composite Montana tropical ®

Status:

Version 1

Abstract

Figures

Background

Results

Population structure

Runs of Homozygosity

Discussion

Conclusion

Methods

Animals

Pedigree Rating

PCA

Admixture

Runs of homozygosity

Abbreviations

Declarations

References

Additional Declarations

Status:

Version 1