Search

Abstract Background Previous fine mapping studies in Norwegian Red cattle (NRC) in the region 86-90.4 Mb on Bos taurus chromosome 6 (BTA6) has revealed a quantitative trait locus (QTL) for protein yield (PY) around 88 Mb and a QTL for clinical mastitis (CM) around 90 Mb. The close proximity of these QTLs may partly explain the unfavorable genetic correlation between these two traits in NRC. A long range haplotype covering this region was introduced into the NRC population through the importation of a Holstein-Friesian bull (1606 Frasse) from Sweden in the 1970s. It has been suggested that this haplotype has a favorable effect on milk protein content but an unfavorable effect on mastitis susceptibility. Selective breeding for milk production traits is likely to have increased the frequency of this haplotype in the NRC population. Results Association mapping for PY and CM in NRC was performed using genotypes from 556 SNPs throughout the region 86-97 Mb on BTA6 and daughter-yield-deviations (DYDs) from 2601 bulls made available from the Norwegian dairy herd recording system. Highest test scores for PY were found for single-nucleotide polymorphisms (SNPs) within and surrounding the genes CSN2 and CSN1S2, coding for the β-casein and αS2-casein proteins. High coverage re-sequencing by high throughput sequencing technology enabled molecular characterization of a long range haplotype from 1606 Frasse encompassing these two genes. Haplotype analysis of a large number of descendants from this bull indicated that the haplotype was not markedly disrupted by recombination in this region. The haplotype was associated with both increased milk protein content and increased susceptibility to mastitis, which might explain parts of the observed genetic correlation between PY and CM in NRC. Plausible causal polymorphisms affecting PY were detected in the promoter region and in the 5'-flanking UTR of CSN1S2. These polymorphisms could affect transcription or translation of CSN1S2 and thereby affect the amount of αS2-casein in milk. Highest test scores for CM were found in the region 89-91 Mb on BTA6, very close to a cluster of genes coding for CXC chemokines. Expression levels of some of these CXC chemokines have previously been shown to increase in bovine mammary gland cell lines after exposure to bacterial cell wall components. Conclusion Molecular characterization of the long range haplotype from the Holstein-Friesian bull 1606 Frasse, imported into NRC in the 1970s, revealed polymorphisms that could affect transcription or translation of the casein gene CSN1S2. Sires with this haplotype had daughters with significantly elevated milk protein content and selection for milk production traits is likely to have increased the frequency of this haplotype in the NRC population. The haplotype was also associated with increased mastitis susceptibility, which might explain parts of the genetic correlation between PY and CM in NRC.

Abstract Background Comparison of recent patterns of recombination derived from linkage maps to historical patterns of recombination from linkage disequilibrium (LD) could help identify genomic regions affected by strong artificial selection, appearing as reduced recent recombination. Norwegian Red cattle (NRF) make an interesting case study for investigating these patterns as it is an admixed breed with an extensively recorded pedigree. NRF have been under strong artificial selection for traits such as milk and meat production, fertility and health. While measures of LD is also crucial for determining the number of markers required for association mapping studies, estimates of recombination rate can be used to assess quality of genomic assemblies. Results A dataset containing more than 17,000 genome-wide distributed SNPs and 2600 animals was used to assess recombination rates and LD in NRF. Although low LD measured by r2 was observed in NRF relative to some of the breeds from which this breed originates, reports from breeds other than those assessed in this study have described more rapid decline in r2 at short distances than what was found in NRF. Rate of decline in r2 for NRF suggested that to obtain an expected r2 between markers and a causal polymorphism of at least 0.5 for genome-wide association studies, approximately one SNP every 15 kb or a total of 200,000 SNPs would be required. For well known quantitative trait loci (QTLs) for milk production traits on Bos Taurus chromosomes 1, 6 and 20, map length based on historic recombination was greater than map length based on recent recombination in NRF. Further, positions for 130 previously unpositioned contigs from assembly of the bovine genome sequence (Btau_4.0) found using comparative sequence analysis were validated by linkage analysis, and 28% of these positions corresponded to extreme values of population recombination rate. Conclusion While LD is reduced in NRF compared to some of the breeds from which this admixed breed originated, it is elevated over short distances compared to some other cattle breeds. Genomic regions in NRF where map length based on historic recombination was greater than map length based on recent recombination coincided with some well known QTL regions for milk production traits. Linkage analysis in combination with comparative sequence analysis and detection of regions with extreme values of population recombination rate proved to be valuable for detecting problematic regions in the Btau_4.0 genome assembly.

A novel approach for revealing patterns of proteome variation among series of 2-DE gel images is presented. The approach utilises image alignment to ensure that each pixel represents the same information across all gels. Gel images are normalised, and background corrected, followed by unfolding of the images to 1-D pixel vectors and analysing pixel vectors by multivariate data modelling. Information resulting from the data analysis is refolded back to the image domain for visualisation and interpretation. The method is rapid and suitable for automatic routines applied after the gel alignment. The approach is compared with spot volume analysis to illustrate how this approach can solve persistent problems like mismatch of protein spots, erroneous missing values and failure to detect variation in overlapping proteins. The method may also detect variation in the border area of saturated proteins. The approach is given the name pixel-based analysis of multiple images for the identification of changes (PMC). The method can be used for multiple images in general. Effects of pretreatment of the images are discussed.