Search

Progress made in applying agent systems to molecular computational biology is reviewed and strategies by which to exploit agent technology to greater advantage are investigated. Communities of software agents could play an important role in helping genome scientists design reagents for future research. The advent of genome sequencing in cattle and swine increases the complexity of data analysis required to conduct research in livestock genomics. Databases are always expanding and semantic differences among data are common. Agent platforms have been developed to deal with generic issues such as agent communication, life cycle management and advertisement of services (white and yellow pages). This frees computational biologists from the drudgery of having to re-invent the wheel on these common chores, giving them more time to focus on biology and bioinformatics. Agent platforms that comply with the Foundation for Intelligent Physical Agents (FIPA) standards are able to interoperate. In other words, agents developed on different platforms can communicate and cooperate with one another if domain-specific higher-level communication protocol details are agreed upon between different agent developers. Many software agent platforms are peer-to-peer, which means that even if some of the agents and data repositories are temporarily unavailable, a subset of the goals of the system can still be met. Past use of software agents in bioinformatics indicates that an agent approach should prove fruitful. Examination of current problems in bioinformatics indicates that existing agent platforms should be adaptable to novel situations.

Background: Existing linkage maps of the bovine genome primarily contain anonymous microsatellite markers. These maps have proved valuable for mapping quantitative trait loci (QTL) to broad regions of the genome, but more closely spaced markers are needed to fine-map QTL, and markers associated with genes and annotated sequence are needed to identify genes and sequence variation that may explain QTL., Results: Bovine expressed sequence tag (EST) and bacterial artificial chromosome (BAC)sequence data were used to develop 918 single nucleotide polymorphism (SNP) markers to map genes on the bovine linkage map. DNA of sires from the MARC reference population was used to detect SNPs, and progeny and mates of heterozygous sires were genotyped. Chromosome assignments for 861 SNPs were determined by twopoint analysis, and positions for 735 SNPs were established by multipoint analyses. Linkage maps of bovine autosomes with these SNPs represent 4585 markers in 2475 positions spanning 3058 cM. Markers include 3612 microsatellites, 913 SNPs and 60 other markers. Mean separation between marker positions is 1.2 cM. New SNP markers appear in 511 positions, with mean separation of 4.7 cM. Multi-allelic markers, mostly microsatellites, had a mean (maximum) of 216 (366) informative meioses, and a mean 3-lod confidence interval of 3.6 cM Bi-allelic markers, including SNP and other marker types, had a mean (maximum) of 55 (191) informative meioses, and were placed within a mean 8.5 cM 3-lod confidence interval. Homologous human sequences were identified for 1159 markers, including 582 newly developed and mapped SNP., Conclusion: Addition of these EST- and BAC-based SNPs to the bovine linkage map not only increases marker density, but provides connections to gene-rich physical maps, including annotated human sequence. The map provides a resource for fine-mapping quantitative trait loci and identification of positional candidate genes, and can be integrated with other data to guide and refine assembly of bovine genome sequence. Even after the bovine genome is completely sequenced, the map will continue to be a useful tool to link observable phenotypes and animal genotypes to underlying genes and molecular mechanisms influencing economically important beef and dairy traits.

This study describes the physical and linkage mapping of 42 gene-associated markers developed from mammary gland-derived expressed sequence tags to the cattle genome. Of the markers, 25 were placed on the USDA reference linkage map and 37 were positioned on the Roslin 3000-rad radiation hybrid (RH) map, with 20 assignments shared between the maps. Although no novel regions of conserved synteny between the cattle and the human genomes were identified, the coverage was extended for bovine chromosomes 3, 7, 15, and 29 compared with previously published comparative maps between human and bovine genomes. Overall, these data improve the resolution of the human-bovine comparative maps and will assist future efforts to integrate bovine RH and linkage map data.

A system to use bovine EST data in conjunction with human genomic sequence to improve the bovine linkage map over the entire genome or on specific chromosomes was evaluated. Bovine EST sequence was used to provide primer sequences corresponding to bovine genes, while human genomic sequence directed primer design to flank introns and produce amplicons of appropriate size for efficient direct sequencing. The sequence tagged sites (STS) produced in this way from the four sires of the MARC reference families were examined for single nucleotide polymorphisms (SNPs) that could be used to map the corresponding genes. With this approach, along with a primer/extension mass spectrometry SNP genotyping assay, 100 ESTs were placed on the bovine genetic linkage map. The first 70 were chosen at random from bovine EST-human genomic comparisons. An additional 30 ESTs were successfully mapped to bovine Chromosome 19 (BTA19), and comparison of the resulting BTA19 map to the position of the corresponding human orthologs on the HSA17 draft sequences revealed differences in the spacing and order of genes. Over 80% of successful amplicons contained SNPs, indicating that this is an efficient approach to generating EST-associated genetic markers. We have demonstrated the feasibility of constructing a linkage map based on SNPs associated with ESTs and the plausibility of utilizing EST, comparative mapping information, and human sequence data to target regions of the bovine genome for SNP marker development.

The starting point of the present study was the reported identification of a chromosomal region on bovine Chromosome (Chr) 15 (BTA15) carrying loci affecting meat tenderness. A comparative linkage map of BTA15 and human Chr 11 (HSA11) was constructed to identify potential positional candidate genes and to provide a resource of genetic markers to support marker-assisted selection (MAS). Relative rearrangements between the bovine and human genomes for these chromosomes are the most complex observed in comparative mapping between the two species, with nine alternating blocks of conserved synteny between HSA11 and bovine Chrs 15 and 29. The results of this study were the addition of nine genes to the HSA11/BTA15 comparative linkage map, and development of five microsatellite markers within the quantitative trait locus (QTL) interval. One gene with known effects on muscle development (MYOD1) was mapped to the interval. A second gene (CALCA) involved in regulation of calcium levels, a key factor in postmortem tenderization, also mapped within the interval. Refinement of the comparative map and QTL position will reduce the interval on the human transcription map to be scanned in search of candidates, reducing the effort and resources required to identify the allelic variation responsible for the genetic effect.

The goal of this study was to identify and map genes expressed during the elongation phase of embryogenesis in swine. Expressed sequence tags were analysed from a previously described porcine cDNA library prepared from elongating swine embryos. Average insert length of randomly selected clones was approximately 600 bp, with a range from < 100 to > 2500 bp. Single-pass, coding strand sequences from 1132 independent clones were compared with the GenBank non-redundant (nr) database via BLASTN analysis to identify potential porcine homologous of known genes. Among these sequences, 781 (69%) showed significant (score > 300) homology to non- mitochondrial sequences previously deposited in GenBank. Sequences matching interleucin 1 beta and thymosin beta 10 were most frequently observed (24 and 18 clones, respectively), in addition to matches with 310 other distinct genes. No significant match in the GenBank nr database was obtained for 303 sequences. Analysis demonstrated that 151 (50%) had open reading frames (ORF) extending at least 50 codons from the first base of the clone insert. Genetic markers were developed and used to map a subset of 17 genes, selected on the basis of function or of the ability to design primers that successfully amplified porcine genomic DNA, to 10 different porcine chromosomes, providing a set of mapped markers corresponding to genes expressed during conceptus elongation.

Micromolar calcium activated neural protease (CAPN1) was investigated as a potential candidate gene for a quantitative trait locus (QTL) on BTA29 affecting meat tenderness. A 2,948-bp bovine cDNA containing the entire coding region of the gene was obtained, showing 91% identity to human CAPN1. The 716 AA protein predicted from this sequence shows 97% similarity (95% identity) to the 714 AA human protein. Analysis of the gene structure revealed that CAPN1 mRNA is encoded by at least 19 exons, and 11,055 bp of the gene were sequenced, including 17 introns. Two single nucleotide polymorphisms (SNP) were detected in intron 12 and were used to map bovine CAPN1 to the telomeric end of the BTA29 linkage group. This approximately coincides with the position of the QTL, demonstrating that CAPN1 protease is a positional candidate gene potentially affecting variation in meat tenderness in a bovine resource mapping population.

Polymorphic markers at bovine gene loci facilitate the integration of cattle genetic maps with those of humans and mice. To this end, 31 single nucleotide polymorphism (SNP) markers were developed for seven bovine chemokine genes. Loci were amplified from bovine genomic DNA by the polymerase chain reaction, and candidate amplicons were sequenced to determine their identity. Amplified loci from 24 founding parents and select progeny from a beef cattle reference population were sequenced and analyzed for SNPs. SNP haplotype alleles were determined by examining segregation patterns and used to establish the locus position on the bovine linkage map. Loci for growth-related proteins (GRO3, GRO1, and GROX) were clustered with the related CXC chemokine genes, interleukin (IL) 8, and epithelial cell inflammatory protein 1, at 84 cM from the centromeric end of the bovine chromosome (BTA) 6 linkage group. Bovine loci for a cluster of IL8 receptors, a stromal cell-derived factor 1, interferon-gamma, and tumor necrosis factor-alpha were mapped at 90, 55, 59, and 34 cM, respectively, from the centromeric ends of the BTA 2, 28, 5, and 23 linkage groups. The positions of these bovine loci were compared with those of orthologous loci on the human map to refine the boundaries of conserved synteny. These seven loci provide examples of SNP development in which the efficiency was largely dependent on the availability of bovine genomic or cDNA sequence. The polymorphic nature of these SNP haplotype markers suggests that they will be useful for mapping complex traits in cattle, such as resistance to infectious disease.

The locus responsible for the appearance of muscular hypertrophy (mh) in double muscled cattle breeds has recently been shown to encode a secreted growth factor designated myostatin (MSTN). This conclusion was based in part on the placement of MSTN in the interval to which mh had been mapped on bovine chromosome 2 (BTA2). During the mapping phase of the study, numerous yeast artificial chromosome (YAC) clones were isolated that contained genetic markers closely linked to mh. Other YACs and cosmids were identified that contained genes selected from human chromosome 2q (HSA2q), with the goal of defining the position of breakpoints in conserved synteny between the bovine and human comparative maps, thereby permitting accurate selection of positional candidate genes. An efficient subcloning procedure was developed to obtain microsatellites (ms) from YAC clones, to increase the number of informative meioses in herds segregating for mh. The same procedure was used to place the human orthologues of engrailed-1 (EN1), interleukin 1 beta (IL1B), and paired-box-containing 8 (PAX8) genes on the cattle map to further define the positions of breakpoints in conserved synteny and gene order. Twenty-three of 28 ms identified from YAC subclone libraries were informative in the mapping families. Seven mapped to the centromeric end of BTA2, which contains the mh locus, improving marker density and informativeness. The two MSTN and four EN1 gene-associated ms markers developed from YACs, map to positions 1.5 and 61.6 CM in the BTA2 linkage group, respectively. In addition, ms markers developed from cosmids containing either IL1B or PAX8, map to positions 56.6 and 56.9 CM in the BTA11 linkage group, respectively. These linkage data confirm the location and orientation of orthologous segments of HSA2q that were previously indistinguishable on the bovine map, and demonstrates the presence of microrearrangements of gene order (segments < 10 CM) and conserved synteny between the human and bovine genomes.

Genotypic and phenotypic data were collected to estimate chromosomal position of the callipyge (CLPG) gene and to test gene action. Nine Dorset rams of extreme muscling phenotype and 114 Romanov ewes composed the grandparent generation of a resource flock of 362 F2 lambs segregating at the CLPG locus. The parent generation consisted of eight F1 sires and 138 F1 dams. The F2 lambs were serially slaughtered in six groups at 3-wk intervals starting at 23 wk of age to allow comparisons at different end points. A linkage group of 25 marker loci (mean of 708 informative meioses per marker) spanning 87.2 cM was developed and improved the previous known coverage and precision of marker order and interval distance from available maps of ovine chromosome 18. Probabilities of each CLPG genotype were calculated at 1-cM intervals (0 to 107 cM). Statistical models included effects of year, sex, sire, regressions on genotypic probabilities, and genotype-specific linear and quadratic regressions on appropriate covariates. Orthogonal contrasts of CLPG genotypic effects evaluated additive, maternal dominance, and paternally derived polar overdominance models of gene action. The most parsimonious model did not include the additive and maternal dominance genetic contrasts. From analyses of four key traits, a consensus for position of CLPG was obtained at 86 cM relative to the most centromeric marker. An F-test with 3 df representing polar overdominance was maximum at position 86 cM (F = 407.4; P < .00001) with leg score as the dependent variable. These results are consistent with assignment of the CLPG locus to the telomeric region of chromosome 18 and support the polar overdominance model of gene action proposed by Cockett et al. (1996). Furthermore, recombinant individuals with definitive phenotypes confined the position of CLPG to a 3.9-cM interval, facilitating positional cloning experiments.

A report of the first workshop on the genetic map of bovine chromosome 1 (BTA1) is presented. Five laboratories contributed 31,962 informative meioses from 70 loci. Thirty-two loci which had been typed by at least two laboratories were used to construct a framework genetic map with a likelihood ratio support of at least 1000:1 for locus order. The resulting sex-averaged framework map contained 26 loci and spanned 163.6 CM. The lengths of the female and male maps were 159.5 CM and 165.3 CM, respectively, and there was evidence for an expansion in the telomeric one-third of the male map. Of the four cases where order for closely linked loci differed among the maps produced for each of the contributing laboratories, a consensus order was obtained for three in the framework map. The average genetic distance between framework loci on the sex-averaged map was 6.3 CM.

A genetic map of Ovis aries (haploid n = 27) was developed with 519 markers (504 microsatellites) spanning approximately 3063 cM in 26 autosomal linkage groups and 127 cM (female specific) of the X Chromosome (Chr). Genotypic data were merged from the IMF flock (Crawford et al., Genetics 140, 703, 1995) and the USDA mapping flock. Seventy-three percent (370/504) of the microsatellite markers on the map are common to the USDA-ARS MARC cattle linkage map, with 27 of the common markers derived from sheep. The number of common markers per homologous linkage group ranges from 5 to 22 and spans a total of 2866 cM (sex average) in sheep and 2817 cM in cattle. Marker order within a linkage group was consistent between the two species with limited exceptions. The reported translocation between the telomeric end of bovine Chr 9 (BTA 9) and BTA 14 to form ovine Chr 9 is represented by a 15-cM region containing 5 common markers. The significant genomic conservation of marker order will allow use of linkage maps in both species to facilitate the search for quantitative trait loci (QTLs) in cattle and sheep.

A locus near the centromere of bovine chromosome 2 is responsible for muscle hypertrophy (mh) in cattle. The objectives of this study were to refine the genomic region in which the locus resides and to assess the effects of a single copy of the mh allele on carcass and birth traits. Two half-sib families were developed using a Belgian Blue x MARC III (n = 246) or a Piedmontese x Angus (n = 209) sire. Traits analyzed were calving ease (CE), birth weight (BWT, kg), longissimus rib eye area (REA), retail product yield (RPYD), USDA yield grade (YG), marbling (MAR), fat thickness (FAT), estimated kidney, pelvic, and heart fat (KPH), and longissimus tenderness measured as Warner-Bratzler shear force at 3 (S3) and 14 (S14) d postmortem. Six microsatellites were used to determine the presence or absence of the mh allele and to confirm the location of the locus affecting the traits, which was assessed to be 4 cM from the beginning of the linkage group, with the 95% confidence interval between 2 and 6 cM. Cattle with an mh allele had increased (P < .01) REA, RPYD, and BWT and decreased MAR, YG, FAT, and KPH, compared with those without the allele. The effects of the mh allele (mh/+ vs +/+) were 1.35, 1.6, .41, -1.01, -1.42, -.84, and -.86 residual standard deviations, respectively. There were no effects (P > .10) for CE, S3, and S14. Allelic differences due to the mh locus were similar for both sources (Belgian Blue or Piedmontese). Individuals inheriting a single mh allele had a leaner, more heavily muscled carcass compared with those inheriting the alternative allele. Thus, mating schemes that maximize production of mh/+ genotypes provide a viable approach for improving carcass composition.

We report a bovine linkage map constructed with 1236 polymorphic DNA markers and 14 erythrocyte antigens and serum proteins. The 2990-cM map consists of a sex-specific, X chromosome linkage group and 29 sex-averaged, autosomal linkage groups with an average interval size of 2.5 cM. The map contains 627 new markers and 623 previously linked markers, providing a basis for integrating the four published bovine maps. Orientation and chromosomal assignment of all the linkage groups, except BTA20 and BTA22, was provided by 88 markers that were assigned previously to chromosomes. This map provides sufficient marker density for genomic scans of populations segregating quantitative trait loci (QTL) and subsequent implementation of marker-assisted selection (MAS) mating schemes.

We report the highest density genetic linkage map for a livestock species produced to date. Three published maps for Sus scrofa were merged by genotyping virtually every publicly available microsatellite across a single reference population to yield 1042 linked loci, 536 of which are novel assignments, spanning 2286.2 cM (average interval 2.23 cM) in 19 linkage groups (18 autosomal and X chromosomes, n = 19). Linkage groups were constructed de novo and mapped by locus content to avoid propagation of errors in older genotypes. The physical and genetic maps were integrated with 123 informative loci assigned previously by fluorescence in situ hybridization (FISH). Fourteen linkage groups span the entire length of each chromosome. Coverage of chromosomes 11, 12, 15, and 18 will be evaluated as more markers are physically assigned. Marker-deficient regions were identified only on 11q1.7-qter and 14 cen-q1.2. Recombination rates (cM/Mbp) varied between and within chromosomes. Short chromosomal arms recombined at higher rates than long arms, and recombination was more frequent in telomeric regions than in pericentric regions. The high-resolution comprehensive map has the marker density needed to identify quantitative trait loci (QTL), implement marker-assisted selection or introgression and YAC contig construction or chromosomal microdissection.

We describe a conceptual model for genome databases that facilitates the process of building, maintaining, and disseminating physically anchored genetic linkage maps. The model has been implemented as a relational database at the Roman L. Hruska U.S. Meat Animal Research Center (MARC). Development of consensus maps using disparate data from different reference pedigrees or laboratories is supported. The model is of use to quantitative and population geneticists interested in loci that affect phenotypes and marker-assisted selection, and it is sufficiently flexible for centralized, species genome databases facilitating comparative mapping. The MARC genome database is used to assemble, maintain, and disseminate physically anchored genetic linkage maps for cattle, swine, and sheep currently based on more than 100,000 genotypes from 1,000 markers. Integrated with linkage analysis software, this database permits frequent updates of physically anchored genetic linkage maps.

We report the most extensive physically anchored linkage map for cattle produced to date. Three-hundred thirteen genetic markers ordered in 30 linkage groups, anchored to 24 autosomal chromosomes (n = 29), the X and Y chromosomes, four unanchored syntenic groups and two unassigned linkage groups spanning 2464 cM of the bovine genome are summarized. The map also assigns 19 type I loci to specific chromosomes and/or syntenic groups and four cosmid clones containing informative microsatellites to chromosomes 13, 25 and 29 anchoring syntenic groups U11, U7 and U8, respectively. This map provides the skeletal framework prerequisite to development of a comprehensive genetic map for cattle and analysis of economic trait loci (ETL).

We report the most extensive genetic linkage map for a livestock species produced to date. We have linked 376 microsatellite (MS) loci with seven restriction fragment length polymorphic loci in a backcross reference population. The 383 markers were placed into 24 linkage groups which span 1997 cM. Seven additional MS did not fall into a linkage group. Linkage groups are assigned to 13 autosomes and the X chromosome (haploid n = 19). This map provides the basis for genetic analysis of quantitative inheritance of phenotypic and physiologic traits in swine.