A physical map of the bovine genome
- Warren M Snelling1Email author,
- Readman Chiu2,
- Jacqueline E Schein2,
- Matthew Hobbs3,
- Colette A Abbey4,
- David L Adelson4,
- Jan Aerts5,
- Gary L Bennett1,
- Ian E Bosdet2,
- Mekki Boussaha6,
- Rudiger Brauning7,
- Alexandre R Caetano8,
- Marcos M Costa8,
- Allan M Crawford7,
- Brian P Dalrymple9,
- André Eggen6,
- Annelie Everts-van der Wind10,
- Sandrine Floriot6,
- Mathieu Gautier6,
- Clare A Gill4,
- Ronnie D Green11,
- Robert Holt2,
- Oliver Jann5,
- Steven JM Jones2,
- Steven M Kappes11,
- John W Keele1,
- Pieter J de Jong12,
- Denis M Larkin10,
- Harris A Lewin10, 13,
- John C McEwan7,
- Stephanie McKay14,
- Marco A Marra2,
- Carrie A Mathewson2,
- Lakshmi K Matukumalli15,
- Stephen S Moore14,
- Brenda Murdoch14,
- Frank W Nicholas3,
- Kazutoyo Osoegawa12,
- Alice Roy16,
- Hanni Salih4,
- Laurent Schibler6,
- Robert D Schnabel17,
- Licia Silveri18,
- Loren C Skow4,
- Timothy PL Smith1,
- Tad S Sonstegard15,
- Jeremy F Taylor17,
- Ross Tellam9,
- Curtis P Van Tassell15,
- John L Williams5, 19,
- James E Womack4,
- Natasja H Wye2,
- George Yang2,
- Shaying Zhao20, 21 and
- the International Bovine BAC Mapping Consortium
© Snelling et al.; licensee BioMed Central Ltd. 2007
Received: 16 May 2007
Accepted: 14 August 2007
Published: 14 August 2007
Cattle are important agriculturally and relevant as a model organism. Previously described genetic and radiation hybrid (RH) maps of the bovine genome have been used to identify genomic regions and genes affecting specific traits. Application of these maps to identify influential genetic polymorphisms will be enhanced by integration with each other and with bacterial artificial chromosome (BAC) libraries. The BAC libraries and clone maps are essential for the hybrid clone-by-clone/whole-genome shotgun sequencing approach taken by the bovine genome sequencing project.
A bovine BAC map was constructed with HindIII restriction digest fragments of 290,797 BAC clones from animals of three different breeds. Comparative mapping of 422,522 BAC end sequences assisted with BAC map ordering and assembly. Genotypes and pedigree from two genetic maps and marker scores from three whole-genome RH panels were consolidated on a 17,254-marker composite map. Sequence similarity allowed integrating the BAC and composite maps with the bovine draft assembly (Btau3.1), establishing a comprehensive resource describing the bovine genome. Agreement between the marker and BAC maps and the draft assembly is high, although discrepancies exist. The composite and BAC maps are more similar than either is to the draft assembly.
Further refinement of the maps and greater integration into the genome assembly process may contribute to a high quality assembly. The maps provide resources to associate phenotypic variation with underlying genomic variation, and are crucial resources for understanding the biology underpinning this important ruminant species so closely associated with humans.
Cattle have played a crucial role throughout recent human agrarian history, providing draft power, milk and meat for human consumption since domestication 8,000 years ago [1, 2]. Cattle studies have contributed to our knowledge of endocrine function, fertilization, and growth, and enhanced our understanding of genetics, selection and evolution [3, 4]. However, much remains to be determined; particularly, how cattle have adapted to intense selection pressures since domestication and how ruminants convert low quality forages into energy and protein-dense meat and milk. Worldwide, roughly 1,000 different breeds and varieties have been recognized . These breeds originated in different locations, were subjected to different environments, and possess somewhat different characteristics as a consequence of ongoing natural and artificial selection. Ancestry of much of today's seedstock can be traced to breed-specific herdbooks established in the mid-1800's . Formalization of the genetic selection process has culminated in extensive estimation of heritabilities, and genetic and environmental relationships between traits [6, 7], coupled with objective approaches to animal evaluation  and selection . These have been widely utilized in the development of modern beef and dairy performance recording and evaluation schemes [10–12]. The phenotype-based selection systems developed and optimized in the last century are now moving towards integration of DNA information to accelerate genetic progress.
While numerous quantitative trait loci (QTL) have been mapped [13, 14], only a small number of quantitative trait nucleotides or causative mutations [15–19] have been identified for economically important cattle QTL. We expect further development of cattle genomic resources to accelerate discovery of causative mutations, and facilitate genome-wide selection that considers whole genome sequence and associated single-nucleotide polymorphisms (SNPs), rather than specific individual loci .
Understanding the genetic basis of breed differentiation through natural and artificial selection, production related traits, and disease will be greatly advanced by the availability of the genomic sequence of cattle. The foundation of the genomic sequence is provided by a clone-based physical map. Development of a clone-based physical map also offers a resource to accelerate discovery of polymorphisms within and between breeds, including causal polymorphisms contributing to a wide variety of bovine traits. The physical map is a source of genomic clones for sequencing templates and functional studies, and can also be employed to assist in the assembly of whole genome shotgun sequence [21–23].
The value of a clone-based map can be enhanced by establishing connections to the annotated genome sequences of closely related species , as well as to sequence tagged site (STS) maps of the same species. Links established between clones and annotated sequence can be used to identify specific clones containing genes of interest, and connections to genetic STS maps can indicate clones harboring QTL. Existing bovine genetic maps, however, have limited utility for identifying clones containing positional candidate genes near QTL, due to a lack of gene-specific genetic markers and a lack of recombination to separate closely linked markers within the existing bovine genetic maps [25–27]. Available gene-rich radiation hybrid (RH) maps [28–30] have greater short-range resolution than genetic maps, but their whole-chromosome ordering may be unreliable , and RH maps often lack many of the polymorphic markers that are needed to refine the locations of QTL. Exploiting the complementary resolution of genetic and RH data, a composite map can consolidate marker information to more efficiently indicate genes and sequence in the vicinity of QTL. Connections between a clone-based physical map, composite marker map, and annotated genome sequences will greatly facilitate the annotation of newly generated and assembled sequence. We report here the generation of a fingerprinted bacterial artificial chromosome (BAC)-based physical map, representing approximately 15.5-fold coverage of the bovine genome; the construction of a composite marker map from two linkage and three RH data sets; and connections with annotated human sequence and the largely unannotated draft bovine genome sequence.
BAC clone fingerprinting
BAC library fingerprinting summary
Library construction enzyme†
Clones in FPC‡
Average size (kb)§
Hereford bull L1 Domino 99373
Angus bull, Angus cow
A small subset of the BAC clones (approximately 5% of all fingerprinted clones) generated low complexity restriction digest fingerprint patterns, containing a single large molecular weight HindIII fragment (> 30 kb, the largest marker fragment), in addition to the expected vector-specific HindIII fragment(s). These clones appeared to lack HindIII sites within the insert, and were therefore unsuitable for fingerprint pattern-based assembly. Digestion of a small number of these clones with EcoRI generated fingerprints of one to four restriction fragments, each present in multiple copies (data not shown), suggesting that the inserts in these BAC clones were probably derived from regions of repetitive sequence . Thus, while represented within the BAC library these genomic regions are under-represented within the fingerprint map.
Aligning bovine BAC clones to the human genome
Summary of BAC clones, sequences, and matches with composite map markers
% Clones with sequence
BAC map assembly
Fingerprint map summary
Number of contigs
Clones in contigs (% of total)
Average number of clones per contig
Largest number of clones in a contig
Contigs assigned to bovine chromosomes
Autosome and X assignment by comparative alignments
Y by SRY-positive probes
Number clones in assigned contigs (% of total)
Average number of clones per assigned contig
Clones per autosome and X contig
Clones per Y contig
N50 size of assigned contigs
Average number of clones per unassigned contig
N50 size of unassigned contigs
Average number of contigs per autosome
When merging contigs, Sulston fingerprint similarity scores  were relaxed from the initial stringency, considering that the search space was limited to contig ends, and comparative alignments provided supporting evidence. A maximum of four unmatched restriction fragments across the merge point (fragments present in only one of the two merged contigs) allowed for: first, minor errors in fragment identification by BandLeader, the automated band calling software; second, the fact that HindIII does not cleanly excise the insert from the vector in the CHORI-240 and RPCI-42 BAC clones, resulting in two vector-insert junction fragments of unpredictable size that are not expected to be shared by neighboring clones; and third, the potential presence of polymorphic restriction fragments produced by indels, duplications, SNPs creating or destroying HindIII restriction sites, and other haplotype and/or breed-specific DNA variations that result in restriction fragment length differences. Automated scripts were employed throughout the merging process, although merges based on comparative alignments were manually inspected. Discrepancies between FPC/CORAL and human-based clone orders were identified, and manually evaluated in conjunction with the fingerprint images. Clones were rearranged to be consistent with human order only when the rearrangements were supported by fingerprint data.
Assigning contigs to bovine chromosomes using comparative mapping data
Summary of map contig coverage of bovine chromosomes
No. of contigs
Estimated size (Mbp)*
Genome assembly size (Mbp)
Syntenic human chromosomes
Human genome coverage (Mbp)†
A composite bovine marker map
Markers contributed by linkage and radiation hybrid data sets to the composite bovine map*
The markers were mapped as 15,627 discrete entities, accounting for RH markers showing identical retention patterns within a panel. Three maps were computed for each chromosome. Only linkage and RH data were considered to determine an unassisted order. Ordering information from the BAC map and Btau3.1 draft assembly  was introduced for BAC- and sequence-assisted maps. Centimorgan (cM) and kilobase pair locations from the most likely of the three maps were interpolated using location database software , and confidence intervals were estimated. Mean separation between projected marker positions was 0.27 cM, or 228 Kbp. Estimated confidence intervals (CIs) indicate that 5,241 markers, occupying 4,639 positions, could not be repositioned relative to other markers without reducing likelihood. Expressed in cM, the median estimated CI is 0.8 cM, and 62.5% of markers have CI ≤ 1 cM. A tiny fraction of markers (0.1%) have extremely ambiguous positions with estimated CI ≥ 25 cM. Most of the extremely ambiguous markers are expressed sequence tag (EST)-based SNPs and have ambiguous placement on the SIAG-USDA linkage map .
Marker-clone, marker-sequence and clone-sequence alignments
In addition to markers derived from BES, direct alignments between marker and clone-based sequences and indirect alignments using alignments of marker and clone sequences to whole genome shotgun sequence (WGS) contigs were used to anchor markers to the BAC fingerprint map. Matches between 10,313 markers and 25,591 BAC clones in 426 contigs were identified using e-PCR  and BLAT . This includes 9,916 markers that matched 23,724 clones in 359 contigs assigned to autosomes or the X chromosome. Composite and BAC map chromosome assignments were consistent for 97.6% of the 8,902 markers that matched clones from a single chromosome. For 97.2% of the 1,014 markers that matched clones from multiple chromosomes (autosomes and X), at least one of the matched clones was assigned to the same chromosome as the marker. Additionally, clone-marker alignments suggested placement of 23 contigs that were not assigned to chromosomes by comparative alignments.
The comparative map-based chromosome assignments were supported by marker-clone alignments for 290 of the 300 contigs assigned to autosomes, and 66 of the 79 contigs assigned to X. The composite map indicated different chromosome assignments for only three autosome-assigned contigs (ctg7970 - BTA10; ctg17005 - BTA6; ctg25050 - BTA4). Each of these is located near a bovine-human breakpoint, so ambiguity in the bovine-human comparative map may have resulted in the apparent misassignment. For 112 of the 121 contigs containing clones that were assayed for a marker, comparative chromosome assignments agree with the physical probe assignments.
Further support for the comparative map-based chromosome assignments was provided by the first-generation bovine physical map produced by the French National Institute for Agricultural Research (INRA) . The IBBMC and INRA BAC maps share 18,980 CHORI-240 clones, which allow 227 of IBBMC contigs to be joined with 653 INRA contigs. Chromosome assignments concur for 192 of the 214 assigned IBBMC contigs containing a shared CHORI-240 clone. Markers associated with INRA clones suggest chromosome assignments for the 13 unassigned IBBMC contigs containing a shared clone, although the assignments for two of these contigs remain ambiguous.
Strong agreement between composite and BAC map orders is indicated by Spearman's (ρ) and Kendall's (τ) rank correlation coefficients. Spearman's ρ, used to measure strength of the relationship between marker order along the two maps, ranges from 0.97 to 1.00 among autosomal markers matched to BAC clones. Coefficients are > 0.99 for 21 of the 29 autosomes. Kendall's τ indicates rearrangement necessary to reconcile orders, and ranges from 0.92 to 0.99 for the autosomal maps. Agreement between orders along the X chromosome is slightly weaker, with ρ = 0.91 and τ = 0.72. For all chromosomes, ρ exceeds τ, suggesting any rearrangements between the maps are predominately local, involving markers in close proximity to each other.
Composite map markers were also matched to the Btau3.1 draft assembly using e-PCR and BLAT. The draft assembly consists of contig sequences and intermediate scaffolds assembled with ATLAS , which were arranged on whole-chromosome scaffolds according to a set of bovine markers. A total of 15,746 markers matched 14,952 contig sequences, including15,498 markers matching whole-chromosome scaffolds, and 706 markers matching unassigned scaffolds. Further, 3,897 markers matched multiple contigs; for 3,495 of these, all contigs were within the same intermediate scaffold, 175 matched multiple intermediate scaffolds assigned to the same chromosome, and 227 matched scaffolds assigned to different chromosomes. Composite map and draft assembly chromosome assignments were consistent for 96.4% of the markers matching a single chromosome, and one assembly assignment concurred with the composite map for 92.7% of markers matching multiple chromosomes. Markers matched 668 unassigned scaffolds; 542 unassigned scaffolds matched a single marker, 97 matched more than one marker from the same chromosome, and 29 unassigned scaffolds matched markers from two or more chromosomes. Rank correlations suggest somewhat greater rearrangement between the composite map and draft assembly than was observed between the composite and BAC maps; ρ ranged from 0.89 to 1.00, with ρ of 13 chromosomes greater than 0.99; and τ was between 0.87 and 0.96. The level of agreement for the X chromosome was similar to the autosomes, with ρ = 0.98 and τ = 0.95.
End and other partial sequences from 156,783 fingerprinted BAC clones were also aligned to the assembled draft genome sequence. These include: 128,774 autosome- or X-assigned clones matched to whole-chromosome scaffold sequences; 26,587 autosome- or X-assigned clones matched to unassigned scaffolds; and 14,586 singleton BAC clones or clones in unassigned fingerprint contigs matched to whole-chromosome scaffold sequence. BAC map and draft assembly chromosome assignments were consistent for 96.1% of the aligned clones. All 379 autosome- and X-assigned fingerprint contigs contained at least one clone with sequence matching the bovine assembly. All autosomal BAC contigs contained at least one clone matching sequence assembled for that autosome, and the most frequently matched chromosome was consistent with BAC map chromosome assignment for 98.7% of the 300 autosome-assigned BAC contigs. Thirty-seven contigs exclusively matched sequence assembled for the assigned chromosome. Agreement between BAC map and assembly orders was less than that observed between the composite map and either the BAC map or assembly, with ρ between 0.88 and 0.95, and τ between 0.81 and 0.91 for the autosomes; for X, ρ = 0.79 and τ = 0.62.
After removing markers that matched multiple chromosomes, or had widely separated matches on a single chromosome, a set of 14,123 markers was identified to evaluate the likelihood of marker order along the Btau3.1 draft assembly, and to provide the starting point for sequence-assisted reordering of the composite map. Similarly, 7,780 markers with consistent BAC and composite map chromosome assignments matched to single or close clones on the BAC map were used to evaluate the likelihood of markers in the BAC map order, and initiate a BAC-assisted reordering of the composite map. Log10-likelihoods of markers ordered according to the assembly are lower than for the same markers in the unassisted composite map order (Additional data file 4). Differences in log10-likelihoods between markers ordered according to the BAC or unassisted composite map are not as pronounced; the BAC map orders for BTA17 and BTA20 are more likely than the unassisted order. After rearranging and adding markers not included in the starting order, the BAC-assisted order was more likely than the unassisted order for 20 autosomes (Additional data file 5). Sequence-assisted orders were never the most likely computed order, but the log10-likelihoods are intermediate between the unassisted and BAC-assisted orders for ten chromosomes.
A BAC physical map that spans the majority of the bovine genome has been constructed. Genome coverage by the BAC map is at least equal to that of the Btau 3.1 7X draft sequence assembly. Genome size estimated from the BAC map is 3.1 Gbp, somewhat larger than the 2.9 Gbp estimated by the bovine genome sequencing project, but at the low end of the 3.1 to 3.8 Gbp range of estimates obtained from different measures of haploid DNA content [61,62]. The use of three different BAC libraries each constructed using different restriction enzymes may have increased coverage over that possible with a single restriction enzyme, because certain genomic regions may not be clonable due to recognition site biases. Coverage by the BAC map may still be incomplete, however, because other highly repetitive, complex regions, such as telomeric and centromeric regions, may not be clonable with any enzyme, and may not be represented in any of the BAC libraries or the BAC map. Error in our estimates of genome size and genome coverage by individual contigs cannot be fully ascertained without a more finished assembly of bovine genomic sequence.
A composite linkage/RH map was also developed. This map consolidates available data to place markers in a consensus order and approximate positions in common cM and Kbp scales. The CarthaGene [31, 51] procedures used to construct the bovine composite map have also been used to order markers on composite linkage maps of a parasitoid wasp , grapevine , rapeseed  and conifers , and a porcine map that combines linkage and RH data . A related approach using weighted least squares is implemented in JoinMap  to compute composite linkage maps produced from different populations. Combining four linkage data sets for the Picea mariana × Picea rubens species complex  reported that similar maps were obtained with either JoinMap or CarthaGene.
Examination of preliminary composite and BAC maps showed that discrepancies could not readily be resolved. Beyond the inevitable laboratory errors contributing to incorrect marker and clone order and contig membership, potential causes of discrepancies include sub-optimal orders of both maps, ambiguous orders, and error arising from spurious marker and clone sequence alignments. The composite map orders are probably less than optimal, because explicitly evaluating n!/2 possible orders, where n may represent a hundred or more markers on each chromosome, is computationally infeasible. The traveling salesman problem (TSP) approach [69, 70] can be taken to implicitly determine optimal order for some RH data sets, but the combined data sets were not suitable for TSP analysis. Our approach of adding markers to an initial order, and using iterative flips and polish consistently produced the most likely order when testing alternative map construction strategies. The final order from this approach, however, is influenced by the starting order, and identifying the optimal order is not guaranteed. The process of computing maps from three different starting orders was intended to avoid bias towards any of the contributing marker maps, and to introduce fine-scale arrangements supported by external data that are not explicitly considered in the ordering process. Unassisted maps started from a pair of markers from opposite ends of each chromosome, to avoid bias towards any of the contributing maps. BAC-assisted orders started with markers ordered according to the BAC map, to introduce arrangements supported by fingerprint data. Similarly, sequence-assisted orders starting from markers ordered according to the draft assembly introduced assembly-based arrangements. This strategy of seeding the marker order according to the BAC map or assembly introduces marker arrangements that may not otherwise be evaluated. By considering these arrangements, the BAC map and sequence have some influence on the resulting composite map, but testing rearrangements of the seeded order ensures that the composite map will not include arrangements that are not supported by the marker data.
Some apparent discrepancies may result from ambiguously placed markers, which can occupy several positions on the composite map with equal likelihood. Ambiguity on the composite map also affects orders within regions between markers shared by two or more data sets, where there are no data to indicate the best merged order of markers that are unique to each data set. Other discrepancies may be attributed to spurious marker-clone alignments that appear only because of the erroneous match between a marker and clone, while some real discrepancies may be hidden by undetected links between markers and clones. Erroneous alignments may also contribute to errors on the composite map, if markers from different data sets are matched by incorrect alignments to the same GenBank bovine sequence or EST cluster (Bos taurus gene index or NCBI UniGene). The sequence based marker matching procedures were implemented to overcome inconsistent marker nomenclature, and result in identification of many more common markers than matches based solely on marker name. The sequence based procedures are dependent, however, on correct assembly of the sequences used to link markers. The two-point procedures to resolve inconsistent chromosome assignments will correctly break matches between markers that should be placed on different chromosomes, but will not detect incorrect matches between markers that should be separated on the same chromosome.
Repetitive sequence, particularly segmental duplications and other large scale genome variants , may exacerbate discrepancies between the marker and BAC maps. BAC clones that encompass or overlap a large duplication may be incorrectly assigned to the same contig, or correctly assigned but incorrectly ordered within a contig due to similarity of the fingerprints. Markers that match non-overlapping clones may indicate presence of repetitive sequence in the marker, clone and/or intermediate WGS sequence. Some rearrangements between the marker and BAC maps may be real, considering the diversity of genetic material represented by the maps, which include Bos taurus × Bos indicus crosses and several Bos taurus breeds. Breed-specific rearrangements have been shown in sheep , and structural variations, including inversions and inter- and intra-chromosomal translocations, have been detected among humans of diverse ethnic origin [73, 74].
The level of disagreement between the composite map and assembled sequence can be reduced by rearranging the intermediate scaffold sequences according to the average (mean or median) composite map position of markers matching each scaffold, instead of ordering scaffolds according to minimum marker position within each scaffold, as was done for the current draft assembly. Basing the order of scaffolds on average rather than minimum marker position will be more accurate, especially if the minimum position is for a marker misplaced relative to other markers matching the scaffold. Remaining discrepancies between the composite map and assembly may have similar explanations as the inconsistencies between the composite and BAC maps: laboratory errors, ordering errors and ambiguous orders, spurious alignments, complications arising from repetitive DNA sequence, and structural variation among the genomes represented by the composite map and sequence.
Rearranging scaffolds to more accurately represent the marker map will also reduce disagreement between the assembled sequence and BAC map. Further refinement of the draft assembly may be achieved through greater consideration of the BAC map in the assembly process. Genomes represented by the BAC map and bovine reference sequence should be highly similar; reference DNA was obtained from a daughter of the bull used to construct the CHORI-240 library, which predominates the BAC map. The relationship coefficient (rxy)  of 0.954 between these two partially inbred individuals indicates their genomes will not be exactly the same, but will be much more similar than sire and progeny genomes resulting from mating unrelated parents (rxy = 0.5).
Besides the high density of alignments between WGS and BAC sequences, which can anchor and orient more genomic sequence than any marker map, paired BES alignments are especially valuable to orient and space adjacent sequence segments. Paired BES information has been used along with sequenced mammalian genomes to construct detailed framework maps . In the current context, it can also be used as an independent check of the draft assembly and map coherency. Paired BES alignments  against the current draft assembly revealed possible clone identification and assembly errors. Systematic identification errors may affect BACs on 25 or more plates of the CHORI-240 library, containing approximately 3% of the BAC clones (A Ratnakumar and B Dalrymple, unpublished). Mis-identification may account for some chromosome assignment discrepancies between the composite and BAC map, and between the BAC map and draft assembly, but will not affect within-chromosome order comparisons, which do not include discrepant chromosome assignments. Considering estimated clone size, orientation and separation between paired BES alignments may improve the genome assembly. In a preliminary examination of a 50-scaffold region of BTA1, links between paired BES suggested a number of currently unassigned scaffolds should be placed in the region, and the neighbors and/or orientation of most scaffolds should change (B Dalrymple, W Barris and A Ratnakumar, unpublished). The discrepancies in order are predominantly local, consistent with the Kendall's correlation analysis.
Currently observed inconsistencies between the composite map, BAC map, and assembled sequence suggest that all may be improved to provide more accurate representations of the bovine genome. Some improvement of the composite map was realized by including fingerprint-scale information to reorder markers on the BAC-assisted map. Re-examination of the BAC map in light of the refined marker map may suggest further refinement of the BAC map, involving merges between contigs, splitting of contigs and other rearrangements. Similarly, examination of the draft sequence assembly relative to the BAC map may increase consistency with the BAC map and composite map. An iterative approach to refining the various maps is suggested. However, no rearrangement should occur for the sake of increasing consistency between the maps and sequence, unless appropriate data supporting the rearrangement are available. Discrepancies between the maps are relevant, pointing to regions where more experimental data are needed, where caution is required when examining contents of particular regions, and where bovine genomes may be prone to differ due to naturally occurring large-scale polymorphisms.
The composite map consolidates available bovine mapping data, and leverages the complementary resolution of linkage and RH maps to provide a comprehensive marker map of the bovine genome. The BAC fingerprint map provides a resource to define comparative synteny, order and orient bovine genomic sequence, and estimate genome size and complexity. Further elucidation of the bovine genome is obtained from integration of the composite and BAC maps with annotated human sequence, draft bovine genomic sequence, and QTL describing genomic regions associated with phenotypic variation. QTL, described relative to the marker maps, can be anchored to underlying bovine genomic sequence through the BAC map. Annotation, transferred through the high-resolution bovine-human BAC-based comparative map, supplies information about genes and gene function needed to enhance our understanding of biological mechanisms affecting agriculturally important traits. The BAC map is a valuable resource for the development of genomic tools to further our knowledge of evolution of this species, which has undergone natural, undocumented and documented artificial selection, and which may contribute further insight into human conditions.
Materials and methods
BAC library resources
Clones from three BAC libraries were fingerprinted: CHORI-240 derived from Hereford bull L1 Domino 99375 DNA  (PJ de Jong, K Osoegawa and C Shu, unpublished) and RPCI-42 derived from Holstein bull DNA [32, 33] were constructed at BACPAC Resources Centre . The TAMBT library, containing clones derived from Angus bull and cow DNA, was constructed at Texas A&M University  (CA Gill and SL Davis, unpublished).
Based on recorded pedigree, the Hereford bull used for CHORI-240 has an inbreeding coefficient (Fx) of 0.31 . His daughter used for the bovine genome sequencing project has Fx = 0.30, and the rxy between these two individuals is 0.954 (MD MacNeil, personal communication).
BAC clone fingerprinting
Fingerprints were generated using an agarose-gel based methodology [23, 78, 79]. Briefly, BAC clones were cultured overnight in 96-well format and DNA was extracted using an alkaline lysis procedure. The BAC DNA was digested with HindIII (New England Biolabs, Ipswich, MA, USA) and the resulting fragment sizes were resolved by electrophoresis on agarose (Cambrex BioWhittaker, Walkersville, MD, USA) gels. Gels were stained after electrophoresis with SYBR Green I (Invitrogen, Carlsbad, CA, USA) and scanned using a Molecular Dynamics Fluorimager 595. The digitized images were lane tracked interactively using Image software [46, 80, 81] and restriction fragments were automatically identified and sized using BandLeader . Restriction fragments within a size range of approximately 600 bp to 30 Kbp were collected. HindIII fragments predicted by their size as being derived from sequences internal to the vector were removed from the fragment list for each clone.
BAC end sequencing
Summary of CHORI-240, RPCI-42 and TAMBT BAC clones represented by sequences deposited in the genome survey sequence division of GenBank
Anchoring BAC clones to the human genome assembly
BES were aligned to repeat-masked human genome sequence assemblies (UCSC hg17, based on NCBI Build 35; and UCSC hg18, based on NCBI Build 36.1)  using BLASTN  with options -z 3095016460 -m 8. Only the best achieved hits were considered. Paired-end alignments were required to satisfy the following criteria: E-value ≤ 1e-2 for both alignments, with alignments ≤ 400 Kbp apart and in opposite orientations. Single end alignments were required to have E-values ≤ 1e-8. As described below, alignments to hg17 were used in conjunction with the bovine-human comparative map described by  to merge contigs and arrange contigs on chromosomes. Alignments to hg18 were used to define the bovine BAC-human comparative map (Additional data file 2).
Initial fingerprint map assembly
BAC fingerprints were assembled using FPC [43, 44]. The initial assembly was performed using the default parameters (tolerance 7; min bands 3; best = 10; no CpM) and a cutoff value of 1 × 10-16. The fingerprinted clones were then screened and filtered as follows to remove fingerprint patterns indicative of artifactual data. First, a software application, MapMop, was developed to calculate the distributions of insert size and number of restriction fragments for the clones. Based on these distributions, filtering parameters were determined to identify 'outlier' clones that represented extremes for insert size and/or number of restriction fragments, suggesting problematic data such as that resulting from partial digestion of the BAC DNA, or the presence of DNA from more than a single BAC clone (cross-well contamination). Second, initial filtering parameters were assessed by visual inspection of representative clone fingerprints and were adjusted to minimize the retention of poor quality data while limiting the loss of high quality data. Filtering parameters were determined separately for individual libraries, for clones within contigs and for clones that were not assembled into contigs ('singletons').
Following filtering, the remaining fingerprints were again assembled at a cutoff value of 1 × 10-16. The DQer function in FPC was used to reassemble any contigs that contained 'Q' clones, which exhibit an unusual number of extra bands or gaps between bands matching the Q clone to other clones, resulting in false positive overlaps between putative contigs. The presence of Q clones within a contig is suggestive of misassembly due to false-positive fingerprint matches. The DQer function reassembled the clones within the contig with successively increasing stringency until no Q clones remained. The resulting contigs were subsequently processed with an automated clone ordering application, CORAL , to refine clone order within the contigs.
Merging map contigs
Contigs were merged in multiple steps using automated scripts to identify and execute joins between contigs. Initial merges were performed using only the fingerprint data to identify candidate merge points for contigs that satisfied the criteria of Sulston scores of 9 × 10-8 for at least two contig end clones, and that had no more than four unconfirmed fragments at the merge point. An unconfirmed fragment is one that is present in the fingerprint of the end clone of one contig but without a match to fragments in fingerprints of end clones in either of the contigs at the merge point. A subset of merged contigs was manually reviewed to evaluate whether these parameters resulted in incorrect merges.
After making the initial merges, BES alignments to the human sequence assembly were examined to identify candidate contig merges based on their human genome sequence coordinates. Merges were made in cases where the Sulston score match between end clones was 9 × 10-7 and there were four or fewer unconfirmed fragments. Automated scripts and human inspection were used to identify and eliminate incorrect clone orders and merges at multiple points throughout the merging process.
Estimating map contig size
Contig sizes were estimated from fingerprint data with an algorithm that compared the restriction fragments of overlapping clone pairs in the canonical clone set for each contig. Canonical clones are the set of non-redundant overlapping clones spanning a contig that each represent a unique complement of restriction fragments in their fingerprint, such that the remaining non-canonical clones within the contig are subsumed by the canonical clones (that is, all the restriction fragments in the fingerprint of a non-canonical clone are completely represented in one of the canonical clones). The unique fragments for each canonical clone were identified, and their sizes were summed to estimate the overall size of the contigs. Specifically, the algorithm performed the following for each contig: first, sum the sizes of all the fragments in the left-most canonical clone in the contig to create a cumulative size estimate; second, identify the next canonical clone immediately to the right and identify its unique fragments (any fragments not shared with the previous canonical clone to the left or the next canonical clone to the right), then add the sizes of these unique fragments to the cumulative size estimate; third, repeat step 2 until all unique fragments in the canonical clones have been identified and summed to give a total size estimate for the contig. Fragments were considered to be the same if their calculated standard mobilities were within ten mobility units of each other.
Composite bovine map
The composite map was built using data from two independent linkage maps, as well as markers scored for presence or absence on three independent whole-genome RH panels (Table 5). Genotypes and pedigrees were those used for the SIAG-USDA linkage map  with EST-based SNP , and the UAMU SNP linkage map . The two maps anchor 7,404 markers along 30 bovine chromosomes (29 autosomes and the X chromosome), albeit at relatively low resolution with many markers not separated by observed recombination. Markers scored on at least one of the three independent RH panels provided the basis to separate closely linked markers as well as place 9,850 additional markers that were not represented in the linkage maps. Radiation hybrid data included those used for the ILTX-2005 map scored on the 5,000 rad Illinois-Texas panel , the SIAG map scored on a 7,000 rad panel , and the BovGen map  with markers scored on the 3,000 rad ComRad panel . For this work, additional markers scored on the ComRad panel are also considered part of the BovGen data set. These include Illumina BeadStation-scored SNP  identified from the bovine genome sequencing initiative, and a number of amplified fragment length polymorphism markers. The UAMU linkage map SNPs are a subset of the sequencing project SNPs scored on the ComRad RH panel.
Independent two-point analyses of each RH data set were conducted to identify markers to be included on the composite map. Markers common to each RH data set and the SIAG-USDA linkage data were identified. Starting from this set, markers linked with a two-point LOD > 7.0 and < 40 centiRad (cR) separation from previously linked marker were identified in successive passes through the two-point results. Markers assigned to each chromosome-specific linkage group were ordered by an automated process , seeded with three different orders. An initial, unassisted order started with a pair of markers from opposite ends of the chromosome. The BAC-assisted order started with the BAC map order of markers that were matched to BAC clones, and the sequence-assisted order started with the assembly order of markers that were anchored to the assembly. Only markers with consistent chromosome assignments, and unambiguous placement on the BAC map or sequence were included in the starting orders. Log10-likelihoods of markers in the BAC- and sequence-based starting orders were computed, iterative polish and flips procedures were applied to determine a more likely order of markers included in the starting order, remaining markers assigned to the chromosome were sequentially added, and the final order was determined with iterative polish and flips. Confidence intervals surrounding marker placement were approximated from polish applied to the final order. Cytogenetic band, cM and Kbp positions for each marker were computed with a modification of the ldbf program , using the native unit (cM and cR) maps in the most likely consensus order determined from the three different starting orders.
Marker, clone and genomic sequence alignment
Markers placed on the composite map were associated with BAC clones in the fingerprint map via direct alignment of marker and BAC end- or sub-clone sequences, and indirect alignment of marker and BAC sequences to the same WGS contig sequence. Direct marker-clone, and indirect marker-WGS-clone alignments were identified by e-PCR and BLAT. Alignments of primer pairs against BAC clone and WGS by e-PCR allowed no more than one mismatch or gap. BLAT was used to align marker sequences with BES and WGS contigs, and to align BAC clones with WGS contigs. BLAT alignments required exact matches and a bit score of at least 100, 99.5% identity with a score of 250, or 99% identity and a score of 500. Direct matches were identified from markers matching BAC clone sequences, and indirect matches from markers and BAC clones matching the same WGS contig.
Marker sequence used for BLAT alignments was obtained from GenBank when information to associate the marker with a GenBank accession number was available, or from STS sequence supplied with the marker. When only primer sequences were available, the e-PCR results used to match markers across data sets were also used to construct an amplimer sequence. Sequence between primer locations was extracted from each e-PCR match, and assembled with Phrap (version 0.990329) . This assembled sequence was used to represent the marker only when the Phrap assembly resulted in a single contig.
The collection of GenBank bovine sequence used for e-PCR alignments to match markers across linkage and RH data sets includes bovine sequences deposited in the genome survey sequence (GSS), STS, patent (PAT), mammalian (MAM), and EST divisions of GenBank after release of the second bovine draft assembly. Bovine sequences were identified by taxonomy identification numbers for Bos taurus (txid 9913), Bos indicus × Bos taurus (txid 30522), Bos taurus × Bos indicus (txid 30523), and Bos sp. (txid 29061, associated with some patent [PAT] division sequences). The MAM division sequences included WGS contig sequences from the second bovine sequence assembly [GenBank:AAFC02000000]. Markers and BAC sequences were also aligned with WGS contigs from the third bovine draft assembly [GenBank:AAFC03000000] and scaffold sequences (whole-chromosome scaffolds [GenBank:CM000177-CM000206]; unassigned scaffolds [GenBank:CH974204-CH980624]).
Alignment of the BAC fingerprint and composite marker maps
Associations between markers and clones were used to assess and refine chromosome assignments and the order of fingerprint contigs, where the contigs were initially assigned and ordered along chromosomes using end-sequence alignments to human sequence and bovine-human comparative maps [29, 47]. For each fingerprint contig containing clones associated with markers mapped on a preliminary composite map, marker positions unambiguously linked to a single contig were identified, and uninterrupted runs of marker positions associated with a single contig were determined. Marker positions were regressed on corresponding clone position within a contig for the run with the greatest number of markers to predict clone position on the composite map. Predicted positions of clones within contigs linked to the composite map by a single marker position were set to that marker position. Discrepancies between predicted clone positions, and composite map positions of markers linked to those clones were identified. Fingerprints of clones linked to the observed discrepancies were examined, and clones were rearranged to eliminate discrepancies only if the rearrangement was supported by fingerprint data. The preliminary composite map used included SIAG-USDA linkage data, and RH markers and vectors from SIAG, the second generation Illinois-Texas map , and a subset of the BovGen data used for the current map.
BAC-assisted composite map
Direct marker-clone and indirect marker-WGS-clone alignments were used to identify the set of markers from the current composite map with consistent, unambiguous alignments to the BAC map. The order of these markers on the BAC map was used to seed the ordering process for each chromosome. A map of the markers matching the BAC map was computed, followed by iterative polish and flips to reorder markers in that set. Markers assigned to the chromosome, but not matching BAC clones, were added with buildfw, and a final BAC-assisted order obtained from iterative polish and flips after all assigned markers were included.
Additional data files
The following additional data are available with the online version of this paper. Additional data file 1 describes analyses of BAC end sequences. Additional data file 2 contains figures showing the bovine-human comparative map defined by the BAC map and BES alignments. Additional data file 3 is a table containing placement of markers on the composite map. Additional data file 4 is a table of log10 likelihoods of markers ordered according to the BAC map, Btau3.1 draft assembly, and unassisted composite map. Additional data file 5 is a table of log10 likelihoods for the unassisted, BAC-assisted and sequence-assisted composite maps.
bacterial artificial chromosome
BAC end sequence
expressed sequence tag
International Bovine BAC Mapping Consortium
French National Institute for Agricultural Research
quantitative trait loci
sex-determining region Y
sequence tagged site
US Department of Agriculture
whole genome shotgun sequence.
The authors acknowledge the support of the United States Department of Agriculture (National Research Initiative Tools and Reagents Grant AG2004-3520-14196; National Research Initiative Grant 2003-35205-12825; USDACooperative State Research, Education, and Extension Service Livestock Genome Sequencing Initiative andUSDA Agricultural Research Service Agreement 58-5438-2-313), the Alberta Science and Research Authority, Agriculture and Agri-Foods Canada, the Biotechnology and Biological Sciences Research Council, The European Commission, the Commonwealth Scientific and Industrial Research Organization of Australia, AgResearch Ltd, New Zealand Foundation for Research, Science and Technology, EMBRAPA Brazil, National Council for Scientific and Technological Development (CNPq) Brazil, the Roslin Institute Scotland, the French National Institute for Agricultural Research (INRA), Analysis of the Genomes of Farmed Animals Project (AGENAE), Genoscope, and the Texas Agricultural Experiment Station. The Alliance for Animal Genomics and specifically Kellye Eversole's valuable efforts in coordinating industry and institutional support is acknowledged. We also wish to acknowledge the BAC end sequencing work done by the TIGR BAC End Team, and members of Mapping and Sequencing groups at the GSC, especially Martin Krzywinski, who developed essential software. Special thanks to Wes Barris and Sean McWilliam for assistance with the Gbrowse sites, and to Abhirami Ratnakumar and Wes Barris for sharing preliminary results from detailed analysis of paired BES alignments. MAM, RH and SJMJ are scholars of the Michael Smith Foundation for Health Research. The authors gratefully acknowledge the early pre-publication access under the Fort Lauderdale conventions to the draft bovine genome sequence provided by the Baylor College of Medicine Human Genome Sequencing Center and the Bovine Genome Sequencing Project Consortium.
- Diamond J: Guns, Germs and Steel: the Fates of Human Societies. 1997, New York, NY, USA: WW Norton & CompanyGoogle Scholar
- Willham RL: From husbandry to science: A highly significant facet of our livestock heritage. J Anim Sci. 1986, 62: 1742-1758.Google Scholar
- Wright S: Evolution in Mendelian populations. Genetics. 1931, 16: 97-159.PubMedPubMed CentralGoogle Scholar
- Lynch M, Walsh B: Genetics and Analysis of Quantitative Traits. 1998, Sunderland, MA, USA: Sinauer Associates, IncGoogle Scholar
- Mason IL: World Dictionary of Livestock Breeds. 1998, Wallingford, UK: CAB International, 3Google Scholar
- Koots KR, Gibson JP, Smith C, Wilton JW: Analyses of published genetic parameter estimates for beef production traits. 1. Heritability. Anim Breed Abst. 1994, 62: 309-338.Google Scholar
- Koots KR, Gibson JP, Wilton JW: Analyses of published genetic parameter estimates for beef production traits. 2. Phenotypic and genetic correlations. Anim Breed Abstr. 1994, 62: 829-853.Google Scholar
- Henderson CR: Applications of Linear Models in Animal Breeding. 1984, Guelph, ON, Canada: University of GuelphGoogle Scholar
- Harris DL, Newman S: Breeding for profit: synergism between genetic improvement and livestock production (a review). J Anim Sci. 1994, 72: 2178-2200.PubMedGoogle Scholar
- de Mattos D, Bertrand JK, Misztal I: Investigation of genotype × environment interactions for weaning weight for Herefords in three countries. J Anim Sci. 2000, 78: 2121-2126.PubMedGoogle Scholar
- Pollak EJ, Quaas RL: Multibreed genetic evaluations of beef cattle. Proc 6th World Congr Genet Appl Livest Prod. 1998, 23: 81-88.Google Scholar
- Powell RL, VanRaden PM: International dairy bull evaluations expressed on national, subglobal, and global scales. J Dairy Sci. 2002, 85: 1863-1868.PubMedView ArticleGoogle Scholar
- Khatkar MS, Thomson PC, Tammen I, Raadsma HW: Quantitative trait loci mapping in dairy cattle: review and meta-analysis. Genet Sel Evol. 2004, 36: 163-190.PubMedPubMed CentralView ArticleGoogle Scholar
- Polineni P, Aragonda P, Xavier SR, Furuta R, Adelson DL: The bovine QTL viewer: a web accessible database of bovine Quantitative Trait Loci. BMC Bioinformatics. 2006, 7: 283-PubMedPubMed CentralView ArticleGoogle Scholar
- Blott S, Kim JJ, Moisio S, Schmidt-Kuntzel A, Cornet A, Berzi P, Cambisano N, Ford C, Grisart B, Johnson D, et al: Molecular dissection of a quantitative trait locus: a phenylalanine-to-tyrosine substitution in the transmembrane domain of the bovine growth hormone receptor is associated with a major effect on milk yield and composition. Genetics. 2003, 163: 253-266.PubMedPubMed CentralGoogle Scholar
- Kambadur R, Sharma M, Smith TP, Bass JJ: Mutations in myostatin (GDF8) in double-muscled Belgian Blue and Piedmontese cattle. Genome Res. 1997, 7: 910-916.PubMedGoogle Scholar
- Cohen-Zinder M, Seroussi E, Larkin DM, Loor JJ, Everts-van der Wind A, Lee JH, Drackley JK, Band MR, Hernandez AG, Shani M, et al: Identification of a missense mutation in the bovine ABCG2 gene with a major effect on the QTL on chromosome 6 affecting milk yield and composition in Holstein cattle. Genome Res. 2005, 15: 936-944.PubMedPubMed CentralView ArticleGoogle Scholar
- Schnabel RD, Kim JJ, Ashwell MS, Sonstegard TS, Van Tassell CP, Connor EE, Taylor JF: Fine-mapping milk production quantitative trait loci on BTA6: Analysis of the bovine osteopontin gene. Proc Natl Acad Sci USA. 2005, 102: 6896-6901.PubMedPubMed CentralView ArticleGoogle Scholar
- Winter A, Kramer W, Werner FA, Kollers S, Kata S, Durstewitz G, Buitkamp J, Womack JE, Thaller G, Fries R: Association of a lysine-232/alanine polymorphism in a bovine gene encoding acyl-CoA:diacylglycerol acyltransferase (DGAT1) with variation at a quantitative trait locus for milk fat content. Proc Natl Acad Sci USA. 2002, 99: 9300-9305.PubMedPubMed CentralView ArticleGoogle Scholar
- Meuwissen TH, Hayes BJ, Goddard ME: Prediction of total genetic value using genome-wide dense marker maps. Genetics. 2001, 157: 1819-1829.PubMedPubMed CentralGoogle Scholar
- Krzywinski M, Wallis J, Gosele C, Bosdet I, Chiu R, Graves T, Hummel O, Layman D, Mathewson C, Wye N, et al: Integrated and sequence-ordered BAC- and YAC-based physical maps for the rat genome. Genome Res. 2004, 14: 766-779.PubMedPubMed CentralView ArticleGoogle Scholar
- Marra M, Kucaba T, Sekhon M, Hillier L, Martienssen R, Chinwalla A, Crockett J, Fedele J, Grover H, Gund C, et al: A map for sequence analysis of the Arabidopsis thaliana genome. Nat Genet. 1999, 22: 265-270.PubMedView ArticleGoogle Scholar
- McPherson JD, Marra M, Hillier L, Waterston RH, Chinwalla A, Wallis J, Sekhon M, Wylie K, Mardis ER, Wilson RK, et al: A physical map of the human genome. Nature. 2001, 409: 934-941.PubMedView ArticleGoogle Scholar
- Gregory SG, Sekhon M, Schein J, Zhao S, Osoegawa K, Scott CE, Evans RS, Burridge PW, Cox TV, Fox CA, et al: A physical map of the mouse genome. Nature. 2002, 418: 743-750.PubMedView ArticleGoogle Scholar
- Barendse W, Vaiman D, Kemp SJ, Sugimoto Y, Armitage SM, Williams JL, Sun HS, Eggen A, Agaba M, Aleyasin SA, et al: A medium-density genetic linkage map of the bovine genome. Mamm Genome. 1997, 8: 21-28.PubMedView ArticleGoogle Scholar
- Ihara N, Takasuga A, Mizoshita K, Takeda H, Sugimoto M, Mizoguchi Y, Hirano T, Itoh T, Watanabe T, Reed KM, et al: A comprehensive genetic map of the cattle genome based on 3802 microsatellites. Genome Res. 2004, 14: 1987-1998.PubMedPubMed CentralView ArticleGoogle Scholar
- Snelling WM, Casas E, Stone RT, Keele JW, Harhay GP, Bennett GL, Smith TP: Linkage mapping bovine EST-based SNP. BMC Genomics. 2005, 6: 74-PubMedPubMed CentralView ArticleGoogle Scholar
- Everts-van der Wind A, Larkin DM, Green CA, Elliott JS, Olmstead CA, Chiu R, Schein JE, Marra MA, Womack JE, Lewin HA: A high-resolution whole-genome cattle-human comparative map reveals details of mammalian chromosome evolution. Proc Natl Acad Sci USA. 2005, 102: 18526-18531.PubMedView ArticleGoogle Scholar
- Itoh T, Watanabe T, Ihara N, Mariani P, Beattie CW, Sugimoto Y, Takasuga A: A comprehensive radiation hybrid map of the bovine genome comprising 5593 loci. Genomics. 2005, 85: 413-424.PubMedView ArticleGoogle Scholar
- Jann OC, Aerts J, Jones M, Hastings N, Law A, McKay S, Marques E, Prasad A, Yu J, Moore SS, et al: A second generation radiation hybrid map to aid the assembly of the bovine genome sequence. BMC Genomics. 2006, 7: 283-PubMedPubMed CentralView ArticleGoogle Scholar
- Schiex T, Chabrier P, Bouchez M, Milan D: Boosting EM for radiation hybrid and genetic mapping. Proceedings of the First International Workshop on Algorithms in Bioinformatics. Edited by: Gascuel O, Moret BME. 2001, London, UK: Springer-Verlag, 2149: 41-51. [Lecture Notes In Computer Science].View ArticleGoogle Scholar
- Children's Hospital Oakland Research Institute BACPAC Resources Center. [http://bacpac.chori.org/]
- Warren W, Smith TP, Rexroad CE, Fahrenkrug SC, Allison T, Shu CL, Catanese J, de Jong PJ: Construction and characterization of a new bovine bacterial artificial chromosome library with 10 genome-equivalent coverage. Mamm Genome. 2000, 11: 662-663.PubMedView ArticleGoogle Scholar
- Cai L, Taylor JF, Wing RA, Gallagher DS, Woo SS, Davis SK: Construction and characterization of a bovine bacterial artificial chromosome library. Genomics. 1995, 29: 413-425.PubMedView ArticleGoogle Scholar
- Schibler L, Roig A, Mahe MF, Save JC, Gautier M, Taourt S, Boichard D, Eggen A, Cribiu EP: A first generation bovine BAC-based physical map. Genet Sel Evol. 2004, 36: 105-122.PubMedPubMed CentralView ArticleGoogle Scholar
- Ng SH, Artieri CG, Bosdet IE, Chiu R, Danzmann RG, Davidson WS, Ferguson MM, Fjell CD, Hoyheim B, Jones SJ, et al: A physical map of the genome of Atlantic salmon, Salmo salar. Genomics. 2005, 86: 396-404.PubMedView ArticleGoogle Scholar
- Lander ES, Linton LM, Birren B, Nusbaum C, Zody MC, Baldwin J, Devon K, Dewar K, Doyle M, Fitzhugh W, et al: Initial sequencing and analysis of the human genome. Nature. 2001, 409: 860-921.PubMedView ArticleGoogle Scholar
- UCSC Genome Bioinformatics, Sequence and Annotation Downloads. [http://hgdownload.cse.ucsc.edu/downloads.html]
- Bovine BAC Map in FPC Format. [ftp://ftp.bcgsc.bc.ca/pub/bovine_mapping/]
- The Bovine Genome Database Home Page (contains links to BAC and Composite map Gbrowse sites). [http://www.bovinegenome.org]
- Bovine BAC Map Gbrowse (Texas A&M). [http://genomes.tamu.edu/cgi-bin/gbrowse/bacmap/]
- Bovine BAC Map Gbrowse (CSIRO). [http://www.livestockgenomics.csiro.au/perl/gbrowse.cgi/bacmap/]
- Soderlund C, Longden I, Mott R: FPC: a system for building contigs from restriction fingerprinted clones. Comput Appl Biosci. 1997, 13: 523-535.PubMedGoogle Scholar
- Soderlund C, Humphray , Dunham A, French L: Contigs built with fingerprints, markers, and FPC V4.7. Genome Res. 2000, 10: 1772-1787.PubMedPubMed CentralView ArticleGoogle Scholar
- Flibotte S, Chiu R, Fjell C, Krzywinski M, Schein JE, Shin H, Marra MA: Automated ordering of fingerprinted clones. Bioinformatics. 2004, 20: 1264-1271.PubMedView ArticleGoogle Scholar
- Sulston J, Mallett F, Staden R, Durbin R, Horsnell T, Coulson A: Software for genome mapping by fingerprinting techniques. Comput Appl Biosci. 1988, 4: 125-132.PubMedGoogle Scholar
- Everts-van der Wind A, Kata SR, Band MR, Rebeiz M, Larkin DM, Everts RE, Green CA, Liu L, Natarajan S, Goldammer T, et al: A 1463 gene cattle-human comparative map with anchor points defined by humangenome sequence coordinates. Genome Res. 2004, 14: 1424-1437.PubMedPubMed CentralView ArticleGoogle Scholar
- Bovine Composite Map Gbrowse (Texas A&M). [http://genomes.tamu.edu/cgi-bin/gbrowse/bosmap2/]
- Bovine Composite Map Gbrowse (CSIRO). [http://www.livestockgenomics.csiro.au/perl/gbrowse.cgi/bosmap/]
- McKay SD, Schnabel RD, Murdoch BM, Aerts J, Gill C, Gao C, Matukumalli LK, Stothard P, Wang Z, Van Tassell CP, et al: High-throughput genotyping facilitates rapid construction of whole genome radiation hybrid and linkage maps. Anim Genet. 2007, 38: 120-125.PubMedPubMed CentralView ArticleGoogle Scholar
- Williams JL, Eggen A, Ferretti L, Farr CJ, Gautier M, Amati G, Ball G, Caramorr T, Critcher R, Costa S, et al: A bovine whole-genome radiation hybrid panel and outline map. Mamm Genome. 2002, 13: 469-474.PubMedView ArticleGoogle Scholar
- de Givry S, Bouchez M, Chabrier P, Milan D, Schiex T: CARHTAGENE: multipopulation integrated genetic and radiation hybrid mapping. Bioinformatics. 2005, 21: 1703-1704.PubMedView ArticleGoogle Scholar
- Snelling WM, Gautier M, Keele JW, Smith TP, Stone RT, Harhay GP, Bennett GL, Ihara N, Takasuga T, Takeda H, et al: Integrating linkage and radiation hybrid mapping data for bovine chromosome 15. BMC Genomics. 2004, 5: 77-PubMedPubMed CentralView ArticleGoogle Scholar
- Human Genome Sequencing Center at Baylor College of Medicine, Bovine Genome Project. [http://www.hgsc.bcm.tmc.edu/projects/bovine/]
- Liao W, Collins A, Hobbs M, Khatkar MS, Luo J, Nicholas FW: A comparative location database (CompLDB): map integration within and between species. Mamm Genome. 2007, 18: 287-299.PubMedView ArticleGoogle Scholar
- Schuler GD: Sequence mapping by electronic PCR. Genome Res. 1997, 7: 541-550.PubMedPubMed CentralGoogle Scholar
- Kent WJ: BLAT - the BLAST-like alignment tool. Genome Res. 2002, 12: 656-664.PubMedPubMed CentralView ArticleGoogle Scholar
- Havlak P, Chen R, Durbin KJ, Egan A, Ren Y, Song XZ, Weinstock GM, Gibbs RA: The Atlas genome assembly system. Genome Res. 2004, 14: 721-732.PubMedPubMed CentralView ArticleGoogle Scholar
- Stein LD, Mungall CJ, Shu S, Caudy M, Mangone M, Day A, Nickerson E, Stajich J, Harris TW, Arva A, Lewis S: The generic genome browser: a building block for a model organism system database. Genome Res. 2002, 12: 1599-1610.PubMedPubMed CentralView ArticleGoogle Scholar
- Bovine QTL Viewer. [http://bovineqtl.tamu.edu/]
- Gregory TR, Nicol JA, Tamm H, Kullman B, Kullman K, Leitch IJ, Murray BG, Kapraun DF, Greilhuber J, Bennett MD: Eukaryotic genome size databases. Nucleic Acids Res. 2007, 35: D332-338.PubMedPubMed CentralView ArticleGoogle Scholar
- Animal Genome Size Database. [http://www.genomesize.com/]
- Laurent V, Wajnberg E, Mangin B, Schiex T, Gaspin C, Vanlerberghe-Masutti F: A composite genetic map of the parasitoid wasp Trichogramma brassicae based on RAPD markers. Genet Soc Am. 1998, 150: 275-282.Google Scholar
- Doligez A, Adam-Blondon AF, Cipriani G, Di Gaspero G, Laucou V, Merdinoglu D, Meredith CP, Riaz S, Roux C, This P: An integrated SSR map of grapevine based on five mapping populations. Theor Appl Genet. 2006, 113: 369-382.PubMedView ArticleGoogle Scholar
- Piquemal J, Cinquin E, Couton F, Rondeau C, Seignoret E, Doucet I, Perret D, Villeger MJ, Vincourt P, Blanchard P: Construction of an oilseed rape (Brassica napus L.) genetic map with SSR markers. Theor Appl Genet. 2005, 111: 1514-1523.PubMedView ArticleGoogle Scholar
- Pelgas B, Bousquet J, Beauseigle S, Isabel N: A composite linkage map from two crosses for the species complex Picea mariana × Picea rubens and analysis of synteny with other Pinaceae. Theor Appl Genet. 2005, 111: 1466-1488.PubMedView ArticleGoogle Scholar
- Demeure O, Pomp D, Milan D, Rothschild MF, Tuggle CK: Mapping of 443 porcine EST improves the comparative maps for SSC1 and SSC7 with the human genome. Anim Genet. 2005, 36: 381-389.PubMedView ArticleGoogle Scholar
- Stam P: Construction of integrated genetic linkage maps by means of a new computer package: JoinMap. Plant J. 1993, 3: 739-744.View ArticleGoogle Scholar
- Agarwala R, Applegate DL, Maglott D, Schuler GD, Schaffer AA: A fast and scalable radiation hybrid map construction and integration strategy. Genome Res. 2000, 10: 350-364.PubMedPubMed CentralView ArticleGoogle Scholar
- Ben-Dor A, Chor B: On constructing radiation hybrid maps. J Comp Biol. 1997, 4: 517-533.View ArticleGoogle Scholar
- Freeman JL, Perry GH, Feuk L, Redon R, McCarroll SA, Altshuler DM, Aburatani H, Jones KW, Tyler-Smith C, Hurles ME, et al: Copy number variation: New insights in genome diversity. Genome Res. 2006, 16: 949-961.PubMedView ArticleGoogle Scholar
- McRae AF, Beraldi D: Examination of a region showing linkage map discrepancies across sheep breeds. Mamm Genome. 2006, 17: 346-353.PubMedView ArticleGoogle Scholar
- Tuzun E, Sharp AJ, Bailey JA, Kaul R, Morrison VA, Pertz LM, Haugen E, Hayden H, Albertson D, Pinkel D, et al: Fine-scale structural variation of the human genome. Nat Genet. 2005, 37: 727-732.PubMedView ArticleGoogle Scholar
- Feuk L, Carson AR, Scherer SW: Structural variation in the human genome. Nature Rev Genet. 2006, 7: 85-97.PubMedView ArticleGoogle Scholar
- Wright S: Coefficients of inbreeding and relationship. Am Nat. 1922, 56: 330-338.View ArticleGoogle Scholar
- Dalrymple BP, Kirkness EF, Nefedov M, McWilliam S, Ratnakumar A, Barris W, Zhao S, Shetty J, Maddox JF, O'Grady M, et al: Using comparative genomics to reorder the human genome sequence into a virtual sheep genome. Genome Biol. 2007, 8: R152-PubMedPubMed CentralView ArticleGoogle Scholar
- Bovine Genome Assembly Gbrowse. [http://www.livestockgenomics.csiro.au/perl/gbrowse.cgi/bova3/]
- Marra MA, Kucaba TA, Dietrich NL, Green ED, Brownstein B, Wilson RK, McDonald KM, Hillier LW, McPherson JD, Waterston RH: High throughput fingerprint analysis of large-insert clones. Genome Res. 1997, 7: 1072-1084.PubMedPubMed CentralGoogle Scholar
- Schein J, Kucaba T, Sekhon M, Smailus D, Waterston R, Marra M: High-throughput BAC fingerprinting. Methods Mol Biol. 2004, 255: 143-156.PubMedGoogle Scholar
- Sulston J, Mallett F, Durbin R, Horsnell T: Image analysis of restriction enzyme fingerprint autoradiograms. Comput Appl Biosci. 1989, 5: 101-106.PubMedGoogle Scholar
- The Fingerprint Image Analysis System. [http://www.sanger.ac.uk/Software/Image/]
- Fuhrmann DR, Krzywinski MI, Chiu R, Saeedi P, Schein JE, Bosdet IE, Chinwalla A, Hillier LW, Waterston RH, McPherson JD, et al: Software for automated analysis of DNA fingerprinting gels. Genome Res. 2003, 13: 940-953.PubMedPubMed CentralView ArticleGoogle Scholar
- Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ: Basic local alignment search tool. J Mol Biol. 1990, 215: 403-410.PubMedView ArticleGoogle Scholar
- TIGR Cattle Gene Index. [ftp://occams.dfci.harvard.edu/pub/bio/tgi/data/Bos_taurus/BTGI.release_11.zip]
- Bos taurus UniGene Repository. [ftp://ftp.ncbi.nih.gov/repository/UniGene/Bos_taurus/]
- Laboratory of Phil Green, Genome Sciences Department, University of Washington. [http://www.phrap.org]
This article is published under license to BioMed Central Ltd. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.