- Research article
- Open access
- Published:
Re-annotation of the physical map of Glycine max for polyploid-like regions by BAC end sequence driven whole genome shotgun read assembly
BMC Genomics volume 9, Article number: 323 (2008)
Abstract
Background
Many of the world's most important food crops have either polyploid genomes or homeologous regions derived from segmental shuffling following polyploid formation. The soybean (Glycine max) genome has been shown to be composed of approximately four thousand short interspersed homeologous regions with 1, 2 or 4 copies per haploid genome by RFLP analysis, microsatellite anchors to BACs and by contigs formed from BAC fingerprints. Despite these similar regions,, the genome has been sequenced by whole genome shotgun sequence (WGS). Here the aim was to use BAC end sequences (BES) derived from three minimum tile paths (MTP) to examine the extent and homogeneity of polyploid-like regions within contigs and the extent of correlation between the polyploid-like regions inferred from fingerprinting and the polyploid-like sequences inferred from WGS matches.
Results
Results show that when sequence divergence was 1–10%, the copy number of homeologous regions could be identified from sequence variation in WGS reads overlapping BES. Homeolog sequence variants (HSVs) were single nucleotide polymorphisms (SNPs; 89%) and single nucleotide indels (SNIs 10%). Larger indels were rare but present (1%). Simulations that had predicted fingerprints of homeologous regions could be separated when divergence exceeded 2% were shown to be false. We show that a 5–10% sequence divergence is necessary to separate homeologs by fingerprinting. BES compared to WGS traces showed polyploid-like regions with less than 1% sequence divergence exist at 2.3% of the locations assayed.
Conclusion
The use of HSVs like SNPs and SNIs to characterize BACs wil improve contig building methods. The implications for bioinformatic and functional annotation of polyploid and paleopolyploid genomes show that a combined approach of BAC fingerprint based physical maps, WGS sequence and HSV-based partitioning of BAC clones from homeologous regions to separate contigs will allow reliable de-convolution and positioning of sequence scaffolds (see BES_scaffolds section of SoyGD). This approach will assist genome annotation for paleopolyploid and true polyploid genomes such as soybean and many important cereal and fruit crops.
Background
Soybean (Glycine max) is the second most valuable crop in the U.S., accounting for $12–17 billion in annual revenue (USDA-NASS Agricultural Statistics 2000–2007). Genomics has had a profound effect on plant biology, but the impact on major crop species such as soybean remains limited to a few marker characterized disease resistant germplasm releases [1, 2]. A primary difficulty is that the soybean genome is 4–10 times larger than the model plants Arabidopsis thaliana, Medicago truncatulata or Lotus japonicus. Further, the soybean genome shows evidence of a paleopolyploid origin with gene-rich islands that were highly conserved following duplication [3, 4].
Shultz et al., [4] used BAC fingerprint derived contig clone density to estimate that 25–30% of the genome was highly conserved after both duplications, leading to 50–60% of the genome existing in a two- or four-copy state. That conclusion was supported by the gene number in gene families inferred from EST hybridizations to BAC minimum tile paths (MTPs) [5]. Ultimately, Shultz et al., [4] predicted the genome could be resolved into about four thousand segments (each about 150–350 Kbp in size) that differed in copy number per haploid genome. The regions appear interspersed at random, with no evidence for conserved neighbor relationships.
Toward the end of developing a complete map describing where duplicated regions were located, contigs representing each of the genomic segments were rebuilt at high stringency and a minimum number of merges allowed [4]. Despite the high stringency, homeologous regions coalesced to single contigs. Consequently, each contig was measured for the number of BAC clones per unique DNA band. Six clones per unique band in a clone fingerprint was expected, yet regions of 12 and 24 clones per unique band were common. Since homeology could not be distinguished from over-representation of regions in the BAC libraries, contigs were labeled to distinguish their expected copy number. The 2,408 contigs in the 1 to 3,500 series were expected to be largely single copy (1,092 numbers were removed when contig merges were made). The 240 contigs in the 8,000 to 8,999 series were predicted to be present in two copies and derive from the more recent tetraploidy event. Therefore, with further analyses the 8,000 series of contigs were each expected to be separated into two, resulting in 480 different regions [6]. The 406 contigs in the 9,000 to 9,999 series were predicted to be largely coalescences of 4 genomic regions derived from both the genome duplication and hybridization events that produced an octaploid-like genome (though an octaploid-like soybean may never have existed since the two events were separated by millions of years). With further analyses, contigs containing clones from 4 genomic regions were expected to separate into 1,624 different regions. In total, 2,104 multi-copy regions and 2,408 single-copy regions were expected.
DNA markers that anchored the soybean physical map to the genetic map also showed evidence of variation in copy numbers derived from ancient ploidy shifts [4]. All RFLP markers hybridized to clones in two or more contigs. Even the majority (239/363) of microsatellite markers could generate amplicons from clones in two or more contigs. Markers were labeled with an alphabetic suffix, with -a the smallest amplicons, or band, -b the next smallest (sometimes up to -z in cases where many amplicons were found). Alignment of contigs with the genetic map using these anchors was error prone, requiring each marker anchored contig to be shown at each possible location.
To resolve the problem of genetic map placement, microsatellite markers derived from BAC end sequences were used to align contigs with the genetic map [7]. Here, only one outcome was expected, the placement of a single contig to a single location. Maps generated with the markers did show single locations, often in gaps in the existing maps [8]. There were 25,123 BES reads available from the physical map of the 'Forrest' cultivar of soybean that provided about two thousand potential satellite markers. These markers should be enough to locate and orientate every contig at a single map location. These markers, however, cannot separate the polyploid-like regions that are composed of nearly identical homeologous BACs as markers in these regions produce multiple amplicons. New approaches are needed to map these regions correctly.
SoyGD, based on a distributed annotation systems (DAS) called the generic model organism databases (e.g. GMOD) [9], was developed to show polyploid genomic regions to users (Figure 1). A track for duplications can be inserted into GBrowse [10], however, for highly duplicated genomes, new classifications and ontology for representing large-scale genome duplications had to be developed [4]. Exciting new tools that were developed to compare syntenic regions among genomes [11] demonstrate that the GMOD platform has the potential to display homeology.
The cultivars Forrest and 'Williams 82' provide a large set of useful genomic tools for soybean genomics [6, 12, 13]. The two cultivars can be thought models in the same way as are cultivars 'Columbia' and 'Landsberg erecta' to Arabidopsis thaliana, or 'Mo17' and 'B73' are to Zea mays. The soybean community is committed to advancing both resources, with Williams 82 as the lead for a complete genome sequence. In 2007 there were 7.4 million trace sequences at NCBI. Some preliminary sequence contigs with annotations for about 90% of the genome were available. Sequence contigs can be viewed on the genetic map at a new section of SoyGD [4]. These sequence resources represent tools for in silico biology that can resolve the physical map and de-convolute the complete genome sequence. Here, these resources were tested for usefulness as tools to determine if the existing contig annotations truly reflect genomic regions that are polyploidy-like, to identify HSVs that can distinguish homeologs within cultivars and to identify HSVs that can distinguish among soybean cultivars.
Results and discussion
Genomic regions
A MegaBlast of Forrest BES (Table 1) to Williams 82 WGS showed some homeologous regions had diverged significantly and were essentially single-copy (Figure 2). For example, one of the pair of BES from B47P08 (ctg 312; LG A1; Figure 1) showed only 90% identity between Forrest and Williams 82. Only one sequence read matched this BES in Williams 82 trace files, with about 60 SNPs and 6 SNIs. That degree of divergence between cultivars had been reported in sequenced AFLP bands [14, 15]. At this degree of divergence though, the Williams 82 sequences may represent a homeologous region. In that case the region being examined was not represented in the Williams 82 WGS. Under-representation may be due to cloning bias against some regions since the WGS is based on a single insertion site in a single high copy number vector.
Examination of the second BES from B47P08 (ctg 312; LGA1; Figure 1) showed three sequences with 99% identity between Forrest and Williams 82 (Figure 3). The differences represent probable SNPs between cultivars. However, in addition there were 43 sequences with >95% identity. Those homeologous sequences could be clustered into 4 different groups based on SNPs among HSVs (H-SNPs). Contig 321 appears to be octaploid-like at one end and diploid-like at the other end. Contigs of this type are dangerous to merge with neighboring contigs, as eight different regions might match, only one of which would be correct.
Contig 9077, also from LG A1 (Figure 1), is a potentially octaploid-like contig. Examination of the four BES from H53F21 (Figure 4 and not shown) and H51D13 (Figure 5 and not shown) showed evidence to support the octaploid-like nature of the contig. For example, at H53F21 there were three sequences with 99% identity between Forrest and Williams 82 (Figure 4). The differences among these three and the Forrest BES may represent SNIs between cultivars (few or no SNPs). However, in addition there were 22 sequences with >90% identity. Those clearly homeologous sequences could be clustered into 3 different groups based on H-SNPs.
At BES H51D13 (Figure 5), about 300 Kbp away on LG A1 (Figure 1), the genomic region was less well represented among trace files. There were three sequences with 99% identity between Forrest and Williams 82 (Figure 5). The differences represent probable SNIs between cultivars (few or no SNPs were found). However, in addition there were 5–6 different sequences with >90% identity. Those clearly homeologous sequences could be clustered into 3 different groups based on H-SNPs.
Fourteen BAC clones were chosen from contig 9077 and used for PCR amplification of the BES. Sequencing these amplicons revealed two sequences, each representing one of two homeologs mixed together throughout the contig (Figure 4; Panel C). The A type and the G type were present but the T type and the C type found in WGS were not present. Therefore, the third and fourth homeologs predicted to be in the contig by WGS to BES alignments could not be distinguished by the >600 bp of DNA sequence. The G type and A type clones can each be used to form a new contig. The SNHs will be used to split ctg9077 (Figure 1) in two. Map locations for the split contigs may be determined if cultivar differences can be found linked closely to the HSVs among common mapping population parents.
Of further note, micro-satellite marker Sat_368 anchoring contig 9077 was on the G-type clone ISO56K20. Sat_368 did not appear to have any close homeologs (Additional File 1). Therefore, the octaploid-like regions can be quite heterogeneous across contig-sized regions and suggests diploidization acts on regions less than the size of a BAC clone.
A similar pattern was observed on LG G (Figure 6). Examination of the pair of BES from H77P02 (Figure 7) in Contig 9354 shows that four genomic regions potentially coalesced and indicates the octaploid-like nature of the contig. For example, at BES SOYFK12TH there were three sequences with 99% identity between Forrest and Williams 82 (Figure 7; Panel A). The differences represent probable SNIs between cultivars (no SNPs). In addition, there were 65 homeologous sequences with >95% identity which could be clustered into 4–9 different groups based on H-SNPs (Figure 7; Panel B).
Even the satellite marker Satt688, anchoring BAC H77P02 to contig 9354 and the genetic map showed evidence for conserved homeologs (Figure 8). There were four different HSV sequences with 99% identity. Differences among homeologs were evident as 1 (trace file -021), 2 (trace file -796) and 11 (trace file -795) probable SNPs compared to the trace file -270. None of the four genomic regions was adequately represented among the trace files, suggesting a cloning bias against each. An alternate hypothesis is that the sequence variations are errors in the reads of a single region, which is unlikely given the genomic context of the marker.
Whole genome comparisions
Because BAC clones were selected from the MTP to generate BES, there are many fewer BES from polyploid-like regions than diploidized regions (Table 2). Nonetheless, there are sufficient numbers to examine the usefulness ploidy annotations found at SoyGD. Allowing for 90% identity, the mean number of homeologs inferred from the BES to WGS matches was 8.4 for octaploid-like contigs; 4.5 for tetraploid-like contigs and 1.5 for diploidized contigs. Those means were in good agreement with those inferred from clones per unique band metric used to label contigs in the physical map as potentially repeated and octaploid-like [4]. At 95% identity there were about half as many matches on average, suggesting divergence following the most recent genome duplication has been less than after the first event. At 98 and 99% identity there was evidence for sequence pairs that share significant identity in the polyploid-like regions but not in the diploidized regions.
Genes and markers
BAC end sequences anchored to a robust physical map are important tools for genome analysis. BES have been developed from MTP2BH, MTP4BH and MTPE4 (Table 1). Enquiries to GenBank nr and pat databases identified 12,919 potentially geneic homologs. Analysis of the locations of the inferred genes showed evidence of gene rich islands on each chromosome (Figure 1; Figure 6).
Eighty one homologs of DNA markers found in genetic maps were detected in the BES, i.e. forty two BES's contained sequence highly homologous (over 80–341 bp from e-30 to e-300) to 80 different genetic markers (20 RFLPs, 61 microsatellites), or about 4% of the markers with sequences in GenBank. About three thousand new microsatellite markers were identified within the whole BES collection.
SNPs among the HSVs were found in nearly every BES examined. SNIs among the HSVs were found among 24% of sequences (Figure 2, 3, 4, 5, 6, 7, 8). Clones in plates 11 and 12 were re-sequenced and so have 2 records for each BAC end in GenBank. Re-sequenced clones help determine the sequence error rate and greatly facilitate SNP detection. Along with the few clones tested directly by mapping (data not shown), about 67.5% of SNPs and SNIs detected in single pass sequence are expected to be validated [8].
Conclusion
The comparison of Forrest and Williams 82 sequences represents a powerful tool for soybean geneticists. There are abundant SNPs and SNIs among the sequences, with many linked to predicted gene sequences (Table 1). The high frequency of single nucleotide changes between genomic regions of soybean cultivars has been reported previously [14, 15] and stands in contrast to the very low frequency between ESTs [16]. Clearly, further investment in genomic SNP identification is called for. MTP BES [17] make an excellent starting point, providing markers spaced at regular intervals in the genome.
In comparison to SNIs, indels larger than 2 bp are very rare. This bias against indels may explain why RFLPs and AFLPs are rare in soybean [3, 14]. Further the scarcity of indels will have contributed to the inability of FPC to separate BACs into different contigs, once their sequence identity exceeded 90% [4, 18]. The use of SNPs and SNIs to characterize BACs will improve contig building methodology. For example, plate 13 of MTP4BH was developed from just 6 octaploid-like contigs by picking redundant clones from putatively octaploid-like contigs [4]. This set of 748 sequences should resolve into 48 regions when the genome sequence is properly de-convoluted.
Bioinformatic and functional annotation of polyploid genomes can be greatly improved using a combination of BAC fingerprint based physical maps, WGS sequence and HSV partitioning of BAC clones in polyploid regions. The separation of contigs will allow the de-convolution of sequence and allow whole genome annotation in polyploids. Preliminary results from stringent BLAT analysis of BES to sequence scaffolds can be viewed at SoyGD [4]
Major challenges will have to be overcome in assigning function to the duplicated regions. Reverse genetic approaches like gene silencing and mutation would be expected to be effective only in certain small gene families and particular genomic regions. Gene silencing should work when duplicated genes of redundant function are close enough in sequence to be inhibited by the same probe [19]. During over-expression [20], the co-suppression response of the endogenous gene family will have to be considered. Will co-suppression actually reduce the activity of the members of the gene family in patterns not predicted by the experimenters? In the case of the identification of mutations for loss of function by TILLING [21–23], the functions of the homeologous genes must have been sufficiently diverged over evolutionary time for success to be expected. Secondly, no aneupleurotic pathways with functional redundancy must exist. The physical map should be used to guide these approaches. A complete map of homeologous regions can help identify genes in regions likely to be unique, single copy, and others likely to be redundant in 2 or 4 copies like an allo-polyploid.
Methods
Source of sequences
Forrest genome resources used included all three MTPs described in Table 1. There were 13,473 BES reads from MTP2 (CG826126 to CG812653). There were 7,700 BES reads from MTP4BH (DX406713 to DX414412) and 3,324 reads from MTP4E (ER962965 to ER966289). After trimming, the mean read length for these BES was about 736 bp. The total sequence generated was 18.5 Mbp, or about 2% of the soybean genome. There were 9,386 paired, forward and reverse reads.
At the time of enquiry (mid 2007), there were 7.4 million reads of Williams 82 genome reported in the trace sequence section of NCBI [29]. The total amount of sequence was 6,000 Mbp, about six fold the soybean genome. Most were paired forward and reverse reads from 2–3 kb inserts. These sequences were not trimmed and most contained 50–60 bp of sequence from pUC18 at the start of the sequence. About 36,000 reads had another tract from pUC18 at the end of the sequence.
In silico polymorphism detections
MegaBlast enquiries were made of the Glycine max WGS database using individual BES [29]. Criteria set were; database Glycine max-WGS; hits computed 250; all low complexity filter selected; expect was set to 10; word size was set to 32 or 64; percent identity used was normally 90% though 99%, 98%, 95%, and 85% were manually tested in instances noted in the results.
Results were assembled into groups of 100 by expected copy number and 600 were examined by a manual editor. Distance trees of the results were selected (some some typical result trees were captured as screen shots (Figures 2, 3, 4). "Show multiple alignment" was selected from the root of the tree. Multiple alignments were examined for the presence of HSVs, SNPs, SNIs and SSRs and illustrative examples used to make Figures 2, 3, 4, 5, 7 and 8. For the means in Table 2, multiple alignment was selected directly and the results automatically recorded.
In vitro HSV polymorphism detection
BACs from homeologous regions that assembled into single contigs were picked from BAC library master plates. DNA was extracted as previously described [4, 17]. Primers were designed from within the BES to encompass HSV. Settings used for primer design were Tm 55° ± 1°C, amplicons 100–500 bp, primer length 20 ± 2 bp. No constraints on GC% were set to avoid potential bias against the AT rich regions of the soybean genome. Repeated DNA amplicons (mini-satellites, transposons etc.) were filtered out by Blast searching, unlike Shultz et al., [7]. Primers were obtained from Sigma Genosys (Woodlands, TX).
Polymerase chain reaction (PCR) was performed in a PE 9700 (Boston, MA). An initial 95°C denaturation for 5 min was followed by 30 cycles of 95° for 30 s, 55° for 30 s, and 72° for 30 s. After PCR was complete, gel electrophoresis was performed in a 2% (w/v) agarose gel or a 4% (w/v) PAGE stained with ethidium bromide and amplicon documented using a BioRad GelDoc (Hercules, CA) system. Bands were isolated in pGEM3T. SNP polymorphism was identified by DNA sequencing of PCR amplicons following plasmid isolation using a CEQ2000 (Beckman Coulter, Fullerton, CA).
Annotation and map representation
All potential SNPs, SNIs and microsatellites that distinguish either cultivars or homeologs were named with the SIUC_ suffix (at each database entry and first mention in the text) followed by N-, I- or S- prefix. For cultivar polymorphisms this was followed by the motif and BAC of origin. For HSVs the letter H- was suffixed, then followed by the motif and BAC of origin. In contrast, earlier markers were assigned a sequential number [30–33]. The altered naming convention used here was designed to help users find the clone of origin in the physical map. All potential markers will be shown at SoyGD in the BES_SSR, BES_SNP or BES_SNI track (not shown). Markers that have been located in the genetic map by DNA polymorphism scored in RIL populations will be shown on the locus track.
Abbreviations
- BAC:
-
bacterial artificial chromosome
- BES:
-
BAC end sequence
- bp:
-
base pair
- contig:
-
contiguous set of overlapping clones
- EST:
-
expressed sequence tag
- HSV:
-
homeolog sequence variant
- LIS:
-
legume information system
- MTP:
-
minimum tiling path
- NCBI:
-
National Center for Biotechnology Information
- RFLP:
-
restriction fragment length polymorphism
- SNH:
-
single nucleotide polymorphism between homeologs
- SNI:
-
single nucleotide insertion
- SNP:
-
single nucleotide polymorphism between alleles
- SSR:
-
simple sequence repeat.
References
Schmidt ME, Suttner RJ, Klein J, Gibson PT, Lightfoot DA, Myers O: Registration of LS-G96 soybean germplasm resistant to soybean sudden death syndrome and soybean cyst nematode race 3. Crop Sci. 1999, 39: 598-
Concibido VC, Diers , Brian W, Arelli PR: A Decade of QTL Mapping for Cyst Nematode Resistance in Soybean. Crop Sci. 2004, 44: 1121-1131.
Shoemaker RC, Polzin K, Labate J, Specht J, Brummer EC, Olson T, Young N, Concibido V, Wilcox J, Tamulonis JP, Kochert G, Boerma HR: Genome duplication in soybean Glycine subgenus soja. Genetics. 1996, 144: 329-338.
Shultz JL, Kurunam DJ, Shopinski KL, Iqbal MJ, Kazi S, Zobrist K, Bashir R, Yaegashi S, Lavu N, Afzal A, Yesudas CR, Kassem MA, Wu C, Zhang HB, Town CD, Meksem K, Lightfoot DA: The Soybean genome database (SoyGD): A browser for display of duplicated, polyploid regions and sequence tagged sites on the integrated physical and genetic maps of Glycine max. Nucleic Acids Res. 2006, 34: D758-765. 10.1093/nar/gkj050. http://soybeangenome.siu.edu/cgi-bin/gbrowse/BES_scaffolds for Forrest BES to scaffolds; http://soybeangenome.siu.edu/cgi-bin/gbrowse/SSR_scaffolds for scaffold to SSR marker map and http://bioinformatics.siu.edu for the Forrest physical map,
Shopinski K, Iqbal J, Afzal J, Shultz J, Jayaraman D, Lightfoot DA: Development of a pooled probe method for locating small gene families in a physical map of soybean using stress related paralogues and a BAC minimum tile path. Plant Methods. 2006, 2: 20-28. 10.1186/1746-4811-2-20.
Lightfoot DA: Soybean Genomics: Developments Through the Use of Cultivar Forrest. Inter J Plant Genom. 2008, 1-22. 10.1155/2008/793158.
Shultz J, Kazi S, Bashir R, Afzal J, Lightfoot DA: The development of BAC-end sequence-based microsatellite markers and placement in the physical and genetic maps of soybean. Theor Appl Genet. 2007, 114: 1081-1090. 10.1007/s00122-007-0501-9.
Bashir R: Minimum tile derived microsatellite markers improve the physical map of the soybean genome and the Essex by Forrest genetic map. MS PSAS, SIUC. 2007, 258-
Stein LD, Mungall C, Shu S, Caudy M, Mangone M, Day A, Nickerson E, Stajich JE, Harris TW, Arva A: The generic genome browser: A building block for a model organism system database. Genome Res. 2002, 12: 1599-1610. 10.1101/gr.403602.
Cheung J, Wilson MD, Zhang J, Khaja R, MacDonald JR, Heng HH, Koop BF, Scherer SW: Recent segmental and gene duplications in the mouse genome. Genome Biol. 2003, 4: R47-R52. 10.1186/gb-2003-4-8-r47.
Pan X, Stein L, Brendel V: SynBrowse: a synteny browser for comparative sequence analysis. Bioinformatics. 2005, 21: 3461-3468. 10.1093/bioinformatics/bti555.
Lightfoot DA, Njiti VN, Gibson PT, Kassem MA, Iqbal JM, Meksem K: Registration of the Essex by Forrest recombinant inbred line mapping population. Crop Sci. 2005, 45: 1678-1681. 10.2135/cropsci2004.0279.
Jackson SA, Rokhsar D, Stacey G, Shoemaker RC, Schmutz J, Grimwood J: Toward a reference sequence of the soybean genome: A multiagency effort. Crop Sci. 2006, 46: S55-S61.
Meksem K, Ruben E, Hyten D, Schmidt M, Lightfoot DA: High-throughput detection of polymorphism physically linked soybean cyst nematode resistance gene Rhg4 using Taqman Probes. Molec Breed. 2001, 7: 63-71. 10.1023/A:1009610009663.
Meksem K, Ruben E, Hyten D, Triwitayakorn K, Lightfoot DA: Conversion of AFLP bands to high-throughput DNA markers. Mol Gen Genomics. 2001, 265: 207-214. 10.1007/s004380000418.
Zhu YL, Song QJ, Hyten DL, Van Tassell CP, Matukumalli LK, Grimm DR, Hyatt SM, Fickus EW, Young ND, Cregan PB: Single-nucleotide polymorphisms in soybean. Genetics. 2003, 163: 1123-34.
Shultz JL, Yesudas Charles, Satsuki Yaegashi, Jawaad Afzal, Samreen Kazi, Lightfoot David: Three minimum tile paths from bacterial artificial chromosome libraries of the soybean (Glycine max cv. 'Forrest'): Tools for, structural and functional genomics. Plant Methods. 2006, 2: 9-18. 10.1186/1746-4811-2-9.
Wu CS, Sun P, Nimmakayala P, Santos FA, Meksem K, Springman R, Ding K, Lightfoot DA, Zhang HB: A BAC- and BIBAC-based physical map of the soybean genome. Genome Res. 2004, 14: 319-26. 10.1101/gr.1405004.
Wesley SV, Helliwell CA, Smith NA, Wang M, Rouse DT, Liu Q, Gooding PS, Singh SP, Abbott D, Stoutjesdijk PA, Robinson SP, Gleave AP, Green AG, Waterhouse PM: Construct design for efficient, effective and high-throughput gene silencing in plants. Plant J. 2001, 27: 581-590. 10.1046/j.1365-313X.2001.01105.x.
Collier R, Burgwyn B, Walter N, Lutke K, Taylor CG: Ex vitro composite plants: an inexpensive, rapid method for root biology. Plant J. 2005, 43: 449-457. 10.1111/j.1365-313X.2005.02454.x.
Bouchez N, Bouchez D: Arabidopsis gene knockout: phenotypes wanted. Curr Opin Plant Biol. 2003, 4: 111-117. 10.1016/S1369-5266(00)00145-X.
Nawy T, Lee JY, Colinas J, Wang JY, Thongrod SC, Malamy JE, Birnbaum K, Benfey PN: Transcriptional profile of the Arabidopsis root quiescent center. Plant Cell. 2005, 17: 1908-25. 10.1105/tpc.105.031724.
Meksem K: Soybean Tilling Project. 2007, [http://www.soybeantilling.org/tilling.jsp]
Yuan J, Njiti VN, Meksem K, Iqbal MJ, Triwitayakorn K, Kassem MA, Davis GT, Schmidt ME, Lightfoot DA: Quantitative trait loci in two soybean recombinant inbred line populations segregating for yield and disease resistance. Crop Sci. 2002, 42: 271-277.
Kazi S, Shultz JL, Bashir R, Afzal J, Njiti VN, Lightfoot DA: Separate loci underlie resistance to soybean sudden death syndrome in 'Hartwig' by 'Flyer'. Theoretical and Applied Genetics. 2008, 116: 967-977. 10.1007/s00122-008-0728-0.
Iqbal J, Meksem K, Njiti V, Kassem M, Lightfoot DA: Microsatellite markers identify three additional quantitative trait loci for resistance to soybean sudden death syndrome (SDS) in Essex x Forrest RILs. Theor Appl Genet. 2001, 102: 187-192. 10.1007/s001220051634.
Kassem MA, Shultz J, Meksem K, Wood AJ, Iqbal MJ, Lightfoot DA: An updated 'Essex' by 'Forrest' linkage map and first composite interval map of QTL underlying six soybean traits. Theor Appl Genet. 2006, 113: 1015-1026. 10.1007/s00122-006-0361-8.
Njiti VN, Meksem K, Iqbal MJ, Johnson JE, Zobrist KF, Kilo VY, Lightfoot DA: Common loci underlie field resistance to soybean sudden death syndrome in Forrest, Pyramid, Essex, and Douglas. Theor Appl Genet. 2002, 104: 294-300. 10.1007/s001220100682.
DOE's Williams 82 sequence assemblies at http://www.phytozome.net/soybean.php and Williams 82 soybean trace files at http://www.ncbi.nlm.nih.gov/blast/mmtrace.shtml
Cregan PB, Mudge J, Fickus EW, Danesh D, Denny R, Young ND: Two simple sequence repeat markers to select for soybean cyst nematode resistance conditioned by the rhg1 locus. Theor Appl Genet. 1999, 99: 811-818. 10.1007/s001220051300.
Meksem K, Zobrist K, Ruben E, Hyten D, Quanzhou T, Zhang HB, Lightfoot DA: Two large-insert soybean genomic libraries constructed in a binary vector: applications in chromosome walking and genome wide physical mapping. Theor Appl Genet. 2000, 101: 747-755. 10.1007/s001220051540.
Song QJ, Marek LF, Shoemaker RC, Lark KG, Concibido VC, Delannay X, Specht JE, Cregan PB: A new integrated genetic linkage map of the soybean. Theor Appl Genet. 2004, 109: 122-128. 10.1007/s00122-004-1602-3.
Triwitayakorn K, Njiti VN, Iqbal MJ, Yaegashi S, Town CD, Lightfoot DA: Genomic analysis of a region encompassing QRfs1 and QRfs2: genes that underlie soybean resistance to sudden death syndrome. Genome/Génome. 2005, 48: 125-138. 10.1139/g04-103.
Acknowledgements
The physical map was based upon work supported by the National Science Foundation under Grant No. 9872635. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation. The continued support of SIUC, College of Agriculture and Office of the Vice Chancellor for Research to DAL and the LTU College of Applied and Natural Sciences to JS is appreciated. The authors thank Dr. Q. Tao and Dr, H.B. Zhang for assistance with fingerprinting. We thank Dr. C. Town and Dr. C. Foo at TIGR for the BES and BLAT. We thank the "Soybean Genome Project, at DoE Joint Genome Institute" for release of the WGS reads and scaffolds. We thank the Government of India for support of the Fellowship for NS.
Author information
Authors and Affiliations
Corresponding author
Additional information
Authors' contributions
DAL and NS conceived of the study and executed all of the analysis. JS classified the contigs for ploidy based on clones per unique band. DAL drafted the manuscript. NS and JS provided critical review, interpretation of results and funding. All authors read and approved the final manuscript.
Electronic supplementary material
12864_2007_1516_MOESM1_ESM.doc
Additional File 1: The WGS trace files homeologous to marker sequence Sat_368. No evidence for multiple homeologs was found. (DOC 32 KB)
Authors’ original submitted files for images
Below are the links to the authors’ original submitted files for images.
Rights and permissions
Open Access This article is published under license to BioMed Central Ltd. This is an Open Access article is distributed under the terms of the Creative Commons Attribution License ( https://creativecommons.org/licenses/by/2.0 ), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
About this article
Cite this article
Saini, N., Shultz, J. & Lightfoot, D.A. Re-annotation of the physical map of Glycine max for polyploid-like regions by BAC end sequence driven whole genome shotgun read assembly. BMC Genomics 9, 323 (2008). https://doi.org/10.1186/1471-2164-9-323
Received:
Accepted:
Published:
DOI: https://doi.org/10.1186/1471-2164-9-323