A BAC pooling strategy combined with PCR-based screenings in a large, highly repetitive genome enables integration of the maize genetic and physical maps

Abstract

Background: Molecular markers serve three important functions in physical map assembly. First, they provide anchor points to genetic maps facilitating functional genomic studies. Second, they reduce the overlap required for BAC contig assembly from 80 to 50 percent. Finally, they validate assemblies based solely on BAC fingerprints. We employed a six-dimensional BAC pooling strategy in combination with a high-throughput PCR-based screening method to anchor the maize genetic and physical maps.

Results: A total of 110,592 maize BAC clones (approximately 6x haploid genome equivalents) were pooled into six different matrices, each containing 48 pools of BAC DNA. The quality of the BAC DNA pools and their utility for identifying BACs containing target genomic sequences was tested using 254 PCR-based STS markers. Five types of PCR-based STS markers were screened to assess potential uses for the BAC pools. An average of 4.68 BAC clones were identified per marker analyzed. These results were integrated with BAC fingerprint data generated by the Arizona Genomics Institute (AGI) and the Arizona Genomics Computational Laboratory (AGCoL) to assemble the BAC contigs using the FingerPrinted Contigs (FPC) software and contribute to the construction and anchoring of the physical map. A total of 234 markers (92.5%) anchored BAC contigs to their genetic map positions. The results can be viewed on the integrated map of maize 12.

Conclusion: This BAC pooling strategy is a rapid, cost effective method for genome assembly and anchoring. The requirement for six replicate positive amplifications makes this a robust method for use in large genomes with high amounts of repetitive DNA such as maize. This strategy can be used to physically map duplicate loci, provide order information for loci in a small genetic interval or with no genetic recombination, and loci with conflicting hybridization-based information.

Figure 1

Schematic display of BAC pooling strategy for six different matrices. Two hundred eighty eight 384-well microtiter plates containing 110,592 individual BAC clones were arranged in a three-dimensional square consisting of 48 ranks in all three axes (x, y, and z). Each pooled BAC DNA consists of DNAs isolated from 2,304 BAC clones simultaneously.

Figure 2

Representative gel images of BAC pool screening. Primers for umc1658 were used to amplify pool DNA. On average 6 positives per dimension are expected. Amplified PCR products were electrophoresed on 4.5% Super Fine Resolution Agarose gels (A). Each gel contained two dimensions of BAC pools, first gel with plate (PP) and face (FP) pools, second gel with side (SP) and row (RP) pools, and the last gel on the right with column (CP) and diagonal (DP) pools. First and last lane of each tier contains 100 bp ladder. Electrophoresed gel images were scored in tab delimited text format (B). Amplified products were deconvoluted by Resolve script using 3 equations (C). BAC addresses were automatically converted using Lab Convert software (D).

Figure 3

Distribution of BAC, Contig/marker hits. Maize HindIII BAC pools (6×) were screened with 237 PCR-base STS markers. The distribution of BAC hits/marker using all 237 markers (A) is illustrated. Among 237 markers tested, marker and BAC association data from 197 markers, and map position on the IBM genetic map, were used for the analysis of distribution of the numbers of identified contigs for each marker (B).

Figure 4

Distribution of BAC, Contig/marker hits. BAC and contig data from both the PCR-based screening on pool and overgo hybridizations are illustrated for the distribution of BAC hits/marker (A) and contig hits/marker (B) for 81 markers.