Accessing genetic information with high-density DNA arrays

Science

Volumn 274, Issue 5287, 1996, Pages 610-614

  Chee, Mark ^a   Yang, Robert ^a   Hubbell, Earl ^a   Berno, Anthony ^a   Huang, Xiaohua C ^a   Stern, David ^a   Winkler, Jim ^a   Lockhart, David J ^a   Morris, Macdonald S ^a   Fodor, Stephen P A ^a  

^a Thermo Fisher Scientific   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

DNA; RNA;

ARTICLE; CLINICAL ARTICLE; DNA SEQUENCE; GENE MAPPING; GENETIC POLYMORPHISM; GENETIC VARIABILITY; GENOME; HUMAN; MOLECULAR GENETICS; PRIORITY JOURNAL;

ALGORITHMS; BASE COMPOSITION; BASE SEQUENCE; CLONING, MOLECULAR; DNA, MITOCHONDRIAL; FLUORESCEIN; FLUORESCEINS; GENE EXPRESSION; GENOME; HUMANS; MITOCHONDRIA; NUCLEIC ACID HYBRIDIZATION; OLIGONUCLEOTIDE PROBES; PHYCOERYTHRIN; POLYMERASE CHAIN REACTION; POLYMORPHISM, GENETIC; SEQUENCE ANALYSIS, DNA; VARIATION (GENETICS);

EID: 10244219858     PISSN: 00368075     EISSN: None     Source Type: Journal    
DOI: 10.1126/science.274.5287.610     Document Type: Article

References (43)

1. F. Sanger, S. Nicklen, A. R. Coulson, Proc. Natl. Acad. Sci. U.S.A. 74, 5463 (1977).
2. A. M. Maxam and W. Gilbert, ibid., p. 560.
3. J. D. Watson and F. H. C. Crick, Nature 171, 737 (1953).
4. W. Bains and G. C. Smith, J. Theor. Biol. 135, 303 (1988); Y. P. Lysov et al., Dokl. Akad. Wank. SSSR 303, 1508 (1988); R. Drmanac, I. Labat, I. Brukner, R. Crkvenjakov, Genomics 4, 114 (1989); E. Southern, U. Maskos, R. Elder, ibid. 13, 1008 (1992); see also R. B. Wallace et al., Nucleic Acids Res. 6, 3543 (1979).
5. W. Bains and G. C. Smith, J. Theor. Biol. 135, 303 (1988); Y. P. Lysov et al., Dokl. Akad. Wank. SSSR 303, 1508 (1988); R. Drmanac, I. Labat, I. Brukner, R. Crkvenjakov, Genomics 4, 114 (1989); E. Southern, U. Maskos, R. Elder, ibid. 13, 1008 (1992); see also R. B. Wallace et al., Nucleic Acids Res. 6, 3543 (1979).
6. W. Bains and G. C. Smith, J. Theor. Biol. 135, 303 (1988); Y. P. Lysov et al., Dokl. Akad. Wank. SSSR 303, 1508 (1988); R. Drmanac, I. Labat, I. Brukner, R. Crkvenjakov, Genomics 4, 114 (1989); E. Southern, U. Maskos, R. Elder, ibid. 13, 1008 (1992); see also R. B. Wallace et al., Nucleic Acids Res. 6, 3543 (1979).
7. W. Bains and G. C. Smith, J. Theor. Biol. 135, 303 (1988); Y. P. Lysov et al., Dokl. Akad. Wank. SSSR 303, 1508 (1988); R. Drmanac, I. Labat, I. Brukner, R. Crkvenjakov, Genomics 4, 114 (1989); E. Southern, U. Maskos, R. Elder, ibid. 13, 1008 (1992); see also R. B. Wallace et al., Nucleic Acids Res. 6, 3543 (1979).
8. W. Bains and G. C. Smith, J. Theor. Biol. 135, 303 (1988); Y. P. Lysov et al., Dokl. Akad. Wank. SSSR 303, 1508 (1988); R. Drmanac, I. Labat, I. Brukner, R. Crkvenjakov, Genomics 4, 114 (1989); E. Southern, U. Maskos, R. Elder, ibid. 13, 1008 (1992); see also R. B. Wallace et al., Nucleic Acids Res. 6, 3543 (1979).
9. S. P. A. Fodor et al., Science 251, 767 (1991).
10. A. C. Pease et al., Proc. Natl. Acad. Sci. U.S.A. 91, 5022 (1994).
11. 6 functional copies of a specific probe, which corresponds to a mean distance of about 100 Å between probes (M. O. Trulson, D. Stern, R. P. Rava, unpublished results).
12. S. Anderson et al., Nature 290, 457 (1981).
13. The control region of mtDNA is characterized by high amounts of sequence polymorphism concentrated in two hypervariable regions [B. D. Greenberg, J. E. Newbold, A. Sugino, Gene 21, 33 (1983); C. F. Aquardo and B. D. Greenberg, Genetics 103, 287 (1983)].
14. The control region of mtDNA is characterized by high amounts of sequence polymorphism concentrated in two hypervariable regions [B. D. Greenberg, J. E. Newbold, A. Sugino, Gene 21, 33 (1983); C. F. Aquardo and B. D. Greenberg, Genetics 103, 287 (1983)].
15. R. L. Cann, W. M. Brown, A. C. Wilson, Genetics 106, 479 (1984)
16. The mt1 and mt2 sequences were cloned from amplified genomic DNA extracted from hair roots [P. Gill, A. J. Jeffreys, D. J. Werrett, Nature 318, 577 (1985); R. K. Saiki et al., Science 239, 487 (1988)]. The clones were sequenced conventionally (1). Cloning was performed only to provide a set of pure reference samples of known sequence. For templates for fluorescent labeling, DNA was reamplified from the clones with primers bearing bacteriophage T3 and T7 RNA polymerase promoter sequences (bold; mtDNA sequences uppercase): L15935-T3, 5′-ctcggaattaaccctcactaaaggAAACCTTTTTCCAAGGA and H667-T7, 5′-taatacgactcactatagggagAGGCTAGGACCAAACCTATT.
17. The mt1 and mt2 sequences were cloned from amplified genomic DNA extracted from hair roots [P. Gill, A. J. Jeffreys, D. J. Werrett, Nature 318, 577 (1985); R. K. Saiki et al., Science 239, 487 (1988)]. The clones were sequenced conventionally (1). Cloning was performed only to provide a set of pure reference samples of known sequence. For templates for fluorescent labeling, DNA was reamplified from the clones with primers bearing bacteriophage T3 and T7 RNA polymerase promoter sequences (bold; mtDNA sequences uppercase): L15935-T3, 5′-ctcggaattaaccctcactaaaggAAACCTTTTTCCAAGGA and H667-T7, 5′-taatacgactcactatagggagAGGCTAGGACCAAACCTATT.
18. 2, and heating at 94°C for 40 min. Fragmentation improved the uniformity and specificity of hybridization (M. Chee et al., data not shown). The extent of fragmentation is dependent on the magnesium ion concentration [ J. W. Huff, K. S. Sastry, M. P. Gordon, W. E. C. Wacker, Biochemistry 3, 501 (1964); J. J Butzow and G. L. Eichorn, Biopolymers 3, 95 (1965)]. Good hybridization results have been obtained with both DNA and RNA targets prepared with a variety of labeling schemes, including incorporation of fluorescent and biotinylated deoxynucleoside triphosphates by DNA polymerases, incorporation of dye-labeled primers during PCR, ligation of labeled oligonucleotides to fragmented RNA, and direct labeling by photo-cross-linking a psoralen derivative of biotin directly to fragmented nucleic acids ( L. Wodicka, personal communication).
19. 2, and heating at 94°C for 40 min. Fragmentation improved the uniformity and specificity of hybridization (M. Chee et al., data not shown). The extent of fragmentation is dependent on the magnesium ion concentration [ J. W. Huff, K. S. Sastry, M. P. Gordon, W. E. C. Wacker, Biochemistry 3, 501 (1964); J. J Butzow and G. L. Eichorn, Biopolymers 3, 95 (1965)]. Good hybridization results have been obtained with both DNA and RNA targets prepared with a variety of labeling schemes, including incorporation of fluorescent and biotinylated deoxynucleoside triphosphates by DNA polymerases, incorporation of dye-labeled primers during PCR, ligation of labeled oligonucleotides to fragmented RNA, and direct labeling by photo-cross-linking a psoralen derivative of biotin directly to fragmented nucleic acids ( L. Wodicka, personal communication).
20. 2, and heating at 94°C for 40 min. Fragmentation improved the uniformity and specificity of hybridization (M. Chee et al., data not shown). The extent of fragmentation is dependent on the magnesium ion concentration [ J. W. Huff, K. S. Sastry, M. P. Gordon, W. E. C. Wacker, Biochemistry 3, 501 (1964); J. J Butzow and G. L. Eichorn, Biopolymers 3, 95 (1965)]. Good hybridization results have been obtained with both DNA and RNA targets prepared with a variety of labeling schemes, including incorporation of fluorescent and biotinylated deoxynucleoside triphosphates by DNA polymerases, incorporation of dye-labeled primers during PCR, ligation of labeled oligonucleotides to fragmented RNA, and direct labeling by photo-cross-linking a psoralen derivative of biotin directly to fragmented nucleic acids ( L. Wodicka, personal communication).
21. 2, and heating at 94°C for 40 min. Fragmentation improved the uniformity and specificity of hybridization (M. Chee et al., data not shown). The extent of fragmentation is dependent on the magnesium ion concentration [ J. W. Huff, K. S. Sastry, M. P. Gordon, W. E. C. Wacker, Biochemistry 3, 501 (1964); J. J Butzow and G. L. Eichorn, Biopolymers 3, 95 (1965)]. Good hybridization results have been obtained with both DNA and RNA targets prepared with a variety of labeling schemes, including incorporation of fluorescent and biotinylated deoxynucleoside triphosphates by DNA polymerases, incorporation of dye-labeled primers during PCR, ligation of labeled oligonucleotides to fragmented RNA, and direct labeling by photo-cross-linking a psoralen derivative of biotin directly to fragmented nucleic acids ( L. Wodicka, personal communication).
22. 4, 6 mM EDTA, pH 7.4), 0.005% Triton X-100. Phycoerythrin-conjugated streptavidin (2 μg/ml in 6X SSPE, 0.005% Triton X-100) was added and incubation continued at room temperature for 5 min. The chip was washed again and scanned at a resolution of ∼74 pixels per probe cell. Two scans were collected: a fluorescein scan was obtained with a 515- to 545-nm band-pass filter, and a phycoerythrin scan with a 560-nm long-pass filter. Signals were separated to remove spectral overlap and average counts per cell determined.
23. 4, 6 mM EDTA, pH 7.4), 0.005% Triton X-100. Phycoerythrin-conjugated streptavidin (2 μg/ml in 6X SSPE, 0.005% Triton X-100) was added and incubation continued at room temperature for 5 min. The chip was washed again and scanned at a resolution of ∼74 pixels per probe cell. Two scans were collected: a fluorescein scan was obtained with a 515- to 545-nm band-pass filter, and a phycoerythrin scan with a 560-nm long-pass filter. Signals were separated to remove spectral overlap and average counts per cell determined.
24. Each 2.5-kb target sequence was PCR-amplified directly from genomic DNA with the primer pair L14675-T3 (5′-aattaaccctcactaaagggATTCTCGCACGGACTACAAC) and H667-T7 (11).
25. To scale the sample to the reference intensities, we constructed a histogram of the base 10 logarithm of the intensity ratios for each pair of probes. The his-togram had a mesh size of 0.01 and was smoothed by replacing the value at each point with the average number of counts over a five-point window centered at that point. The highest value in the histogram was located, and the resulting intensity ratio was taken to be the most probable calibration coefficient.
26. Base identification was accomplished with a Bayesian classification algorithm based on variable kernel density estimation. The likelihood of each identification associated with a set of hybridization intensity values was computed by comparing an unknown set of probes to a set of example cases for which the correct base identification was known. The resulting four likelihoods were then normalized so that they summed to 1. Data from both strands were combined by averaging the values. If the most likely base identification had an average normalized likelihood greater than 0.6, it was called, otherwise the base was called as an ambiguity. The example set was derived from two different samples, ib013 and ief005, which have a total of 35 substitutions relative to mt1, of which 19 are shared with the 12 samples analyzed and 16 are not. Identification performance was not sensitive to the choice of examples.
27. To provide an independently determined reference sequence, each 2.5-kb PCR amplicon was sequenced on both strands by primer-directed fluorescent chain-terminator cycle sequencing with an ABI 373A DNA sequencer and assembled and manually edited with Sequencher 3.0. The analysis presented here assumes that the sequence amplified from genomic DNA is essentially clonal [R. J. Monnat and L A. Loeb, Proc. Natl. Acad. Sci. U.S.A. 82, 2895 (1985)] and that its determination by gel-based methods is correct. A frequent length polymorphism at positions 303 to 309 was not detected by hybridization under the conditions used. It was excluded from analysis and is not part of the set of 180 polymorphisms discussed in the text. However, polymorphisms at this site have previously been differentiated by oligonucleotide hybridization [M. Stoneking, D. Hedgecock, R. G. Higuchi, L. Vigilant, H. A. Erlich, Am. J. Hum. Genet. 48, 370 (1991)].
28. To provide an independently determined reference sequence, each 2.5-kb PCR amplicon was sequenced on both strands by primer-directed fluorescent chain-terminator cycle sequencing with an ABI 373A DNA sequencer and assembled and manually edited with Sequencher 3.0. The analysis presented here assumes that the sequence amplified from genomic DNA is essentially clonal [R. J. Monnat and L A. Loeb, Proc. Natl. Acad. Sci. U.S.A. 82, 2895 (1985)] and that its determination by gel-based methods is correct. A frequent length polymorphism at positions 303 to 309 was not detected by hybridization under the conditions used. It was excluded from analysis and is not part of the set of 180 polymorphisms discussed in the text. However, polymorphisms at this site have previously been differentiated by oligonucleotide hybridization [M. Stoneking, D. Hedgecock, R. G. Higuchi, L. Vigilant, H. A. Erlich, Am. J. Hum. Genet. 48, 370 (1991)].
29. 0) the mean value was -0.05, with a standard deviation of 0.25.
30. R. L. Cann, M. Stoneking, A. C. Wilson, Nature 325, 31 (1987); M. Zeviani et al., Am. J. Hum. Genet. 47, 904 (1990); D. C. Wallace, Annu. Rev. Biochem. 61, 1175 (1992); S. Horai, K. Hayasaka, R. Kondo, K. Tsugane, N. Takahata, Proc. Natl. Acad. Sci. U.S.A. 92, 532 (1995); T. Hutchin and G. Cortopassi, ibid., p. 6892.
31. R. L. Cann, M. Stoneking, A. C. Wilson, Nature 325, 31 (1987); M. Zeviani et al., Am. J. Hum. Genet. 47, 904 (1990); D. C. Wallace, Annu. Rev. Biochem. 61, 1175 (1992); S. Horai, K. Hayasaka, R. Kondo, K. Tsugane, N. Takahata, Proc. Natl. Acad. Sci. U.S.A. 92, 532 (1995); T. Hutchin and G. Cortopassi, ibid., p. 6892.
32. R. L. Cann, M. Stoneking, A. C. Wilson, Nature 325, 31 (1987); M. Zeviani et al., Am. J. Hum. Genet. 47, 904 (1990); D. C. Wallace, Annu. Rev. Biochem. 61, 1175 (1992); S. Horai, K. Hayasaka, R. Kondo, K. Tsugane, N. Takahata, Proc. Natl. Acad. Sci. U.S.A. 92, 532 (1995); T. Hutchin and G. Cortopassi, ibid., p. 6892.
33. R. L. Cann, M. Stoneking, A. C. Wilson, Nature 325, 31 (1987); M. Zeviani et al., Am. J. Hum. Genet. 47, 904 (1990); D. C. Wallace, Annu. Rev. Biochem. 61, 1175 (1992); S. Horai, K. Hayasaka, R. Kondo, K. Tsugane, N. Takahata, Proc. Natl. Acad. Sci. U.S.A. 92, 532 (1995); T. Hutchin and G. Cortopassi, ibid., p. 6892.
34. R. L. Cann, M. Stoneking, A. C. Wilson, Nature 325, 31 (1987); M. Zeviani et al., Am. J. Hum. Genet. 47, 904 (1990); D. C. Wallace, Annu. Rev. Biochem. 61, 1175 (1992); S. Horai, K. Hayasaka, R. Kondo, K. Tsugane, N. Takahata, Proc. Natl. Acad. Sci. U.S.A. 92, 532 (1995); T. Hutchin and G. Cortopassi, ibid., p. 6892.
35. Long-range PCR amplification was carried out on genomic DNA with Perkin-Elmer GeneAmp XL PCR reagents according to the manufacturer's protocol. Primers were L 14836-T3 (5′-aattaaccctcactaaagggATGAAACTTCGGCTCACTCCTTGGCG) and RH1066T7 (5′-taatacgactcactatagggaTTTCATCATGCGGAGATGTTGGATGG), based on RH 1066 [S. Cheng. R. Higuchi, M. Stoneking, Nature Genet. 7, 350 (1994)]. Each 100-μl reaction contained 0.2 μM concentration of each primer and ∼ 10 to 50 ng of total genomic DNA. Transcription reactions were carried out in 10 μl with Ambion MAXIscript kit according to the manufacturer's protocol. The concentration of the 16.6-kb PCR template was ∼ 2 nM, and the reaction contained Ambion 1X biotin-14-CTP/NTP mix and 0.2 mM biotin-16-UTP. Incubation was al 37°C for 2 hours. Fragmentation and hybridization were as described (13), except that 3.5 M TMACI and the biotin-labeled oligonucleotide 5′-CTGAACGGTAGCATCTTGAC were used in the hybridization buffer, which also contained fragmented baker's yeast RNA (100 μg/ml) (Sigma): Hybridization was carried out at 40°C for 4 hours.
36. A custom telecentric objective lens with a numerical aperture of 0.25 focuses 5 mW of 488-nm argon laser light to a 3-μm-diameter spot, which is scanned by a galvanometer mirror across a 14-mm field at 30 lines per second. Fluorescence collected by the objective is descanned by the galvanometer mirror, filtered by a dichroic beamsplitter (555 nm) and a band-pass filter (555 to 607 nm), focused onto a confocal pinhole, and detected by a photomultiplier. Photomultiplier output is digitized to 12 bits. A 4096 by 4096 pixel image is obtained in less than 3 min. Pixel size is 3.4 μm. The data from four sequential scans were summed to improve the signal-to-noise ratio.
37. M. D. Brown, A. S. Voljavec, M. T. Lott, I. MacDonald, D. C. Wallace, FASEB J. 6, 2791 (1992).
38. Mitochondrial DNA populations can contain more than one sequence type, in a condition known as heteroplasmy. The LHON mutations shown in Fig. 3C were characterized as being homoplasmic by conventional sequencing and restriction endonuclease digestion (M. Brown, personal communication). In controlled mixing experiments, we have shown that sequences present at the level of 10% can easily be detected by hybridization (M. Chee and R. Yang, unpublished results; N. Shen, personal communication). The sensitivity of detection is sequence dependent. Importantly, hybridization can be used to detect heterozygous nuclear DNA sequences (J. Hacia et al., in preparation).
39. Mitochondrial DNA populations can contain more than one sequence type, in a condition known as heteroplasmy. The LHON mutations shown in Fig. 3C were characterized as being homoplasmic by conventional sequencing and restriction endonuclease digestion (M. Brown, personal communication). In controlled mixing experiments, we have shown that sequences present at the level of 10% can easily be detected by hybridization (M. Chee and R. Yang, unpublished results; N. Shen, personal communication). The sensitivity of detection is sequence dependent. Importantly, hybridization can be used to detect heterozygous nuclear DNA sequences (J. Hacia et al., in preparation).
40. Mitochondrial DNA populations can contain more than one sequence type, in a condition known as heteroplasmy. The LHON mutations shown in Fig. 3C were characterized as being homoplasmic by conventional sequencing and restriction endonuclease digestion (M. Brown, personal communication). In controlled mixing experiments, we have shown that sequences present at the level of 10% can easily be detected by hybridization (M. Chee and R. Yang, unpublished results; N. Shen, personal communication). The sensitivity of detection is sequence dependent. Importantly, hybridization can be used to detect heterozygous nuclear DNA sequences (J. Hacia et al., in preparation).
41. Mitochondrial DNA populations can contain more than one sequence type, in a condition known as heteroplasmy. The LHON mutations shown in Fig. 3C were characterized as being homoplasmic by conventional sequencing and restriction endonuclease digestion (M. Brown, personal communication). In controlled mixing experiments, we have shown that sequences present at the level of 10% can easily be detected by hybridization (M. Chee and R. Yang, unpublished results; N. Shen, personal communication). The sensitivity of detection is sequence dependent. Importantly, hybridization can be used to detect heterozygous nuclear DNA sequences (J. Hacia et al., in preparation).
42. D. J. Lockhart et al., Nature Biotech., in press.
43. We thank M. Brown and D. Wallace for the gift of the LHON sample and R. Ward for the 10 African samples, M. Trulson for assistance in two-color hybridization, P. Fiekowsky for image analysis, and P. Berg and E. Lander for comments on the manuscript. R. Davis contributed to the initial concepts in oligonucleotide tiling. We especially thank L. Stryer for his incessant and persistent encouragement. Support-ed in part by Human Genome grant 5RO1HG00813 from NIH (S.P.A.F.).