53706
`
`'
`- L
`7,
`git-h.
`.‘h‘ A .41 J
`fiBXNCCNG WW CAR-RT SORT ** CR03
`4000009058363§AS 01/07/94 N 2493
`52
`GEOLOGY GEOPHY LIB
`UNIV WISCONSIN
`1215 N DAYTON ST
`MADISON
`
`7
`
`HI
`
`Ariosa Exhibit 1022, p. 1
`
`

`

`
`
`Fossil shells of the gastropods Hystrivasum lock/inf
`nd Hysi‘n'vasum horridum from the upper Pliocene
`g'necrest Beds (2.0 to 3.5 million years old), Sarasota,
`Fiorida, These species are among many that became
`extinct during the late Pliocene only to be replaced by
`new species, yielding a biota whose diversity has not
`
`
` REPORTS '
`K 1624
`Diversity and Extinction of Tropical
`American Mollusks and Emergence of the
`Isthmus of Panama
`]_ B. C. Jackson, P. Jung, A. G. Coates, L. S. Collins
`
`Diversity of Atlantic Coastal Plain
`Mollusks Since the Pliocene
`W. D. Allmon, G. Rosenberg, R. W. Porrell,
`K, S. Schindler
`
`V 1626
`
`Divergence in Proteins, Mitochondrial K 1629
`DNA, and Reproductive Compatibility
`Across the Isthmus of Panama
`N. Knowlton, L. A. Weigt, L. A. Solérzano,
`D, K. Mills, E. Bermingham
`
`Large Odd-Numbered Carbon Clusters
`from Fullcrene-Ozone Reactions
`S. W. McElvany, J H. Callahan, M. M. Ross,
`L. D. Lamb, D. R. Huffman
`
`1632
`
`Identification of a Sex Pheromone from
`3 Spider
`S. Schulz and S. Toft
`
`Structural Basis of Amino Acid 0t Helix
`Propensity
`M. Blaber, X. Zhang, B. W. Matthews
`
`1635
`
`1637
`
`1 640
`
`A Nonpeptidyl Growth Hormone
`Secretagogue
`R. G. Smith, K. Cheng, W. R. Schoen, S.—S. Pong,
`G. Hickey, T. Jacks, B. Butler, W. W.—S. Chan,
`L.-Y. P. Chaung. F. Judith,J. Taylor, M. J. Wyvratt,
`M. H. Fisher
`
`Unidirectional Spread of Secondary
`Sexual Plumage Traits Across an Avian
`Hybrid Zone
`T. J. Parsons, S. L. Olson, M. J. Braun
`
`1643
`
`
`
`changed in 3.5 million years. See page 1626. Other
`evolutionary events related to the closing ofthe Isthmus
`of Panama are discussed on pages 1603, 1624, and
`1629. [Photo: Joe Traver; specimens are from the
`Paleontological Research Institution, Ithaca, NY]
`
`Evidence of DNA Bending in
`Transcription Complexes Imaged by
`Scanning Force Microscopy
`W. A. Rees, R. W. Keller, J P. Vesenka, C. Yang,
`C. Bustamante
`
`1646
`
`1649
`
`DNA Sequence Determination by
`Hybridization: A Strategy for Efficient
`Large-Scale Sequencing
`R. Drrnanac, S. Dnnanac, Z. Strezoska, T. Paunesku,
`I. Labat, M. Zeremski,J. Snodcly, W. K. Funkhouser,
`B. Koop, L. Hood, R. Crkvenjakov
`
`Synthesis of Polycrystalline Chalcopyrite
`Semiconductors by Microwave Irradiation
`C. C. Landry and A. R. Barron
`
`1653
`
`Proliferation of Human Smooth Muscle
`Cells Promoted by Lipoproteinta)
`D. J. Grainger, H. L. Kirschenlohr, J. C. Metcalfe,
`P. L. Weissberg, D. P. Wade, R. M. Lawn
`
`1655
`
`1658
`Complexes of Ras'GTP with Raf—1
`and Mitogen-Activated Protein Kinase Kinase
`S. A. Moodie, B. M. Willumsen, M. J. Weber,
`A. Wolfman
`
`Effects of CAMP Simulate a Late Stage of
`LTP in Hippocampal CA1 Neurons
`U. Frey, Y.—Y. Huang, E. R. Kandel
`
`1661
`
`TECHNICAL COMMENTS
`
`1664
`Explaining Fruit Fly Longevity
`A. Kowald and T. B. L. Kirkwood; J. M. Robine
`i
`and K. Ritchie; J. R. Carey, J W. Curtsinger
`J. W. Vaupel
`
`Compositional Interpretations of Medfly
`Mortality
`J. W. Vaupel and J. R. Carey
`
`1666
`
`
`
`
`
` MARIEHEAD
`
`
`
`Courtship display of the
`golden-collared manakin
`
`
`
`
`Kl
`
`Indicates accompanying feature
`
`I SCIENCE (lSSN 0036-8075) is published weekly on Frlday, except the last week in
`Decamber, by the American Association for the Advancement of Science, 1333 H Street,
`NW. Washington, DC 20005. Second-class postage (publication No. 484460) paid at Washington,
`.and additional mailing offices. Copyright © 1993 by the American Association for the Advance-
`Men! 01 Science. The title SCIENCE is a registered trademark ol the AAAS. Domestic individual
`”Embership and subscription (51 issues): 587 ($47 allocated to subscription). Domestic institutional
`Subscription (51 issues): $205. Foreign postage extra: Mexico, Caribbean (surface mail) $50;
`Other countries (air assist delivery) $95. First class, airmail. student and emeritus rates on
`'eqUest. Canadian rates with GST available upon request, GST #11254 88122. Change of address:
`an0W 6 weeks, giving old and new addresses and 11-digil account number. Postmaster: Send
`LI“little or address to Science, PO. Box 2033. Marion. OH 43305-2033. Single copy sales: $6.00
`99' lSSue prepaid includes suriace postage; Guide to Biotechnology Products and Instruments, $20.
`
`Bulk rates on request. Authorization to photocopy material tor internal or personal use under
`circumstances not falling within the tair use provisions oi the Copyright Act is granted by AAAS to
`libraries and other users registered with the Copyright Clearance Center (000) Transactional
`Reporting Service, provided that the base lee 0181 per copy plus $0.10 per page is paid directly
`to CCC, 27 Congress Street, Salem, MA 01970. The idcntiiication code for Science is 0036A8075/83
`$1 + .10. Science is indexed in the Reader's Guide to Periodical Literature and in several
`specialized indexes.
`I The American Association for the Advancementot Science was lounded in 184Band incorporated
`in 1874. Its objectives are to further the work of sctenlisis. to lacrlitate cooperation among them, to
`loster scientilic freedom and responsibility. to improve the attentiveness of science in the promotion
`01 human welfare, to advance education in science, and to increase public understanding and appre-
`ciation of the importance and promise of the methods 01 science in human progress.
`
`SCIENCE ° VOL. 260 ' 11 JUNE 1993
`
`Ariosa Exhibit 1022, p. 2
`
`Ariosa Exhibit 1022, p. 2
`
`

`

`l
`
`11.
`
`12.
`
`i3.
`
`14.
`
`10 Anhough it is possible to image macromolecules
`under aqueous envtronments (36).
`the yield ol
`stably bound complexes on the mica surface with
`the use 01 the current deposition method is insuf-
`ficient for the statistical analysis described here.
`This low yield is probably attributable to screen-
`ing. by the ionic medium. of
`the electrostatic
`interactions that hold the complexes to the mica.
`Although the molecules are not fully bathed in
`water under the humidity-controlled conditions
`described here. the macromolecules are likely to
`retain strongly bound and structurally essential
`water (37). These conditions are then less harsh
`than the desiccating environments required for
`the imaging of macromolecules by more tradition-
`al electron microscopic methods.
`J. Ft. Levin. B. Krummel. M. J. Chamberlin, J. Moi.
`Biol. 196. 85 (1987).
`The RNA polymerase holoenzymes (Epicentre
`Technologies)
`and DNA fragments
`(30
`nM
`(RBz-BstU1 DNA fragment) (38) were incubated
`for 10 min at 30°C at an ~221 ratio in a buffer
`containing 20 mM Iris—Cl
`(pH : 7.6). 0.1 mM
`EDTA. 100 mM NaCl. 0.5 mM dithiothreitol (DTD.
`and 5% glycerol. The reaction was then diluted
`~10-lold in the above buffer without NaCl or
`glycerol and was incubated on Parafilrn at ~37°C
`tor an additional 5 min before being deposited on
`freshly cleaved mica. Deposited samples were
`processed and imaged as prevtously described
`(1. 3). The (315 elongation complexes were
`formed after incubation of OPCs with 100 mM
`dinucleotide ApU and 10 mM each of adenosine
`triphosphate (ATP). CTP. and GTP (final concen-
`trations) (no UTP) for 10 min at 30°C and then
`incubation with heparin (75 pg/ml)
`for 10 min
`before dilution and incubation on Parafllm like the
`OPCs. Heparin was added to trap the RNA poly—
`merase not involved in the elongation complexes;
`the highly bent molecules characteristic of 015
`complexes were observed in both the presence
`and the absence of heparin (39). Because of the
`specific sequence of the initial region transcribed
`from the PL promoter. the majority (>80%) of the
`elongation complexes formed by this missing
`nucleotide procedure contained a 15-nucleotide
`transcript (015 complexes) (11. 38).
`D. J. Keller and C. Chih-Chung. Surf. Sci. 268. 333
`(1992).
`Because all lateral dimensions are overestimated
`by about twice the radius of curvature of the tip in
`SFM images (1. 3). the ratios determined here are
`underestimates of the actual values. From the
`apparent width of the DNA molecules (110 1 15
`A). the average radius of curvature of the tip is
`estimated to be ~75 A (1. 3). With use of the
`lateral dimensions of RNA polymerase. obtained
`by electron diffraction of two-dimensional crystals
`as an estimate of the actual values (160 A by 100
`A) (15). and use of the above estimate of the
`radius of curvature of the tip, the ratio of lateral
`dimensions should be ~(160 A + 2 X 75 A)/(100
`A + 2 X 75 A) = 1 24. similar to the observed
`values.
`S. A. Darst, E. W. Kubalek. R. D. Kornberg. Nature
`340. 730 (1989).
`W. Tichelaar et al.. Eur. J. Biochem. 135. 263
`(1983); O. Meisenberger. H. Heumann.
`l. Pilz.
`FEBS Leli.112. 117 (1980).
`H Heumann et at. J. Mol. Biol. 201. 115 (1988).
`The term “bending" is used here to indicate the
`deviation of the flanking arms of the template at
`both sides of the polymerase from the straight
`configuration. even though all the DNA is not in a
`double-stranded form in the region that
`is
`in
`Contact with the polymerase.
`H. Heumann. M. Ricchetti. W. Werel. EMBO J. 7.
`4379 (1988).
`G Kuhnke. H.—J. Fritz. R. Ehring.
`(1987).
`J. Hirsh and R. Schleif. J. Mol. Biol. 108. 471
`(1976).
`R. C. Williams. Proc. Natl. Acad. Sci. USA. 74.
`2311 (1977).
`B. ten HeggelerBordier. W. Wahli. M. Adrian. A.
`Stasiak. J. Dubochot. EMBO J. 11. 667 (1992).
`
`15.
`
`16.
`
`17.
`18.
`
`19.
`
`21.
`
`(bid. 6. 507
`
`24.
`
`25.
`
`26.
`
`27.
`
`28.
`
`29.
`
`30.
`
`31.
`
`ibid. B.
`
`P. Schickor. W. Metzger, W. Werel. H. Lederer. H.
`Heumann. ibid. 9. 2215 (1990).
`W. Metzger. P. Sohickor. H. Heumann.
`2745 (1989).
`A. J Carpousis and J. D. Gralla. J. Mol. Biol. 183.
`165 (1985).
`A. Spassky. K. Kirkegaard. H. Buc. Biochemistry
`24. 2723 (1985).
`D. C. Straney and D. M. Crothers. Cell 43. 449
`(1985).
`To determine if the observed bend-angle distribu-
`tions can be accounted for by random thermal
`fluctuations. we compared the data with a Gaus-
`sian probability distribution
`9})(6) = (n6/P)“’e)<p(—R12/€)
`where 9(0) is the probability of finding a given
`bond angle 6 over a length 6 and P is the
`persistence length of doublcestranded DNA (40).
`Here. 6 is assumed to be the length covered by
`the polymerase as determined from protection
`studies (6 = 80 bp for OPCs and 40 bp for C15
`complexes) (25. 41). The Gaussian is centered at
`0°. as would be expected for a DNA fragment
`possessing no intrinsic bends or curvature. The
`DNA within 100 bp of the transcription start site of
`PL does not appear to be intrinsically bent (42)
`[also confirmed by the SFM images (Fig. 4C)].
`The width of the distribution is determined by the
`bending rigidity of the DNA template as characi
`terized by its persistence length. P = 210 bp at
`25°C (43). The Kolmogorov-Smirnov test was
`used to compare the observed and the theoretical
`distributions. This test indicated that the observed
`data are not likely to be drawn from the theoretical
`distribution (significance = 0.000 for OPC data
`and <0.000 lor C15 data).
`U. M. Hansen and W R. McClure. J. Biol. Chem.
`255. 9564 (1980).
`B. Krummel and M. J. Chamberlin. Biochemistry
`
`32.
`
`33.
`
`34.
`
`35.
`
`3B.
`
`37.
`
`38.
`
`28. 7829 (1989).
`D. Rhodes and M. J. Chamberlin, J. Biol. Chem.
`249. 6675 (1974).
`R. Schafer. W. Zillig. K. Zechel. Eur. J. Biochem.
`33. 207 (1973).
`K. Arndt and M. J. Chamberlin, J. MOI. Biol. 211.
`79 (1990).
`Ft. L. Novak and P. Doty. J. Biol. Chem. 243. 6068
`(1968).
`C. Bustamante. D. J. Keller. G. Yang. Curr. Opin,
`Struct. Biol. in press.
`J. T. Edsall.
`in The Proteins. H. Neurath and K.
`Bailey. Eds. (Academic Press, New York, 1953).
`vol. 1. part B. p. 549.
`W. A. Rees. P. H. von Hippet. A. K. Das. unpub
`lished data.
`. W. A. Rees. R. W. Keller. J. P. Vesenka. G. Yang.
`C. Bustamante, unpublished data.
`40.
`MD. Barkley and B. H. Zimm. J. Chem. Phys. 70.
`2991 (1979).
`B. Krummel and M. J. Chamberlin, J. Mol. Biol.
`225. 239 (1992).
`H. Giladi. M. Gottesman, A. B. Oppenheim. ibid.
`213. 109 (1990).
`S. B. Smith. L. Finzi. C. Bustamante. Science 258.
`1122 (1992).
`We thank S. E. Weitzel fortechnical assistance. G.
`P. Harhay for assistance with data analysis. and
`P. H. von Hippel. D. Erie. and A. Das for com-
`ments on the manuscript. Supported by US.
`Public Health Service (USPHS) research grants
`(SM-32543 (C.B.) and (SM-15792 and GM-29158
`(P. H, von Hippel). National Science Foundation
`grant MCB~9118482 (CB). and a grant from the
`Lucille P. Markey Charitable Trust to the Institute
`of Molecular Biology. WAR. was a predoctoral
`trainee on USPHS institutional training grant GM,
`07759.
`
`41.
`
`42.
`
`43.
`
`44.
`
`8 January 1993; accepted 13 April 1993
`
`This material may be protected by Copyright law (Title 17 US. Code)
`
`DNA Sequence Determination by Hybridization:
`A Strategy for Efficient Large-Scale Sequencing
`
`R. Drmanac, S. Drmanac, Z. Strezoska,* T. Paunesku.* l. Labat,*
`M. Zeremski,* J. Snoddyn‘ W. K. Funkhouser, B. Koop,t
`L. Hood,§ R. Crkvenjakov||
`
`The concept of sequencing by hybridization (SBH) makes use of an array of all possible
`n-nucleotide oligomers (n-mers) to identify n-mers present in an unknown DNA sequence.
`Computational approaches can then be used to assemble the complete sequence. As a
`validation of this concept. the sequences of three DNA fragments. 343 base pairs in length,
`were determined with octamer oligonucleotides. Possible applications of SBH include
`physical mapping (ordering) of overlapping DNA clones. sequence checking, DNA fin-
`gerprinting comparisons of normal and disease—causing genes. and the identification of
`DNA fragments with particular sequence motifs in complementary DNA and genomic
`libraries. The SBH techniques may accelerate the mapping and sequencing phases of the
`human genome project.
`
`The success of the human genome project
`will depend on whether DNA sequencing
`approaches can greatly increase through-
`put [at least
`IOO—fold more than the cur—
`rent value of ~104 base pairs (bp) per day
`per machine) and decrease cost. Strategies
`that may help to accomplish this task
`include a greatly improved method of
`sequencing based on the conventional au—
`tomated fluorescent DNA sequencers and
`SCIENCE ' VOL. 260
`
`‘IJUNE 1993
`
`°
`
`the advent of new sequencing technolo-
`gies (l—3).
`Any linear sequence is an assembly of
`overlapping,
`shorter
`subscquences. Se-
`quencing by hybridization (SBH)
`is based on the use of oligonuclcotide
`hybridization to determine the set of con—
`stituent subsequences
`(such as 8—mers)
`present
`in a DNA fragment. Unknown
`DNA samples can be attached to a support
`
`Ariosa Exhibit 1022, p. 3 1649
`
`Ariosa Exhibit 1022, p. 3
`
`

`

`and sequentially hybridized with labeled
`oligonucleotides (format 1); alternatively,
`the DNA can be labeled and sequentially
`hybridized to an array of support-bound
`oligonucleotides (format 2). Highly dis-
`criminative hybridization is
`required to
`distinguish between perfect DNA frag-
`ment and oligonucleotide complementar—
`ity and all hybridizations exhibiting one or
`more nucleotide mismatches. Reliable
`conditions for
`this discrimination have
`recently been determined for format 1 (5).
`The sets of n-mer oligomers used as
`hybridization probes can vary in number
`from several hundred to all possible combi—
`nations (65,536 for octamers), depending
`on the type of sequence information re-
`quired. The completeness of the probe set
`and its design, which can vary according to
`such parameters as length of probe and
`internal (7) or flanking positioning (4) of
`unspecified bases, determines the kind of
`sequence information that can be extracted
`from individual DNA fragments (7) or li-
`braries of fragments (4). Mapping informa
`tion that determines clone overlap (I 1) can
`be obtained with 100 to 200 probes. The
`positioning and identification of genome
`structural elements
`(partial
`sequencing)
`(12—14) requires 500 to 3000 probes, and
`complete sequencing (4) requires data from
`3000 or more septamer probes on three to
`five related genomes (14).
`In this report, we present the results of
`a blind test of the SBH method on homol—
`ogous DNA fragments cloned into an 8-kb
`M13 vector. The test consisted of se-
`quencing by traditional methods (15), in
`one of our laboratories, three Z-kb inserts
`containing related variable gene segments
`(92 to 94% similarity)
`from several pri—
`mate T cell
`receptor B loci and then
`resequencing by SBH. in the other labo-
`ratory, homologous
`116—bp
`regions of
`these clones. The experiment was
`de—
`signed to determine the ability of SBH not
`only to produce accurate sequence infor-
`mation, but also to circumvent for practi-
`cal reasons the SBH requirement for thou-
`sands of probes and clones in an initial
`Instead of 65,635 octamers, a re—
`Fi. Drmanac. S. Drmanac. Z. Slrezoska, T. Paunesku. I.
`Labat. M. Zeremski. J. Snoddy, R. Crkvenjakov. Bio-
`logical and Medical Research Division, Argonne Na
`tional Laboratory. Argonne, IL 60439.
`W. K. Funkhouser. B. Koop. L. Hood. Department of
`Biology. California Institute of Technology, Pasadena.
`
`‘Permanent address: Institute for Molecular Genetics
`and Genetic Engineering, Post Olftce Box 794. Bel-
`grade 11000. Yugoslavia.
`tPresent address: Office of Health and Environmental
`Research. EFl-72. Department of Energy. Washington.
`iDepartment of Biology, Universrty of Victoria. Victoria.
`BC V8W 2Y2. Canada
`§Department of Molecular Biotechnology. University of
`Washington. Seattle. WA 98195.
`
`duced probe set based on gel-determined
`sequences was
`synthesized with about
`twice as many nonmarcbing as matching
`probes in relation to each of the 116-bp
`targets. The simultaneous probing of three
`similar sequences containing many single
`base mismatches within the target—probe
`combinations provides a stringent test of
`the accuracy of SBH, which justifies use of
`the reduced probe set,
`
`The lists of 8—mers with more than
`three G + C bases and 9-mers with one to
`two G + C bases (16) occurring in the
`human and two rhesus monkey 116—bp
`DNA segments were scrambled and sent to
`the Argonne group along with number.
`coded samples of the three clones (Clones
`I, 7, and 8). The Argonne group was
`informed that within the list was a sublist
`that allowed unambiguous reconstruction
`
`
`
`
`Fig. 1. Representative hybridiza-
`
`
`
`tion images obtained from one filter
`
`
`
`
`
`
`
`
`with different oligomer probes. The
`
`filter
`is made from one microliter
`
`
`
`
`plate. Duplicate dots are present in
`
`the diagonally adjacent well. below
`
`
`
`and to the right. The remainder of
`
`
`the wells contain M13 clones with
`
`
`
`inserts 01 known sequence, The
`
`
`
`
`
`
`
`
`computer~drawn grid is included to
`
`
`
`
`facilitate comparison. The order of
`
`
`
`
`O
` t."
`
`
`test clones (numbering from top to
`
`I
`
`
`
`Ll
`bottom and from left to right) is as
`
`It:
`-Tfli:slat
`
`follows: Clone 1. line 3 position 3.
`
`line 5 position 23, line 9 position 9.
`
`
`
`and line 13 position 11; clone 7.
`
`
`line 3 position 7. line 7 position 23.
`
`
`line 9 position 13, and line 13 po-
`
`
`
`
`sition 13; and clone 8, line 3 posi-
`
`
`
`
`line 9
`Images (from top to bottom) are as
`
`tion 11.
`line 9 position 23.
`
`position 17, and line 13 position 15.
`
`
`follows:
`Probe GCCAGCTTTC
`
`(with target in M13 vector included
`the hybridization of Clones of unknown sequence [except for probe GAAGCTT. where six positive
`to calibrate the DNA amounts in
`the dots), probe GGGCCCAA.
`probe CTTAATATA. probe CGTGGCCT, probe GCTGGCCT. probe TGAGTGTC. probe TGAAG‘
`CTT, and probe ATAAAGAGT. For clarity. we selected probes with no positive controls to highlight
`
`
`
`
`control clones (indicated by arrows) are visible: three in the upper left corner. and three on the rightly
`Other weaker signals visible in the images are at least tenfold lower than the full-match ones from
`the same image. The lowermost image shows the additional central positions of test clones hidden
`in the other images.
`
`B
`Clone 1
`nnTGTI'CCAAATGGATGAAGGTI'GCTAATCAGCTGGCCTTMTATGGGGAGCGTGTCCTGGATTT
`TCCAGATGGGCCCMTGTAATCACMGGGTCCTAMATGTGGMGAGGGAGGATGMGAGTGAGT
`GTCAGAGTGATGCAGA
`
`Fig. 2. (A) Sequence re—
`A
`construction. A part of
`the clone 1 sequence is El
`
`ll nTGTTCCAAATGGATGAAGGTTGCTAATCAGCTGGCCTTAATATGGGGAGCGTGTCCTGGATTT
`depicted,
`showing the
`probes giving positive
`
`and false-positive
`hy-
`bridization scores (solid
`lines without and with G.
`respectively) and some
`of the negative scores
`(dashed lines). The de
`termination of T at posi»
`tion 10 from right is es-
`pecially instructive be-
`cause it has four probes
`determining it as T, three
`as non-G. and two as G.
`giving odds of seven to
`two against G being at
`this
`position.
`(B) See
`quences of
`three seg-
`ments from T cell recep-
`tor genes determined by
`SBH. Clone 1. human
`
`Clone 7
`n nTG'I'I'CCAAATGGATGAAGCTTGCTAATCAGCTGBCCT TAATATAGGGAGCCTGGATTTTCCAG
`GTGGGCCCAATGTAATCCACAAt3GGTCCTAMATGTGGAAGAGGGAGAATGAAGAGTGAGCCTCAG
`AGTGAT GCAGA
`
`Clone 8
`nnTGTTCCAAATGGACMAGCTTGCTAATCATCTG(30GT TAATATAGGGAGCGTGGCCTGGATTT
`TCCAGGTGGGCCCAATGTAATCACAAGGGTCCTAAAATGTAGAAGAGGGAGAATAAAGTGAGO
`CTCAGAGTGATGCAGA
`
`Ariosa Exhibit 1022, p. 4
`
`

`

`1‘0” 100— to 140-bp DNA segment in at
`least one of the clones and that the other
`No sequences sent were similar. Because
`QM list constructed from these sequences
`(would have allowed a complete recon-
`r sanction without any of the probes being
`excludcd,
`the probe information derived
`from a fourth similar sequence (obtained
`from gorilla DNA whose sample was not
`‘56,“) was included in the filial 272-oligo-
`may probe list. This information made the
`rest more challenging by increasing the
`number of complete sequences that could
`beobtained from the list.
`A filter containing 96 DNA samples
`‘ sPotted in duplicate was prepared directly
`from M13 phage cultures grown in Stan»
`dard microtiter plates (17, 18). Four inde-
`Pendent phage cultures from each of the
`[huge unknown clones were used. The
`remaining samples consisted of recombi-
`nant M13 clones for which the complete
`(sequence was known (19). These latter
`(DNAS served as positive and negative
`‘wntrols against which the hybridization
`signal for each probe could be calibrated.
`lThe
`total complexity of these control
`DNAs (~45 kb) was such that they pro-
`vided frilly matched targets for less than
`40% of the supplied probes. Groups of
`sixteen 10- and 11—mer oligonuclcotides
`sharing given octamers or nonamers as the
`informatic cores (4) were used as probes
`because shorter probes gave an insufficient
`
`i
`
`signal with the available amounts of M13
`DNA in the dots.
`Each filter was first hybridized with a
`probe specific for the M13 vector so that
`the relative amount of DNA and hence
`the potential hybridization strength per
`dot could be measured. The filters were
`then successively hybridized with different
`probes. The discrimination values of pos—
`itive signals for a given probe (average
`ratio of signals from matched and mis»
`matched targets normalized in relation to
`variation of DNA molarity among dots)
`ranged from 2 to 92 for different probes,
`with the average value above 10 (18). In
`Fig.
`l are shown images representing the
`discrimination patterns of hybridization
`found with clones 1, 7, and 8. The suc-
`cessive hybridizations were stopped when
`156 probes (20) out of the 272 supplied for
`the test sequence were used because suffi-
`cient data were obtained for a correct
`determination at each base position (Fig.
`2A). On average, each base was covered
`by positive hybridization with more than
`five probes. For each clone, a continuous
`sequence was reconstructed (Fig. 28), and
`all
`three test sequences were unambigu—
`ously determined.
`During the hybridization of 156 probes
`to the three different DNA samples, 22 of
`468 probe hybridization scores gave false
`positive results
`that were incompatible
`with the final
`reconstructed sequences.
`
`A
`
`1
`
`2 3
`
`45678
`
`Fig. 3. (A) Southern blot analysis
`of PCB-amplified inserts of un-
`known clones 1, 7, and 8. Probe
`NTGGCCTGGN hybridized to the
`same filter in the absence (left) or
`presence (right) of the competing
`miabeled oligomers NGGAGCGi
`TGN and NCTGGATTTN specific
`for the fully matched or a mis—
`matched variant of the target with-
`nsequenced segments of clones
`
`
`
`.
`
`Clone a
`eronei
`
`
`...........NTG CCTGGN'
`....cccrcGCAchsAccrAAAA
`NGGAGCGTGNNCTGGATITN
`.....................NTGGCCTGGN'
`
`CCCICGCACCGGACCTAAAA
`NGGAGCGTGNNCTGGATTIN
`
`land 1, respectively. Lanes 1, 2,
`
`and 3. digests of a 2—kb insert of
`'
`'
`'
`Clone 1 With restriction enzymes
`Psi l. Aiu II and Fisa l, respective,
`if lanes 4. 5, 6, 7, and 8, digests
`.
`.
`0i an Insert 0i Clone 8 Willi ECO RI,
`iaq i, Alu l, Hind Ill. and Hpa ||,
`lfispectively. The unmarked lane
`betvveen lanes 1 and 2 is an Hae
`lidigest of insert 1; no hybridiza-
`J7On is expected because of the
`leSlriction site being within the
`itiger. The unmarked lanes be-
`illeen lanes 3 and 4 are various
`iIgests of insert 7; no hybridiza-
`tion is expected from dot blot results. The sizes of visible fragments range from 159 to 1780 bp. (B)
`Overcoming of the context-dependent depression of hybridization.
`(Left) Ninety-six—dol array
`”lobed with decamer NATGGATGAN; the positions of clones 1 and 7 that are expected to be
`Positive are underlined. No higher hybridization signals in relation to expected negative control
`Climes (all other positions) are evident. (Right) The same filler allows efficient discrimination when
`”Ybridized with 12vmer probe CAAATGGATGAA complementary to the same target but extended
`5V two bases on the 5' end.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`SCIENCE ' VOL. 260
`
`'
`
`lljUNE1993
`
`Clone a
`Clone 1
`NTG ccreeNu
`Probe
`Target In Clonet CAchGACCT....
`Probe
`NTGGCCTGGN'.
`Target In Clone a cmccmcc‘r... .
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Halfof the inconsistencies could be traced
`to full matches outside the targeted 116—bp
`sequences. Of the 11 remaining false pos-
`itive results (20), 7 could be accounted for
`by signal summation from two or more
`single mismatch hybrids located in differ-
`ent regions of the clones. In only one of
`these seven cases, one of the mismatched
`hybrids was located within the 1 l6-bp test
`sequence, as shown through mapping by
`competitive oligomer hybridization in
`Southern (DNA) analysis (Fig. 3A). The
`remaining four false positives resulted from
`a single mismatched target. The major
`cause of false positive results thus appeared
`to be the additive effects of two or more
`singly mismatched hybridizations.
`in some regions, especially in a 157-bp
`sequence near the 5' end of the three test
`sequences, we also detected an unexpect~
`ed drop in hybridization intensity (Fig.
`313,
`left). An examination of this phe-
`nomenon (which we call ”SBH compres-
`sion") with model oligomer targets sug
`gested that the discrimination of matched
`and mismatched hybrids within this region
`is retained,
`though at severalfold lower
`intensities (21). This hybrid instability
`may be caused by a secondary or tertiary
`DNA conformation. When longer probes
`specific for this target were used, a specific
`signal was detected (Fig. 313, right).
`some
`cases, nonstandard hybridization
`conditions or
`longer probes may be re—
`quired to obtain unambiguous sequence
`data. The octamer AATATAGG fell into
`the false negative probe category. It did
`not reproducibly give the predicted hy-
`bridization pattern under standard condi—
`tions. However, when conditions were
`modified to detect weaker hybridizations
`(5—min wash at 0°C),
`at signal—to-noisc
`ratio greater than five was consistently
`obtained.
`
`The low percentage of false hybridiza-
`tions ill/(156 X 3), or 2.3%] observed in
`this study is similar to that seen previously
`(6). Two considerations indicate that the
`observed error rate may not be a serious
`concern. First, software simulations suggest
`that as much as 10% error does not appear
`to affect the final sequence determination
`(16). Second, recent improvements in re-
`construction algorithms have reduced the
`number of required heptamer probes for
`complete sequencing from 16,384 to about
`3,000, or a fifth of all possible 7—mers (14,
`22, 23), when the sample DNA is attached
`to a support. This fivefold reduction in
`probe repertoire allows for preselection of
`optimal probes in relation to hybridization
`errors. Thus, a well—characterized set of
`reagents reliable under standard hybridiza—
`tion conditions might be sufficient to cover
`all possible sequences.
`The sequences of the three unknown
`
`Ariosa Exhibit 1022, p. 5
`
`"35‘
`
`Ariosa Exhibit 1022, p. 5
`
`

`

`AAGGGTCC 1. 7. 8: ATATAGGG 7. B; ATATGGGG
`1; ATAGGGAG 7. 8; ATCATCTG 8; ATCTGGCC 8*
`ATGAAGAG 1. 7; ATGAAGCT 7; ATGAAGGT 1'
`ATGTCCTG; ATGGGCCC t; ATGGGGAG 1'
`ACAAAGCT 8; ACAAGGGT 1. 7. 8; ACTGTrCC'
`ACGAAGAG: AGAAGAGG 8; AGAGTGAG 1. 7, 3‘
`AGAGGGAG 1. 7, 8; AGAGGGGG; AGTCAGCi-i
`AGTGAGCC 7. 8; AGTGTCAG 1;AGCT iGCT 7.11
`AGCCTGGA 7; AGCGTGTC 1; AGCGTGGC 3'
`AGGATGAA 1; AGGTTGCT 1; AGGGAGAA 7, 3'
`AGGGAGCA: AGGGAGCC 7; AGGGAGCG 8‘
`AGGGAGGA1;AGGGTCCT 1. 7, 8: AGGGGGGA:
`TAATATGG 1: TAATCAGC 1. 7; TATAGGGA 7, 3‘
`TATGGGGA 1; TAGAAGAG 8; more/«30111:
`TCCAG 1. 7, 8; TTI'CCAGA 1; TTI'CCAGG 7, 3-
`TTTOCAGG; TTCCAGAT 1. TTCCAGGT 7, 8f
`TCATCTGG 8; TCACAAGT: TCACAAGG 1, 7. a;
`TCAGCTGG 1. 7; TCTGGCCT 8: TCCAGATG 1;
`TGAAGAGT 1. 7; TGAAGCTT 7; TGAGTGTC 1,
`TGAGCCTC 7, 8; TGTAGAAG 8; TGTCAGAG 1_
`TGTCCTAA. TGTCCTGG 1; TGTGGAAG 1. 7;
`TGCTAATC 1. 7. 8; TGCTAGTC; TGGAAGAG 1, 7;
`TGGCCTTA 1. 7. 8: TGGCCTGG 1. 8; TGGG
`GAGC 1: CAAAGCTT 8: CATCTGGC 8: CAGC-
`TGGC 1, 7; CAGGTGGG 7. 8. CTAGTCAG; CT-
`TGCTAA 7. 8. CTGGATTT 1. 7. 8; CTGGCCTT t,
`7. 8; CCAGATGG 1. CCAGGTGG 7. 8; CCTG.
`GATT 1. 7. 8: CCCAATGT 1. 7. 8; CGAAGAGT 7.
`8; CGTGGCCT 1. 8; GAATGAAG 1.
`7; GMA
`GAGTG 1. 7; GAAGAGGG 1. 7. 8; GAAGCTTG 7.
`GAAGGTTG 1; GATTTTCC 1. 7. 8; GATTI'TGC,
`GACGAAGA. GAGTGAGT 1; GAGCATGT; GA-
`GCCTCA 7. 8; GAGCCTGG 7; GAGCGTGT 1.‘
`GAGCGTGG 8; GAGGATGA 1: GTAGAAGA B.
`GTTGCTAA 1; GTCCTAAA 1. 7. 8: GTOTCAGA‘l;
`GTGTCCTG i: GTGGGCCC 7, 8; GCAGGTGG;
`GCTAATCA 1 . 7. 8. GCTAGTCA; GCTGGCCT1,7;
`GCCTGGAT 7. 8; GCGTGTCC 1; GCGTGGCC 8;
`GGATTTTC 1. 7, 8; GGATTTTG; GGAGCATG l.
`7. 8; GGAGCCTG 7: GGAGCGTG 1. 8; GGAG-
`GATG 1; GGTTGCTA 1; GGTCCTAA 1. 7. 8:
`GGCCTTAA 1. 7. 8: GGCCTGGA 8; GGGAGAAT
`7, 8; GGGAGCAT. GGGAGCCT 7; GGGAGCGT l.
`8; GGGCCCAA 1, 7, 8; GGGGAGCG 1. AAATGA
`TAGA 8; AATCATCTC- 7. 8; ATAAAGAGT 8; ATTT-
`TCCAG 1. 7. 8; ATTTTGCAG; ATGTAGAAG 8:
`TAAAATGTA 8; TAAAGAGTG 8; TAATATAGG 7.
`8; TAATCACAA 1. 7. 8; TTAATATAG 7. 8; TTAA‘
`TATGG 1: TTGCTAATC 1. 7. 8. TGTAATCAC 1. 7.
`8: TGGATTI'TC 1. 7. 8: TGGATTI’TG; CTAAA-
`ATGT 1. 7. 8; CTAATCATC 8; CTTAATATA 7. 3:
`CCTTAATAT 1. 7. 8; GAGAATAAA 8. Eleven false
`positive scores are included as follows: Wilh
`clones 1. 7. and 8. probes ATGTCCTG and
`GGAGCATG: with clones 7 and 8, probe CGAA-
`GAGT; and with clone 1 probes TGGCCTGG-
`CGTGGCCT. and GAATGAAG. An additional
`nine unscorable probes were tested but are "01
`listed (reasons for noninclusion:
`the full-match
`targets in M13 or in surrounding inserts obchi'
`ing the score.
`the strong hybridi7ation of 3"
`clones. control and unknown.
`indicating V66tor
`mismatch origin; and single test clone false-
`positives with no fuIl-match controls available.
`that
`is, specific mismatch hybridization abOVe
`background),
`T, Paunesku and R. Crkveniakov. unpublished
`results.
`R. Drmanac. in Genome Mapping and Sequenc'
`ing (Cold Spring Harbor Laboratory Press. CO'd
`Spring Harbor. NY. 1992), p. 318.
`I, Labat and F1. Drmanac. unpublished results.
`M, o. Adams et al.. Science 252, 1651 (1991): A'
`S. Wilcox er al.. Nucleic Acids Res. 13.
`1
`7
`(1991).
`R. Drmanac and R. Crkvenjakov. Screntia YUQDSI'
`16. 97 (1990); s. P. A. Fodor at al.. Science 251.
`767 (1991); M. Eggers et al..
`in (22). p. 111.
`'
`We thank M Stodolsky of
`the Department 0‘
`Energy. Office of Health and Environmental.I Re
`search. for monitoring the blind test and aidingf
`in its design. Supported by US, Uepartrriefl1 0.
`Energ)’. Office of Health and Environmental RE
`search. under contract W-31-109-ENG'3B'
`
`21.
`
`22.
`
`23.
`24,
`
`25.
`
`26.
`
`REFERENCES AND NOTES
`
`
`lnt. J. Genome
`
`T. Hunkapiller. R, J. Kaiser. B. F Koop. L. Hood.
`Science 254. 59 (1991).
`.l. H. Jen oral, J. Biomol. Struct. Dyn. 7. 301 (1989).
`S. M. Lindsay and M. Philipp. Genet. Anal. Tech.
`Appl. 8. 8 (1991).
`R. Drmanac and R, Crkvcniakov, Yugoslav patent
`application 570/87 (1987); R Drmanac. l, Labat, l.
`Brukner. Fl. Crkvenjakov. Genomics 4. 114 (1989).
`R. Drmanac. Z, Strezoska. I. Labat. S. Drmanac.
`F1. Crkveniakov. DNA Cell Biol. 9. 527 (1990).
`Z. Strezoska et al.. Proc. Natl. Acad. Sci. USA.
`88. 10089 (1991).
`W. Bains and G. C, Smith. J. Theor. Biol. 135, 303
`(1988).
`Y. P. Lysov et al.. Dokl. Akad. Nauk SSSFi 303.
`1508 (1988); K. R. Khrapko of al.. FEBS Lett. 256.
`118 (1989); P. A. Pevzner etal,. J. Biomol. Strucl.
`Dyn. 9. 399 (1991),
`K, R. Khrapko et al.. J. DNA Sequencing Mapp. 1.
`375 (1991),
`E. M. Southern. United Kingdom Patent Applica»
`tion GB 8810400 (1988). E. M. Southern er al..
`Genomics 13. 1008 (1992).
`A. Poustka at al.. Cold Spring Harbor Symp.
`Quant, Biol. 51, 131 (1986).
`R. Drmanac at al.. Electrophorescs. Supercom-
`puters and the Human Genome (World Scienti