(12) INTERNATIONAL APPLICATION PUBLISHED UNDER THE PATENT COOPERATION TREATY(PCT)
`
`B34
`
`(19) World Intellectual Property Organization
`International Bureau
`
`(43) International Publication Date
`4 December 2003 (04.12.2003)
`
`
`
`PCT
`
`(10) International Publication Number
`WO 03/100012 A2
`
`(51) International Patent Classification’:
`
`C12N
`
`(81) Designated States (national): AE, AG, AL, AM,AT, AU,
`AZ, BA, BB, BG, BR, BY, BZ, CA, CH, CN, CO, CR, CU,
`CZ, DE, DK, DM, DZ, EC, EE, ES, FI, GB, GD, GE, GH,
`(21) International Application Number:©PCT/US03/16429
`GM,HR,HU,ID,IL,IN, IS, JP, KE, KG, KP, KR, KZ, LC,
`LK, LR, LS, LT, LU, LV, MA, MD, MG, MK, MN, Mw,
`(22) International Filing Date:=23 May 2003 (23.05.2003)
`Mx, MZ, NI, NO, NZ, OM, PH, PL, PT, RO, RU, SD, SE,
`SG, SK, SL, TJ, TM, TN, TR, TT, TZ, UA, UG, UZ, VN,
`YU, ZA, ZM, ZW.
`
`(25) Filing Language:
`
`English
`
`(26) Publication Language:
`
`English
`
`(30) Priority Data:
`60/383,559
`
`24 May 2002 (24.05.2002)
`
`US
`
`(71) Applicant: NIMBLEGEN SYSTEMS,INC. [US/US]; 1
`Science Court, Madison, WI 53711 (US).
`
`(84) Designated States (regional): ARIPO patent (GH, GM,
`KE, LS, MW, MZ, SD, SL, SZ, TZ, UG, ZM, ZW),
`Eurasian patent (AM, AZ, BY, KG, KZ, MD, RU, TJ, TM),
`European patent (AT, BE, BG, CH, CY, CZ, DE, DK, EE,
`ES, FI, FR, GB, GR, HU, IE, IT, LU, MC, NL, PT, RO,
`SE, SI, SK, TR), OAPI patent (BF, BJ, CF, CG, CI, CM,
`GA, GN, GQ, GW, ML, MR, NE, SN, TD, TG).
`
`Published:
`
`(72) Inventors; ALBERT, Thomas; 3554 Heather Crest,
`Madison, WI 53705 (US). MCCORMICK,Mark; Madi-
`son, WI (US).
`
`without international search report and to be republished
`upon receipt of that report
`
`(74) Agent: SEAY,Nicholas, J.; Quarles & Brady LLP, P.O.
`Box 2113, Madison, WI 53701-2113 (US).
`
`For two-letter codes and other abbreviations, refer to the "Guid-
`ance Notes on Codes andAbbreviations" appearing at the begin-
`ning ofeach regular issue ofthe PCT Gazette.
`
`(54) Title: MICROARRAYS AND METHOD FOR RUNNING HYBRIDIZATION REACTION FOR MULTIPLE SAMPLES ON
`A SINGLE MICROARRAY
`
`(57) Abstract: The present invention provides a microarray for multiple sample analysis that does not require a alignment of well
`walls with corresponding probe sets. Methods for building and using such a microarray are also within the scope of the present
`invention.
`
`/100012A2
`
`en
`
`O00
`
`

`

`WO 03/100012
`
`PCT/US03/16429
`
`MICROARRAYS AND METHOD FOR RUNNING HYBRIDIZATION REACTION FOR
`
`MULTIPLE SAMPLES ON A SINGLE MICROARRAY
`
`CROSS-REFERENCE TO RELATED APPLICATIONS
`This application claims priority from U.S. provisional patent application number
`[0001]
`60/383,559 filed May 24, 2002.
`
`STATEMENT REGARDING FEDERALLY SPONSORED RESEARCH
`
`OR DEVELOPMENT
`
`[0002]
`
`Notapplicable.
`
`BACKGROUND OF THE INVENTION
`The advent of the DNA microarray technology makesit possible to build an array
`{0003]
`of hundreds of thousands of DNA sequences, or probes, in a very small area, typically a few
`square centimeters on the surface of a microscopicslide. See, e.g., PCT patent publication Nos.
`WO 99/42813, 92/10092 and 90/15070, and U.S. Pat. No. 5,143,854, each ofwhich is hereby
`incorporated by referencein its entirety. A DNA microarray-based assay usually involves
`hybridizing a DNA or RNA sample to a microarray and scanning the microarray to detect
`hybridization. The probes in the microarray are organized into areas of similar probes, these
`areas being referred to as features. By hybridizing experimental DNA or RNAto the probes on
`the microarray, and detecting in which features the experimental DNA or RNAhashybridized,it
`becomespossible to obtain much information about the experimental DNA or RNA in a single
`step relatively simple process. Using this ability, DNA microarray technology has been applied
`to areas such as gene expression and discovery, mutation detection,allelic and evolutionary
`sequence comparison, genome mapping, and more.
`[0004]
`A state ofthe art DNA microarray can accommodate hundreds ofthousands of
`features, each containing unique probes. In fact this capacity can exceed the needs ofmany
`commonuseful experiment, many of whichinvolve hybridization assays involving far less
`probes than a microarray’s full capacity. Therefore, some microarrays are constructed in which a
`set of features are repeated multiple times over the area of the microarray, with each set of
`features ultimately being exposed to a separate experimental sample, to conduct multiple data
`collection experiments in parallel. This concept can be thought of as an array of arrays. To do
`this, it is desirable to make DNA microarrays that can be simultaneously used for multiple
`
`-1-
`
`

`

`WO 03/100012
`
`PCT/US03/16429
`
`samples. To make this work, there must be measurestaken to prevent cross-contamination
`between samples intended for differing areas of the microarray. Currently, the microarrays built
`for this purpose(e.g., U.S. Patent No. 5,874,219) use physical wells to separate probe sets for
`different samples and well walls have to align with correspondingprobesets so that each well
`contains the correct probes. However, alignment ofwell walls with corresponding probesets is
`not always easy to achieve and a misalignmentcan lead to inaccurate result.
`
`BRIEF SUMMARY OF THE INVENTION
`
`Thepresentinvention provides a microarray for multiple sample analysis that
`[0005]
`does not require an alignmentofwell walls with correspondingprobe sets. This is achieved by
`providing a microarray that contains continuous andidentical detection blocks (each detection
`block contains a set of probesofinterest) and a signal for identifying a corner point where any
`four adjacent detection blocks connect. In addition, each well used to separate probes on the
`microarrayis slightly larger than a detection block in all dimensions. With such an arrangement,
`evenif well walls do not align with the detection blocks, each well still containsall the probes of
`a complete detection block and the identity of the probes in a well can be determined by referring
`to a comerpoint also contained in the well.
`[0006]
`The microarray of the present inventionis not limited by the type of molecules on
`the microarray. For example, the microarray can be a polynucleotide microarray, a polypeptide
`microarray or a microarray ofother types of molecules. Methods for building and using the
`microarrays of the present invention are also within the scope of the invention.
`
`BRIEF DESCRIPTION OF THE SEVERAL VIEWS OF THE DRAWINGS
`Fig. 1 depicts one microarray embodimentofthe present invention that contains
`[0007]
`continuous and identical detection blocks.
`[0008]
`Fig. 2 shows an example ofa detection block from the microarray in Fig.1.
`[0009]
`Fig. 3 illustrates how information from a complete detection block is pieced
`
`together.
`
`DETAILED DESCRIPTION OF THE INVENTION
`Theintention of the present invention is to overcomethe problem ofaligning
`[00010]
`physical barriers, such as wells, with areas of a microarray,so asto facilitate the use of a
`microarray to perform multiple parallel hybridization procedures on a single microarray. In fact,
`the idea ofthe present inventionis to forego entirely any attempt to align physical barriers, such
`as wells, with the features or areas of features on the microarray. Instead, the wells or barriers
`
`-2-
`
`

`

`WO 03/100012
`
`PCT/US03/16429
`
`are placed on the microarray without any pre-determined alignment. Just by selecting a proper
`well size in relation to the size of the microarray areas, one can besurethat the needed data can
`be collected. The concept behind this inventionis that the relationship ofthe barriers or wells to
`the areas of features is determined after the hybridization data is collected, by analysis of the
`hybridization data, rather than trying to physically align the wells with the microarray before the
`experiment. This technique can be thought of as an alignment donein software with the data
`from the hybridization rather than an alignment done physically before the hybridization.
`[00011]
`To understand this concept, the introduction of some terminology is helpful.
`Again, a feature is a physical area on the microarray in which a numberof nucleic acid probes of
`similar sequenceareall anchored. For the purposeofthe present invention, a detection blockis
`an area on a microarray slide that contains one completeset of the features ofinterest that are to
`be probed with the experimental sample. Thusifthe experimentis to use 264 features (an 8 by 8
`set of features), the detection block would mean one ofthe 8 by 8 feature areas containing a
`complete set of the 264 features of interest. The size of a detection block can be any number of
`features suchthat there can be more than one detection block on a microarray. By “continuous
`
`detection blocks,” we mean that the detection blocks adjacent to each other share borders. The
`idea behind this concept is that the detection blocksare fabricated on the microarray in repeating
`units nextto each other, such that the detection blocks map over the microarray. If the detection
`blocks are square or rectangular, each detection block andthree adjacent detection blocks share a
`vertex at each cornerofthe detection block. This vertex is here referred to as a corner. By
`
`“<dentical detection blocks,” we mean that the probesin the features of interest in each detection
`block are identical in identity and arrangement. It should be noted that the term “identical
`blocks” are defined with regard to the probesof interest for an assay. For example, two detection
`blocks that contain different probes that are notof interest for a particular assay at corresponding
`positions will be considered as identical blocksif they otherwise qualify as identical blocks. A
`detection block mayalso contain blank positions(a position available for a probe butis left with
`no probe). A detection block mayalso contain features designated for control or features
`designated for fiducial alignment purposes. An detection block can contain any desired number
`of features, so long as it can be repeated in the area ofthe microarray. The detection block need
`not be square or rectangular, but could be any geometry that lendsitself to both arrangement on
`the microarray and detection of the fiducial boundaries as envisionedhere.
`[00012]
`Using the technique described here, ultimately set of features on a microarray of
`the present invention are compartmentalized by physical barriers from other features. A
`compartment so formedis also called a well. Each wellis at least slightly larger than a detection
`
`-3-
`
`

`

`WO 03/100012
`
`PCT/US03/16429
`
`block in all dimensions andis notin fluid communication with other wells during the
`hybridization process, so that hybridization reaction in one well does notinterfere with that in
`another. The exact way the probes are compartmentalized is not critical for the present
`invention. For the purposeof the present invention, the shape of each well does not matter and
`can be uniform or varying. Thecloser the shapeof a wellis to the shape ofa detection block,the
`more wells can be formed on a givenslide.
`[00013]
`As mentioned earlier, an advantage of the microarray provided by the present
`inventionis that the physical barriers do not have to be physically aligned with detection blocks.
`No matter where a well is located on a microarray, as long as the well is slightly larger than a
`detection block in all dimensions, it will contain a complete set of features forming a detection
`block and a corner point where four adjacent detection blocks intersect. Althoughtherelative
`locations of the complete set of features in the well maybedifferent from that in another well,
`the identity of the features andtheposition ofthe detection block in the well can always be
`determined by locatingfirst the corner point contained in the well. As long as the detection block
`are formed in a continuous grid on the microarray, and as longas the well is sufficiently larger
`than the detection block, the area inside of the well will inevitable contain at least one corner
`where four detection blocks meet. The ideahereis that in order to create a complete data set for
`the sample in each well, the detection block is created by assemble an complete data set of
`features from the features that surround a corner. Thus for the data collection purposes, a virtual
`
`detection block offeatures is created from the feature surround the corner rather using the
`
`physical detection block designed whenlaying outthe microarray.
`[00014]
`The corneris thus detected when the output ofthe hybridization step is read,
`typically by fluorescent scanning. There are many techniques by which a corner point can be
`identified, and one can use any of them for the present invention. For example, one or more
`positive control probescan be printed onto features arranged on a microarray to identify a corner
`point. One easy wayis to arrange controlor fiducial features to a physical cross formation, using
`the positive control probes, such that the cross intersection coincides with the corner point. One
`can readily envision any numberofothervariations using combinations of negative and positive
`controls that could be usedto create a visible pattern that could be used to identify a corner.
`[00015]
`Oncethe corneris detected, the rest of the hybridization data can be rearranged to
`create complete set of data for a detection block by using the data from the features which
`surround the corner. The location of each feature relative to the corner can be usedto identify
`whichprobesare in which features. One can think of this process are re-arranging the location of
`the features in software to reassemble an entire detection block.
`
`-4.
`
`

`

`WO 03/100012
`
`PCT/US03/16429
`
`The present invention is most useful for an application in which a hybridization
`[00016]
`assay is used to analyze a large numberof samples with the samerelatively small numberof
`probes. Described belowis a preferred microarray embodimentofthe present invention for such
`an application.
`In this embodimentasillustrated in Fig. 1, the microarray contains 64 identical
`[00017]
`blocks labeled as block A, B, C, D, E and so on. Circles 1 and 2 represent two wells onthe
`microarray (Fig. 1). Fig. 2 is an example ofwhat oneofthe identical detection blocks lookslike.
`In this simplified example, there are 12 available feature positions in the block and the number of
`features of interest from which data is sought in the hybridization assayis 10, leaving two
`features for controls or fiducials. The first number in each numberset shown in Fig. 2 defines
`feature position of the block and the second number defines a numberassigned to each probe.
`The 10 probes ofinterest are labeled as probe 1-10 are located at feature positions 1 to 8, 10 and
`11, respectively. The feature at feature position 9 does not have any probes constructed in it, and
`thusis a blank position or negative control. The feature at feature position 12 contains a probe
`which is not ofinterest to the hybridization, but which will hybridize to a nucleotide spiked into
`the experimental sample,i.e. a positive control. All the repeating detection blocksof the
`microarray in Fig. 1 have the same arrangementoffeatures at corresponding positions1 to 12.
`[00018]
`Figs. 1 and3 illustrate how information in a complete block is reconstructed by
`taking partial information from adjacent blocks. In Fig. 1, well 1 contains a complete detection
`block A and thus contains all information from one block. The information from the
`
`hybridization in well 1 is read by determining the comer, from the location of the corner
`understanding that the entire detection block A is in well 1, and then simply reading the data
`
`from detection block A.
`[00019]
`Well 2, however, does not contain any single whole detection block. However,it
`does contain the information from a whole block that can be obtained by piecing together
`
`information contained in the parts of block B-Ethat are enclosed in well 2. First, the corner at
`the intersection of detection blocks B, C, D, and E is located. In this simple example, the corner
`might be located by looking for a lighted feature (feature position 12 ofblock B) adjacentto a
`dark feature (feature 9 of detection block C). From the detection of the corner, a data extraction
`grid 3 of the samesize as a detection block can be envisioned inside well 2 (Fig. 3). The grid is
`placed in such a way so that a corner point 4 where the four blocks connect to each other is
`included in the grid. The data from all of the featuresin the grid can be then read and assembled
`into a data set representing on detection block.
`
`

`

`WO 03/100012
`
`PCT/US03/16429
`
`The methodology used for detecting the corner in this simple exampleistrivial
`{00020}
`and subject to ambiguity, since there will likely be other instancesof a lighted feature next to a
`dark feature in the data set. This example wasjust to illustrate the principle. In actual practice,
`the numberoffeatures in the data set and the construction ofcontrols or fiducials can be as
`
`elaborate as necessary to allow for the unambiguousdetection of corners. Notealso that since
`someofthe features will appear multiple times in a well, that redundant reading and comparison
`of the other features, even features outside of the grid 3, can be used to confirm that the corner
`
`has been found and that the detection block is being read correctly.
`[00021]
`In view of the microarrays and methodsfor piecing together information
`described above, it is well within the capability of one of ordinary skill in the art to build a
`microarray ofthe present invention and use it to analyze multiple samples.
`[00022]
`It is understood that examples and embodimentsofthe present invention set forth
`in the specification are illustrative and not intended to confine the invention. The invention
`embraces all modified forms of the examples and embodiments as come within the scopeofthe
`
`following claims.
`
`

`

`WO 03/100012
`
`PCT/US03/16429
`
`CLAIMS
`
`WE CLAIM:
`
`1.
`
`A microarray comprising:
`a plurality of continuous detection blocks wherein each detection block contains
`the sameset of features containing nucleic acid probes of interest constructed on a microarray
`
`slide, the set of features in each detection block being identically arranged.
`
`The microarray of claim 1 further comprising positive control features in each
`2.
`detection block at the corners thereof for identifying a corner point where four adjacent detection
`
`blocks connect.
`
`3.
`
`4.
`
`The microarray of claim 1, wherein the probes are polynucleotides.
`
`The microarray of claim 1, wherein the probes are polypeptides.
`
`5.
`
`The microarray of claim 1, further comprising:
`a plurality of wells formed by physical barriers connected to the microarray slide
`wherein the size of each well is larger than one detection block in all dimensions, and the wells
`are not in fluid communication with one another during the hybridization process, the plurality of
`
`wells placed on the microarray without alignment with features on the microarray.
`
`6.
`
`7.
`
`The microarray of claim 5, wherein the wells are of the same shape.
`
`The microarray of claim 5, wherein the shape ofthe wells is selected from a
`
`rectangle, a hexagon andacircle.
`
`8.
`
`The microarray of claim 5, wherein the physical barriers can be separated from the
`
`microarrayslide.
`
`9.
`
`The microarray of claim 6, wherein a plurality of detection blocksis at least 4
`
`blocks.
`
`-7-
`
`

`

`WO 03/100012
`
`PCT/US03/16429
`
`10...
`
`The microarray of claim 5, wherein a plurality of detection blocksis at least 10
`
`blocks.
`
`slide.
`
`12.
`
`Amethod for building the microarray of claim 1 comprising the steps of:
`
`selecting a detection block of features containing probesofinterest;
`
`positioning one detection block in one region of the microarray slide; and
`
`positioning another detection block offeatures in another region of the microarray
`
`13.|Amethod for building the microarray of claim 5 comprisingthesteps of:
`
`selecting detection block of features containing probesofinterest;
`
`providing a microarrayslide that is pre-patterned with wells formed by physical
`barriers wherein each well contains an area sufficiently large to accommodate the detection
`
`block;
`
`positioning detection block in one region of the microarray slide; and
`
`positioning another detection block in another region of the microarray slide.
`
`14. A method of conducting multiple hybridization experimentsin parallel on a single
`microarray, the method comprising the steps of
`
`providing a microarray which has a plurality of detection blocks, each detection block
`
`having a plurality of similar features each of the features having a set of defined probes in them;
`physically placing a set of wells on the microarray without aligning the wells to the
`
`detection blocks, the wells each being larger in area than a detection block;
`
`placing different samples in different of the wells;
`
`performing a hybridization experiment;
`
`obtaining the data from the hybridization experiment;
`
`from the data from the hybridization experiment, determining wherethe corner of a
`
`detection block from the data for a well is, and combining the data from features near that corner
`
`to reconstruct the data from an entire detection block.
`
`15. The methodof claim 14 wherein each detection block includes control features
`
`adjacent the corners thereof to assist in locating the corner of each block.
`
`

`

`WO 03/100012
`
`PCT/US03/16429
`
`16.|Amethod for running a plurality of hybridization reactions for a plurality of
`
`samples on a single microarray, the method comprisingthestepsof:
`
`providing a microarray according to claim 6;
`
`hybridizing one sample to probes in one detection block; and
`
`hybridizing another sample to probes in another detection block.
`
`-9.
`
`

`

`WO 03/100012
`
`PCT/US03/16429
`
`FIG 1
`
`TENT
`SPDCCC
`
`
`
`N©i FIG 3
`
`
`