Ex Parte Batzoglou et alDownload PDFPatent Trial and Appeal BoardJun 23, 201612129330 (P.T.A.B. Jun. 23, 2016) Copy Citation UNITED STA TES p A TENT AND TRADEMARK OFFICE APPLICATION NO. FILING DATE 12/129,330 05/29/2008 40581 7590 06/27/2016 CRAWFORD MAUNU PLLC 1150 NORTHLAND DRIVE, SUITE 100 ST. PAUL, MN 55120 FIRST NAMED INVENTOR Serafim Batzoglou UNITED STATES DEPARTMENT OF COMMERCE United States Patent and Trademark Office Address: COMMISSIONER FOR PATENTS P.O. Box 1450 Alexandria, Virginia 22313-1450 www .uspto.gov ATTORNEY DOCKET NO. CONFIRMATION NO. STFD.209PA (S06-059) 1916 EXAMINER BRUSCA, JOHNS ART UNIT PAPER NUMBER 1631 NOTIFICATION DATE DELIVERY MODE 06/27/2016 ELECTRONIC Please find below and/or attached an Office communication concerning this application or proceeding. The time period for reply, if any, is set in the attached communication. Notice of the Office communication was sent electronically on above-indicated "Notification Date" to the following e-mail address( es): USPTO-patent@ip-firm.com PTOL-90A (Rev. 04/07) UNITED STATES PATENT AND TRADEMARK OFFICE BEFORE THE PATENT TRIAL AND APPEAL BOARD Ex parte SERAFIM BATZOGLOU, MOST AF A RONAGHI, and ANDREAS SUNDQUIST Appeal2013-000784 Application 12/129,3301 Technology Center 1600 Before DEMETRA J. MILLS, RICHARD M. LEBOVITZ, and JEFFREY N. FREDMAN, Administrative Patent Judges. LEBOVITZ, Administrative Patent Judge. DECISION ON APPEAL This appeal involves claims directed to a method of genome sequencing. Appellants appeal from the Examiner's rejections of 1-16 as obvious under 35 U.S.C. § 103. We have jurisdiction under 35 U.S.C. § 134. The rejections are affirmed but have been designated as new grounds of rejection under 37 C.F.R. § 41.50(b). STATEMENT OF CASE Claims 1-16 are pending. The claims stand rejected by the Examiner as follows: 1 "The '330 Application." Appeal2013-000784 Application 12/129,330 1. Claims 1, 5-13, 15, and 16 under 35 U.S.C. § 103(a) (pre-AIA) as obvious in view of Meyers (Meyers et al., "A Whole-Genome Assembly of Drosophila," 287 Science 2196-2204, 2002) and Venter (Venter et al., "A new strategy for genome sequencing," 381Nature364--366, 1996). Answer 2. 2. Claims 1, 2, and 3 under 35 U.S.C. § 103(a) (pre-AIA) as obvious in view of Meyers, Venter, and Konno (Konno et al., "Computer-Based Methods for the Mouse Full-Length cDNA Encyclopedia: Real-Time Sequence Clustering for Construction of a Nonredundant cDNA Library," 11 Genome Res. 281-289, 2001). Answer 5. 3. Claims 1, 4, 9, and 14 under 35 U.S.C. § 103(a) (pre-AIA) as obvious in view of Meyers, Venter, and Giardine (Giardine et al., "Galaxy: A platform for interactive large-scale genome analysis," 15 Genome Res. 1451-1455, 2005). Answer 5. Claims 1 and 9 are representative. Claims 1 and 9 are reproduced below. We have annotated them with bracketed numerals to number the steps of the claims. 1. A method for genome sequencing, the method comprising: [ 1] as a random function, selecting a subset of fragments of a target genome, the each fragment having a clone length; [2] replicating each fragment of the subset of fragments into clones; [3] ordering the clones into clone contigs based on sets of shared k-mers, the clone contigs including overlapping clones; [ 4] determining, for each clone contig, potential read overlaps from clone read data and validating base pairs of each read; [5] generating read sets for each clone contig by selecting reads from other clones in the clone contigs, the reads 2 Appeal2013-000784 Application 12/129,330 corresponding to regions smaller than the clone length of the clone, and [ 6] assembling the selected reads into read sets; [7] combining the assembled read sets into clone-sized regions; and [8] assembling the clone-sized regions into clone contigs. 9. A method for genome sequencing that uses clones generated from a subset of fragments of a target genome and ordered into clone contigs based on sets of overlapping clones, the method comprising: [9] reading local assemblies of contigs from clone regions smaller than a clone length and assembling the local assemblies into read sets; [ 10] using a programmed computer to combine the assembled read sets into clone-sized regions; and [ 11] assemble the clone-sized regions into clone contigs. Sundquist publication The '330 Application was filed May 29, 2008, listing Serafim Batzoglou, Mostafa Ronaghi, and Andreas Sundquist as the named inventors. A publication listing the three named inventors as authors, in addition to Haixu Tang and Pavel Pevzner, was published on May 30, 2007. Sundquist et al., "Whole-Genome Sequencing and Assembly with High- Throughput, Short-Read Technologies, PLoS ONE, 2(5): e484, 2007. The Sundquist publication appears to disclose the same method described and claimed in the '330 Application. Because the publication was not published more than a year before the application filing date, it does not bar patentability of the claimed subject matter under 35 U.S.C. § 102(b) (pre-AIA). However, 35 U.S.C. § 102(a) (pre-AIA) is applicable. Under 35 U.S.C. § 102(a) (pre-AIA), ifthe publication is the inventors' own work, it 3 Appeal2013-000784 Application 12/129,330 cannot be used against them. MPEP § 2132.01 (11). In this case, the Sundquist publication states that Serafim Batzoglou, Mostafa Ronaghi, and Andreas Sundquist "Conceived and designed the experiments," and that Haixu Tang and Pavel Pevzner "Contributed reagents/materials/analysis tools." Sundquist e484. Based on this disclosure, it appears that the Sundquist publication is the work is the named inventors, and not the additional two authors listed on the publication. Consequently, the Sundquist publication is not prior art under 35 U.S.C. § 102(a) (pre-AIA). 1. OBVIOUSNESS BASED ON MYERS AND VENTER Steps [1] through [3] of the claimed method involve a hierarchical sequencing type approach where clones are ordered based on the "k-mer content (the set of all sequences of k bases)" of sequence reads from each clone. '330 Appl. 5: 7-16; 6: 10-15. The '330 Application discloses that hierarchical sequencing in which BAC clones are ordered had been performed prior to the claimed invention. Id. at 2: 1-12. Venter, as cited by the Examiner, describe an approach where BAC clones are selected ("selecting a subset of fragments of a target genome" as in step [ 1 ]" of claim 1) and cloned ("[2] replicating each fragment of the subset of fragments into clones" of claim 1 ). Our new approach to genomic sequencing eliminates the need for any prior physical mapping and uses BAC clones as the basic sequencing reagent (see the figure). A human BAC library with an average insert size of 150 kb and about 15-fold coverage of the human genome contains 300,000 clones. These are arrayed into microtitre wells. Venter 365 (cols. l----2). 4 Appeal2013-000784 Application 12/129,330 The clones are selected as a '"random function," as required by claim 1, because Venter explicitly teaches that the clones are selected without "the need for any prior physical mapping." Id. Appellants' arguments about the prior art not describing a "random function" in selecting fragments is thus not supported by a preponderance of the evidence. Appeal Br. 6. Venter describes connecting the BAC clones to each other based on sequence reads from the ends of each clone, Both ends (starting at the vector-insert points) of each BAC clone are then sequenced to generate 500 bases from each end. The 600,000 BAC end sequences are scattered roughly every 5 kb across the genome and make up 10 per cent of the genome sequence. vVe denote them 'sequence-tagged connectors', or STCs, because they allow any one BAC clone to be connected to about 30 others (for example, a 150-kb insert 'divided' by 5 kb will be represented in 30 BACs). Venter 365 (col. 2). The end sequence reads described in Venter represent the "k-mers" as in claim 1, and the sequence reads are utilized to connect the BAC clones to other BAC clones, meeting the claim limitation of "[3] ordering the clones into clone contigs based on sets of shared k-mers, the clone contigs including overlapping clones." Steps [ 4] through [8] of claim 1 accomplish the nucleotide sequencing of the genome. As found by the Examiner, Meyers teaches a whole genome assembly approach utilizing sequence reads from BAC clones. Meyers 2196, col. 3 (paragraph beginning, "For Drosophila.) Meyers describes comparing sequence reads (called "fragments" in Meyers, para. Bridging 2197-2198) and identifying overlaps ("Overlapper", Fig 1 ). Each fragment/reads was compared with all fragments/reads previously examined in search of fragment/reads overlaps., meeting the 5 Appeal2013-000784 Application 12/129,330 limitation of step [ 4] of "determining, for each clone contig, potential read overlaps from clone read data." Meyers 2197-2198. Meyers teaches that the overlaps are utilized to assemble the sequence reads, namely fragments, into "unitags." Id. at 2198 ("Unitigger. Collections of fragments whose arrangement is uncontested by overlaps from other fragments were assembled into what we call unitigs.") This step corresponds to claimed step [5] of "generating read sets for each clone contig by selecting reads from other clones in the clone contigs" (i.e., "fragments"), step [ 6] of "assembling the selected reads into read sets" (i.e., "Collections of fragments"), and step [7] of "combining the assembled read sets into clone-sized regions" (i.e., assembling into "unitigs"). The claim requires that reads sets are assembled from "clones" (step [5] which we interpret to mean physical DNA clones, such as BAC clones. Meyers in the sections titled "Overlapper" and "Unitigger" refer to assembling fragments which are sequence reads: "Almost every unitig is a correct subassembly of fragments." Meyers 2198. However, since the sequence reads of Meyers are obtained from BAC clones, the claimed steps of "[5] generating read sets for each clone," and then "[6] assembling" and "[7] combining" the read sets is necessarily performed on at least some of the BAC clones from which the read sets are derived. In the last step [8] of the claimed method, the read sets from the clone-size regions are assembled into clone contigs, i.e., more than one clone is assembled into a clone contig. Meyers teaches the "Scaffolder" step in which unitigs with "BAC ends were linked into scaffolds consisting of a set of ordered, oriented contigs." Id. at 2198-2199. Meyers teaches: 6 Appeal2013-000784 Application 12/129,330 ... all sets of U-unitigs that were consistently ordered and placed by confirmed bundles, that is, containing two or more 2- kbp or 10-kbp links, were assembled into a scaffold of contigs where a contig is, at this stage, a series of overlapping U- unitigs. We then ordered and placed these scaffolds using a best-first selection of BAC bundles (that is, one involving a BAC mate) ordered on the number of links in the bundle. Id. at 2199 This passage shows assembling unitigs using "two or more 2-kbp or 10-kbp links" into a scaffold. The links appear to be sets of sequence reads which are smaller than the BAC clones, and thus corresponds to steps [ 5] and [6] of the claimed method. The quoted passage expressly refers to BAC clones, and thus meets the limitation of the claims that is based on sequence reads from clones. The clones are "ordered" and used to create scaffolds, which also corresponds to step [8] of "assembling the clone-sized regions into clone contigs." Thus, the "Scaff older" step of Meyers also meets the limitation of steps [5] through [8] of claim 1. Appellants contend that the Meyers does not order clones (Appeal Br. 5; Reply Br. 2-3). However, as the above-quoted passage indicates, the sequence reads ("fragments") are derived from clones and thus implicitly clones are ordered. Meyers also explicitly teaches refers to ordering BAC clones in the "Scaffolder" step. Consequently, we do not find Appellants arguments to have demonstrated error in the rejection. To the extent that the ordered sequence reads in Meyer are not necessarily from a given BAC clone, the concept of ordering clones using read sets of sequence data was known in the art as taught by Pevzner2 which 2 Pevzner et al., "An Eulerian path approach to DNA fragment assembly," Proc. Natl. Acad. Sci., 98(17): 9748-9753, 2001. 7 Appeal2013-000784 Application 12/129,330 describes the Euler algorithm utilized by the inventors to assemble the sequence reads into contigs ('330 Appl. 7: 2--4). Specifically, the abstract Pevzner teaches: "For the last 20 years, fragment assembly in DNA sequencing followed the "overlap-layout-consensus" paradigm that is used in all currently available assembly tools. Although this approach proved useful in assembling clones, it faces difficulties in genomic shotgun assembly." Claim 6 Claim 6, depends from claim 1, and requires "selecting a subset of fragments that cover the genome at high redundancy of at least about 4.0x coverage" and acquiring sequence reads at a redundancy that is lower than "said high redundancy." Appellants contend that the Examiner has not addressed this limitation. Appeal Br. 11. Myers describes 15x coverage of the Drosophila genome which meets the limitation of at least about 4x coverage. Myers 2196 (col. 3). There is no indication that each sequence read is covered 15x in Meyers' s method. Consequently, the actual redundancy analyzed by Meyers would be expected to be less than the 15x coverage, meeting the corresponding limitation of the claim. Claim 9 Independent claim 9 has following limitations: [9] reading local assemblies of contigs from clone regions smaller than a clone length and assembling the local assemblies into read sets; 8 Appeal2013-000784 Application 12/129,330 [1 OJ using a programmed computer to combine the assembled read sets into clone-sized regions; and [ 11] assemble the clone-sized regions into clone contigs. Myers teaches the recited limitations. Specifically, Myers describes reading links of 2-kbp or 10-kbp (id. at 2196, col. 3; 2199, col. 3), which are [9] "smaller than a clone length" because the reads are less than the 100- to 150-kbp size of a BAC clone (id. at 2196, col. 2). The sequence reads, smaller than a clone length, are assembled into read sets of overlapping sequences ("Overlapper. Each fragment was compared with all fragments previously examined in search of overlaps ... " Id. at 2198, col. 1-2). The read sets, using a computer program are assembled into "clone contigs" as required by steps [10] and [ 11] of claim 9 (" Unitigger. Collections of fragments whose arrangement is uncontested by overlaps from other fragments were assembled into what we call unitigs." Id. at 2198, col. 2). Alternatively, where the "clone sized regions" represent BAC clones (step [11]), Meyers also describes a scaffolding procedure which meets the claimed limitations. Meyers teaches assembling BAC clones ("clone sized regions") into a scaffold ("clone contigs") . . . . all sets ofU-unitigs that were consistently ordered and placed by confirmed bundles, that is, containing two or more 2- kbp or 10-kbp links, were assembled into a scaffold of contigs where a contig is, at this stage, a series of overlapping U- unitigs. We then ordered and placed these scaffolds using a best-first selection of BAC bundles (that is, one involving a BAC mate) ordered on the number of links in the bundle. Id. at 2199. 9 Appeal2013-000784 Application 12/129,330 While the terms utilized in Meyers are different from those used in the claim ("contigs" versus "scaffold"), Appellants did not provide argument, evidence, or definitions from the Specification which would distinguish the claimed subject matter from Meyers. Consequently, the step of ordering BAC clones into a scaffold meets the claimed limitation of step [ 11] "assemble the clone-sized regions into clone contigs." OBVIOUSNESS BASED ON MYERS, VENTER, AND KONNO Claims 2 and 3, depend from claim 1, and further require tagging the clones with unique identifiers. The Examiner cited Konno for teaching tagged clones to avoid identification error. Answer 5. Appellants contends that the Examiner did not provide evidence nor an explanation as to how Konno's tags would be utilized in Meyers's process. Reply Br. 6. We interpret claims 2 and 3 to require that the clones are physically tagged with a clone ID. For example, the '330 Application discloses using unique 5-base tags in the claimed method. '330 Appl. 8: 26-31. The Examiner did not provide evidence that Konno tagged its clones. Konno describes sequencing the 3' ends of cDNAs and using these sequences as "tags," clustering them into groups. Konno, Abstract; 283 ("Clustering"). The "tag" sequences are obtained by sequencing the clones ("The first step is a small-scale clustering of the tag sequences that were sequenced during the previous day."). Id. Thus, the Examiner did not meet the burden showing that the limitations in claims 2 and 3 are met by Konno. However, sequence tags appear to be conventionally utilized in sequencing, particularly 454 Sequencing utilized by the inventors. '330 Appl. 8: 26-31; Binladen et al., "The Use of Coded PCR Primers Enables 10 Appeal2013-000784 Application 12/129,330 High-Throughput Sequencing of Multiple Homolog Amplification Products by 454 Parallel Sequencing," PLoS ONE 2(2): e197, Feb. 14, 2007. Consequently, we set forth a new grounds of rejection of claim 1, 2, and 3 as obvious in view of Meyer, Venter, Pevzner, and Binladen. OBVIOUSNESS BASED ON MYERS, VENTER, AND GIARDINE Claims 4 and 14 depend, from claims 1 and 9, and require performing intersection and subtraction operations on the sets of reads to isolate smaller regions. The Examiner found that Giardine describes a computer program which performs such analysis to avoid duplication when sequences from different clones are combined. Answer 5-6. Appellants contend that the Examiner erred in not showing how Giardine would be applied to Meyers' s sequencing method. Appeal Br. 10- 11. Appellants' argument does not provide sufficient evidence that the Examiner erred. Giardine describes the computer program Galaxy that provides sequence analysis tools. Giardine 1451. As explained by the Examiner, Galaxy would be used on the sequence data set of Myers for its benefit in enabling intersection and subtraction analysis. See, e.g., Giardine, 1451, col. 1. Appellants did not explain how this explanation is insufficient when Galaxy's purpose is to analyze genomic information from data sets, such as data obtained from Myers. Id. at 1451, 1452 ("Combining and comparing ENCODE data to find promoter") 11 Appeal2013-000784 Application 12/129,330 NEW GROUNDS OF REJECTION We set forth the following new grounds of rejection under 37 C.F.R. § 41.50(b). 1. Claims 1, 5-13, 15, and 16 under 35 U.S.C. § 103(a) (pre-AIA) as obvious in view of Meyers, Venter, and Pevzner. Answer 2. 2. Claims 1, 2, and 3 under 35 U.S.C. § 103(a) (pre-AIA) as obvious in view of Meyers, Venter, Pevzner, and Binladen. Answer 5. 3. Claims 1, 4, 9, and 14 under 35 U.S.C. § 103(a) (pre-AIA) as obvious in view of Meyers, Venter, Pevzner, and Giardine. Answer 5. NEW GROUNDS OF REJECTION When the Board designates a new ground of rejection under 37 C.F.R. § 41. 50(b ), the appellant, as to each claim so rejected, has the option of: ( 1) Reopen prosecution. Submit an appropriate amendment of the claims so rejected or new evidence relating to the claims so rejected, or both, and have the matter reconsidered by the examiner, in which event the proceeding \'l1ill be remanded to the examiner .... (2) Request rehearing. Request that the proceeding be reheard under 37 C.F.R. § 41.52 by the Board upon the same record .... The amendment and/ or new evidence under 3 7 C.F .R. § 41. 50(b )( 1 ), or the request for rehearing under 37 C.F.R. § 41.50(b)(2), must be filed within 2 months from the date of the Board's decision. In accordance with 37 C.F.R. § 41.50(±), this 2-month time period may not be extended by the filing of a petition and fee under 3 7 C.F .R. § 1.13 6( a), but only under the provisions of 3 7 C.F .R. § 1.136(b). AFFIRMED;§ 41.50(b) 12 Copy with citationCopy as parenthetical citation