DETAILED ACTION Notice of Pre-AIA or AIA Status The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA. Claim Status Claims 1, 6-8, 10-15, 18-23, 34-36, 39, 40, and 99 are pending. Priority This application is a 371 of PCT/ US2021 /016924, filed 02/05/2021, which claims benefit of application no. 63/038,691, filed 06/12/2020 and application no. 62/970,684, filed 02/05/2020. The instant application has the effective filing date of 05 February 2020. Information Disclosure Statement The information disclosure statement (IDS) submitted on 02/06/2023 is in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statement has been considered by the examiner. The following documents have not been considered by the examiner, as a copy does not appear to be provided: RU 2687451, EP 2877490, CN107236741 , CN 105802980, CN 105177038 , CN 106011104, CN 107099850 WO 2016049258, 2016057961, 2016057850, 2014204724 , 2015089046 , 2016049258, 2015148863, 2014207043, 2016179122, 2017192172, 2017190257, 2017197301, 2017197238 , 2019139645, 2018213351, 2019051097 , page 75 (lines 8-end), pages 76-77, page 78 (except line 17), Cox et al . (p. 109, line 8), Buckley et al. (p. 100) Dassa et al. (p. 111) , Hess et al. (p. 132), Lin et al. (p. 152), Osborn et al. (p. 166, line 9) . The following document has not been considered, as an English translation has not been provided: Dup u y et al. (p. 114) . Applicant is reminded that it is desirable to avoid the submission of long lists of documents if it can be avoided. As set forth in MPEP 2004, applicant is directed to eliminate clearly irrelevant and marginally pertinent cumulative information. If a long list is submitted, highlight those documents which have been specifically brought to applicant's attention and/or are known to be of most significance. See Penn Yah Boats, Inc. v. Sea Lark Boats, Inc., 359 F.Supp . 948, 175 USPQ 260 (S.D. Fla. 1972), aft'd , 479 F.2d 1338, 178 USPQ 577 (5th Cir. 1973), cert. denied, 414 U.S. 874 (1974). But cf. Molins PLC v. Textron Inc., 48 F.3d 1172, 33 USPQ2d1823 (Fed. Cir. 1995). Applicant has cited more than 200 pages of references and is cautioned against burying material references and the appearance of inequitable conduct in this application . Drawings The drawings, submitted on 08/04/2022, are accepted by the examiner. Specification: Abstract The abstract of the disclosure is objected to because it appears only the first page of a WIPO publication was provided. A corrected abstract of the disclosure is required and must be presented on a separate sheet, apart from any other text. See MPEP § 608.01(b). Claim Rejections - 35 USC § 101 35 U.S.C. 101 reads as follows: Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title. Claims 1, 6-8, 10-15, 18-23, 34-36, 39, 40, and 99 are rejected under U.S.C 101 because the y are directed to abstract ideas without significantly more, as detailed in the analysis below. Eligibility Step 1: Subject matter eligibility evaluation in accordance with MPEP § 2106: Claims 1, 6-8, 10-15, 18-23, 34-36, and 99 are directed to a statutory category ( method ). Claim 39 is directed to a statutory category ( product ). Claim 40 is directed to a statutory category ( system ). Therefore, in accordance with MPEP § 2106.03 , all claims have patent eligible subject matter. [Eligibility Step 1: YES] Eligibility Step 2A : This step determines whether a claim is directed to a judicial exception in accordance with MPEP § 2106. Eligibility Step 2A -- Prong One: Limitations are analyzed to determine if the claims recite any concepts that could equate to a judicial exception (i.e. abstract idea, law of nature, or natural phenomenon). Recitations of Judicial Exceptions: Claim s 1 , 39, 40 : A method of using at least one machine learning model to identify at least one guide RNA for use in a base editing system for introducing a desired change in a nucleotide sequence, (mental process) (ii) generating , from the first data indicative of the nucleotide sequence, data indicative of the set of guide RNAs; wherein the set of guide RNAs includes a first guide RNA, wherein the input data includes data indicative of at least a part of a nucleotide sequence associated with the first guide RNA, and wherein the data indicative of at least part of a nucleotide sequence associated with the first guide RNA specifies a spacer or a protospacer sequence associated with the first guide RNA; (mathematical concept) generating first input features from the input data; (mathematical concept) applying a first machine learning model to the first input features to obtain first output data indicative, for each guide RNA in the set of guide RNAs, of a base editing efficiency , at one or multiple locations in the nucleotide sequence, of the base editing system when using the each guide RNA, (mathematical concept) generating second input features from the input data; (mathematical concept) applying a second machine learning model to the second input features to obtain second output data indicative, for each guide RNA in the set of guide RNAs, of bystander editing activity , at one or multiple locations in the nucleotide sequence, by the base editing system when using each guide RNA; (mathematical concept) and identifying , using the first output data and the second output data, the at least one guide RNA for use in the base editing system for introducing the desired change in the nucleotide sequence. (mental process) Claim 6 : The method of claim 1, wherein the first machine learning model comprises a non-linear machine learning model selected from the group consisting of a random forest model, a logistic regression model, a support vector machine model, a generalized linear model, a hierarchical Bayesian model, and neural network model. (mental process ) Claim 7: The method of claim 1, wherein the first machine learning model comprises a random forest model. (mathematical concept) Claim s 8 and 19 : wherein generating the first input features comprises generating multiple features to include in the first input features, the multiple features including: (mathematical concept) features encoding at least some nucleotides in a protospacer sequence or spacer sequence associated with the first guide RNA; (mathematical concept) features encoding at least some nucleotides, in the nucleotide sequence, located within a threshold number of nucleotides of the protospacer sequence associated with the first guide RNA. (mathematical concept) Claim 10: The method of claim 8, wherein the multiple features further include one or more of the following features: features encoding at least some dinucleotides at neighboring positions in the protospacer sequence; (mathematical concept) features representing melting temperature of the first guide RNA; (mathematical concept) one or more features representing a total number of G, C, A, and/or T nucleotides in the protospacer sequence ( mathematical concept, m ental process ) one or more features representing a percentage of G, C, A, and/or T nucleotides in the protospacer sequence; (mathematical concept, mental process ) feature representing an average base editing efficiency of the base editing system. (mathematical concept , mental process ) Claim 11: wherein the first output data is indicative of a fraction of sequence reads containing at least one base edit at any nucleotide in a desired window about a protospacer sequence associated with the first guide RNA, among all sequence reads. (mathematical concept, mental process) Claim 12: The method of claim 1, wherein the second first machine learning model comprises a non-linear machine learning model selected from the group consisting of a random forest model, a logistic regression model, a support vector machine model, a generalized linear model, a hierarchical Bayesian model, and a neural network model. (mental process) Claim 15: an encoder neural network mapping input data to a latent representation; (mathematical concept) and a decoder neural network mapping the latent representation to output data, wherein the decoder neural network has an autoregressive structure. (mathematical concept) Claim 18: parameters representing a position-wise bias toward producing an unedited outcome. (mathematical concept) Claim 20: The method of claim 1, wherein the second output data is indicative of frequencies of occurrence of base editing outcomes, each of which includes edits to nucleotides at multiple positions. ( mathematical concept, mental process) Claim 21: The method of claim 1, wherein the second output data is indicative of a frequency distribution of combinations of base editing outcomes. ( mathematical concept, mental process) Claim 22: The method of claim 1, wherein, for a specific combination of base edits , the second output data is indicative of a frequency of occurrence of the specific combination of base edits among all sequenced reads containing at least one base edit at any nucleotide in a desired window about a protospacer sequence associated with the first guide RNA. ( mathematical concept ) Claim 23: t he method of claim 1, wherein the first output data includes a first base editing efficiency value for the first guide RNA, wherein the second output data includes a first bystander editing value for the first guide RNA, ( mathematical concept ) wherein identifying the guide RNA using the first output data and the second output data comprises multiplying the first base editing efficiency value by the first bystander editing value. (mathematical concept, mental process) Claim 36: The method of claim 1 further comprising: determining a likelihood of whether the identified guide RNA and the base editing system, when used in combination, will result in introducing the desired change in a cell. (mathematical concept, mental process) Claim 99: selecting a guide RNA for use in the base editing system in accordance with the method of claim 1 , (mental process) Step 2A – Prong One Analysis: Generating secondary data via multiplication , e ncoding , and applying machine learning algorithms with bias variables represent the transform ation and organiz ation of data using mathematical calculations . Therefore, claims that recite elements of this nature fall under then mathematical concepts grouping of abstract ideas. Selecting data based on available information and performing simple calculations that require no more than the human mind and pen/paper fall under the mental process grouping of abstract ideas. Some recited limitations merely provide additional information pertaining and depending on claims that recite abstract ideas and thus fall within the same respective grouping. Therefore, the claims are found to recite judicial exceptions. [Eligibility Step 2A – Prong One : YES ] Eligibility Step 2A – Prong Two: A claim that integrates a judicial exception into a practical application will apply, rely on, or use the judicial exception in a manner that imposes a meaningful limit on the judicial exception. If the claim contains no additional claim elements beyond the abstract idea, the claim fails to integrate the abstract idea into a practical application (MPEP 2106.04(d)). Additional elements within the claimed invention are recited and analyzed below. Data G athering Elements : Claim s 1 , 39, 40 : obtaining input data indicative of the nucleotide sequence and a set of one or more guide RNAs by ( i ) obtaining, by the software and from at least one source external to the software, first data indicative of the nucleotide sequence Computer Elements : Claim s 1, 39, 40: s oftware, c omputer readable storage medium s , or s ystem s comprising at least one computer hardware processor; at least one computer readable storage medium storing processor executable instructions that, when executed by the at least one computer hardware processor, cause at least one computer hardware processor Claim s 13 - 15 , and 18 : deep neural network compri sing a conditional autoregressive Base E diting Elements : Claim s 1 , 39, 40 : the base editing system comprising a napDNAbp and a deaminase. Claim 34 : The method of claim 1 further comprising: synthesizing the identified guide RNA for use in the base editing system for introducing the desired change in the nucleotide sequence Claim 35: The method of claim 1 further comprising: using the identified guide RNA and the base editing system to introduce the desired change in a cell. Claim 99: A method of editing a target DNA sequence by base editing using a base editor comprising: contacting the genome of the target DNA sequence with the selected guide RNA and the base editor, thereby editing the target DNA sequence. Step 2A – Prong Two Analysis: The noted data gathering elements represent insignificant extra-solution activitie s that do not integrate the judicial exception s into practical application per MPEP 2106.05(g). The computer elements of claims 1, 39, and 40 re cite implementations of a method onto a generic comput ing environment . They provide mere instructions to implement the abstract ideas per Alice Corp., 573 U.S. at 223, 110 USPQ2d at 1983. See also 573 U.S. at 224, 110 USPQ2d at 1984 . The computer elements of claims 13-15 and 18 describe a deep neural network model. However, as the model does not equate to a particular machine (2106.05( b )) , they also merely provide instructions to implement the abstract ideas of the claimed inventio n . The base editing elements such as s ynthesizing a guide RNA , performing a gen ome editing process (using said guide RNA and base editing system) , and contacting DNA , represent insignificant application s of the judicial exceptions per MPEP 2106.05 (g) . As such, the additional elements , when viewed separately and in the context of a whole claimed invention , do not integrate the judicial exceptions into practical application. [Eligibility Step 2A – Prong Two: No ] Eligibility Step 2B : Claim elements are probed for inventive concept equating to significantly more than the judicial exception (MPEP 2106.04(II)). Step 2 B Analysis : D ata gathering activities that obtain data , based on types of information and availability of information for further analysis and complete necessary data inputting and outputting are considered well-known and conventional within the art, as exemplified by Electric Power Group, LLC v. Alstom S.A., 830 F.3d 1350, 1354-55, 119 USPQ2d 1739, 1742 (Fed. Cir. 2016) and See Mayo, 566 U.S. at 79, 101 USPQ2d at 1968; OIP Techs., Inc. v. Amazon.com, Inc., 788 F.3d 1359, 1363, 115 USPQ2d 1090, 1092-93 (Fed. Cir. 2015) . Generic c omputing environments and neural networks that quantify efficiency and select optimal sgRNAs to aid genetic engineering are classified as well-understood, routine, and conventional per Wang et al. (RNA Biol. Vo. 17 (1); 2019), which reviews the topic , and Shen (Water Resources Research; Vol. 54 (11); 2018) which reviews deep learning models , such as C ANNs , and possible genomic applications of them. Base editing via a system of nucleic acid programmable DNA binding proteins and deaminase nucleic acid effector domains is well-understood, routine, and conventional within the art as evidenced by Rees et al. (IDS ref; NPL; p. 172, line 15; 2018) which reviews the process, and Schatoff et al. ( Methods; p. 164-165; PMCID : PMC6684841 ; 2019 ), which affirms conventionality. As such, the additional elements are further found to lack inventive concept. [Eligibility Step 2B : NO] C laims 1, 6-8, 10-15, 18-23, 34-36, 39, 40, and 99 are thus directed to judicial exceptions without significantly more and rejected under 35 U.S.C 101. Claim Rejections - 35 USC § 103 In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis ( i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status. The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action: A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made. The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows: 1. Determining the scope and contents of the prior art. 2. Ascertaining the differences between the prior art and the claims at issue. 3. Resolving the level of ordinary skill in the pertinent art. 4. Considering objective evidence present in the application indicating obviousness or nonobviousness . This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention. Claims 1, 6-8, 11 -12 , 19 -23, 36, and 39-40 are rejected under 35 U.S.C. 103 as being unpatentable over Hwang et al. ( IDS ref ; NPL; p. 135 , line 1; 2018) and Doench et al. ( IDS ref ; NPL; p. 113 , line 8; 2014) in view of Dandage et al. ( IDS ref ; NPL ; p. 110, line 13; 2018 ). Claim s 1 , 39, and 40 are directed to methods, computer readable storage mediums , and systems that perform the following steps: o btain first data , indicative of a nucleotide sequence and a set of at least one guide RNA s, via g enera t ing a set of one or more g uide RNA s that are associated with part of the nucleotide sequence and specif y a spacer or protospacer ; use the data to generate at least two sets of input features ; apply m achine learning M odel 1 to generate base editing efficiency (first output data) and m achine learning Model 2 to generate bystander editing activity (second output data), for at least one location in the nucleotide sequence associated with each guide RNA in the set; and i dentif y at least one G uide RNA via analysis of first and second output data . Hwang et al. describes web-based tools, BE-designer and BE-analyzer, which identify and assess base editor guide and target sequences. Hwang et al. teaches that BE-Designer provides a list of all possible sgRNA sequences from a given input DNA sequence (page 2, column 1 ) and analysis for CRISPR base editors based on SpCas9 from Streptococcus pyogenes, which recognizes a variety of different protospacer-adjacent motif (PAM) sequences (page 2, column 1). Hwang et al . further teaches generating at least two features from the input data including p ossible off target sequences , on target sequences, and their relative position and GC content. (page 3, figure 1) for optimal selection of sgRNAs ( page 6, column 1 ) . Claim 11 is directed to first output data (base editing efficiency) including a ratio of sequence reads with at least one base edit , amongst t he total number of sequence reads . Hwang et al. teaches results in a 9-column table where in , ‘Total Sequence’ indicates the number of all reads present in the Fastq file (page 4, column 2); the 8th column indicates the number of reads that have nucleotide conversions induced by CRISPR base editors in target windows (page 5, column 2); and u ltimately calculate s , the ratio of intended substitutions induced by CRISPR base editors (page 5, figure 3a ) . Claim 20 is directed to the second output data (bystander editing activity) including the frequency of occurrence of different base editing outcomes, including edits made to nucleotides at multiple positions , and claim 21 is directed to outputting a frequency distribution of multiple outcome combinations . Hwang et al. teaches generating g raphic plots that show the occurrence frequency ratio of types of nucleotide changes , such as C to T, C to G, and C to A , at each position (page 5, figure 3c ). Claim 22 is directed to generating a ratio of the frequency of occurrence of a specific combination of outcomes to the frequency of occurrence of all sequenced reads with at least one base edit in a desired window , near a protospacer sequence associated with a guide RNA. Hwang et al. teaches that t he 9th column indicates the intended substitution rate (such as ‘C to T Substitution Rate’), obtained by dividing the number of reads that have intended conversions in the base editing window with the number of reads above the minimum frequency (page 5, column 2). As, the minimum frequency ensures the ratio is compared to sequences with a t least one base edit, the result of this calculation serves as the bystander editing activity value . Hwang et al . does not teach inputting the features into machine learning models in order to identify at least one guide RNA from the output data. Doench et al. describes a sgRNA scoring protocol for a CRISPR- Cas9 editing system. Regarding claims 6, 7, and 12 , Doench et al. teaches inputting data into the following statistical models: ( i ) linear regression, (ii) L1 -regularized linear regression, (iii) L2 -regularized linear regression, (iv) the hybrid SVM plus logistic regression approach used previously, (v) Random Forest , (vi) Gradient-boosted regression tree, (vii) L1 logistic regression (a classifier), (viii) SVM Classification (page 194, column 1) in order to create optimized sgRNA libraries that maximize on-target activity and minimize off-target effects to enable more effective and efficient genetic screens and genome engineering (page 184, column 1). Claim s 8 and 19 are directed to first and second features, respectively, that encode, within the protospacer sequence associated with a guide RNA , at least some nucleotides in a protospacer sequence ; and at least some nucleotides, within a threshold number of nucleotides from the protospacer sequence. Doench et al. teaches one-hot encoding the five nucleotides immediately proximal to the PAM (page 192, column 2) within the protospacer sequence ; and t he two nucleotides in the N and N positions relative to the PAM “ NGGN ” (page 194, column 2) , a threshold number away from the protospacer sequence . Claim 36 is directed to determining a likelihood that the identified guide RNA and base editing system will introduce a desired change in a cell. Doench et al. teaches determining g ene-knockout efficacy scores (page 192, column 1) ; determining if a library pass es quality control when >85% of the sequencing reads map to an intended sgRNA, which corresponds to an oligonucleotide synthesis error rate of 0.75% per base or lower ( page 192, column 2); and analyzing the log2 fold change of each sgRNA relative to control cells treated with DMSO in order to effectively model on-target activity ( page 193, column 1 ) . Therefore, Hwang et al. in view of Doench et al. describe generating scoring values via machine learning models in order to identify guide RNAs , aligned with the claimed invention. Doench et al . further teaches that the on-target and off-target ranks of each sgRNA were combined at equal weight to provide a final rank for each sgRNA targeting a particular transcript (page 195, column 2) and selec ting up to 4 sgRNAs per gene using this heuristic (page 195, column 2) . However, Hwang et. al and Doench et al. do not explicitly teach using the product of the outputted bystander and efficiency metrics to select an optimal base guide RNA ( claim 23 ) , nor explicit motivation that the scoring algorithm, taught by Doench et al. can be applied to a base editing analysis system, taught by Hwang et al. Dandage et al. describes ‘ beditor ’, a computational workflow for designing libraries of guide RNAs for CRISPR-Mediated Base Editing. Dandage et al. teaches that the overall beditor score for a gRNA provides estimates of editing efficiency ( page 377, column 1 ), is determined by multiplying penalties assigned per alignment for all alignments with a penalty assigned to the gRNA ( page 378, column 2 ) , and is used to select the best-performing gRNAs from a designed gRNA library ( page 382, column 1 ). Dandage et al. further teaches that the relative values of such penalties were determined by fitting a third-degree polynomial equation to the mismatch tolerance data the from Doench et al . scoring system (page 378, column 2) . Therefore , Dandage et al. teaches that the scoring system taught by Doench et al. provides a framework that is applicable and integral to a predictive base editing algorithm. As such , it is obvious for one of ordinary skill in the art to similarly improve the base editing system of Hwang et al. with the techniques presented in Doench et al. and Dandage et al. in order to predictably improve the system with known techniques. Claim 10 is rejected under 35 U.S.C. 103 as being unpatentable over Hwang et al. ( IDS ref ; NPL; p. 135 , line 1; 2018) and Doench et al. ( IDS ref ; NPL; p. 113 ; line 8; 2014) in view Dandage et al. ( IDS ref ; NPL ; p. 110, line 13; 2018), as applied to claims 1, 6-8, 11-12, 19-23, 36, and 39-40 above, and in further view of Komor et al. ( IDS ref ; NPL; p. 144, line 1 ; 2017 ) . Hwang et al. , Doench et al ., and Dandage et al. teach a framework that identifies optimal base editing guide RNAs via machine learning scoring analysis . Claim 10 is directed to also including first input features that encode: at least some dinucleotides at neighboring positions in the protospacer sequence ; melting temperatures of the first guide RNA; a total number of one or more types of nucleotides; a percentage of one or more types of nucleotides; and an average base editing efficiency of the base editing system . Doench et al. further teaches encoding all adjacent pairwise nucleotides as features (page 164, column 2); t hermodynamic features from the melting temperatures of the guide sequence, (page 194, column 2); GC counts via count ing the number of Gs and Cs in the 20 mer (sgRNA) (page 194, column 2); and a Rule Set 2 score which averages on-target prediction ( log2 fold change) for sgRNAs within each gene (page 194, column 1). Hwang et al. further teaches that w ithin a given DNA sequence, BE-Designer finds all possible target sites based on input parameters; and in the base editing window, target nucleotides are highlighted, and their relative position and GC content , measured in a percentage of GC nucleotides (page 3, fig. 1) are indicated (page 2, column 2) . Therefore, Hwang et. al teaches a representing a percentage of G and C nucleotides in the sequence , that would be used as an input feature to a model for base editing efficiency prediction in view of Doench et al. and Dandage et al., as described above . Hwang et al. and Doench et al. in view of Dandage et al. do not explicitly t each generating an input feature representing an average base editing efficiency of the base editing system . Komor et al. describes BE4 , a next generation base editor. Komor et al. teaches calculating average efficiencies of base editing syst em s with different architecture components and arrangements , including napDNAbps and deaminases (page 6, fig. 5 ) . Komor et al. teaches that t he average efficiency of C-to-T editing for Target-AID at the same positions analyzed was 1.5 ± 0.5–fold lower than that of BE3 and 2.1 ± 0.5–fold lower than that of BE4 (page 6, column 1) . Komor et al. further teaches that a shifted editing window , caused by a different base editor (page 6, column 1), makes comparisons of efficiency and product purity between Target-AID and BE3 or BE4 difficult because a given target C could lie in more optimal or less optimal position within the different editing windows, even when using the same guide RNA (page 6 column 1) . Therefore, an average base editing efficiency of particular base editing system has a notable impact on effi ciency value of each guide RNA. As such , one of ordinary skill in the art has sufficient motivation to include this data as a n input feature in a base editing eff iciency prediction model , in order to ensure that the guide RNAs are evaluated and identified appropriately , within the specific base editing system they are designed for . Claim 13 is rejected under 35 U.S.C. 103 as being unpatentable over Hwang et al. ( IDS ref ; NPL; p. 135 , line 1; 2018) and Doench et al. ( IDS ref ; NPL; p. 113 , line 8; 2014) in view of Dandage et al. ( IDS ref ; NPL ; p. 110, line 13; 2018) , as applied to claims 1, 6-8, 11-12, 19-23, 36, and 39-40 previously , a nd in further view of Lin et al. (Bioinformatics; Vol. 34; p. 656-663; 2018) . Hwang et al. , Doench et al . , and Dandage et al. teach a framework that identify optimal base editing guide RNAs via machine learning scoring analysis . The previous references do not teach use of a deep neural network to obtain bystander editing data ( claim 13 ) . Lin et al. describes a deep learning approach to off-target prediction , within a CRISPR- Cas9 editing system . Lin et al. teaches that a deep convolutional neural networ k predict ing off-target s (page 661, column 1) , obtained the best performance on two datasets ( CRISPOR and GUIDE-seq ) , outperform ed the current state-of-art prediction m odel ( CFD score ), and surpassed functionality of L ogistic R egression , R andom F orest and G radient B oosting T rees traditional machine learning algorithms (page 662, column 1). Lin et al. further teaches that s uch intelligent prediction approaches can contribute to similar problems in a rigorous manner (page 662, column 2) . Therefore, Lin et. al provides sufficient teachings and motivation for one of ordinary skill in the to apply a deep convolutional neural network model to a bystander editing prediction of a base editing system, in order to improve the traditional machine learning model such as random forest and the CFD scoring system, as taught by Doench et al. Claims 14, 15, and 18 are rejected under 35 U.S.C. 103 as being unpatentable over Hwang et al. ( IDS ref ; NPL; p. 135 , line 1; 2018) and Doench et al. ( IDS ref ; NPL; p. 113 , line 8; 2014), in view of Dandage et al. ( IDS ref ; NPL ; p. 110, line 13; 2018), and Lin et al. (Bioinformatics; Vol. 34; p.656 -663; 2018) as applied to claim 13 above, and in further view of van den Oord et al ( arXiv:1606.05328 ; 2016) . Hwang et al. , Doench et al ., and Dandage et al., in view of Lin et al. teach a framework that identify optimal base editing guide RNAs via deep neural network scoring. Lin et al. further teaches that the adaption of CNN from computer vision to genetic sequence can be accomplished by considering each sgRNA-DNA sequence pair as an image , and i nstead of processing 2-dimensional image with colour channels, we consider a genomic sequence as a 4 L matrix (page 657, column 1). L in et al. does not teach a conditional autoregressive neural network (claim 14); an encoder that maps input to a latent representation , a decoder that maps the latent representation to output (claim 15) ; or parameters representing a position wise bias towards producing an unedited outcome (claim 18); van den Oord et al. describes conditional image generation with convolutional neural network decoders. Regarding claim 14 , van den Oord et al. teaches adapting and improving a convolutional variant of the PixelRNN architecture (page 1, column 1) by m aking it straightforward to apply in domains such as compression and probabilistic planning us ing autoregressive connections architecture to model images and decompo se the joint image distribution as a product of conditionals (page 1, column 1) . Regarding claim 1 5 , van den Oord further teaches a n auto-encoder that consists of two parts: an encoder that takes an input image x and maps it to a low-dimensional representation h, a decoder that tries to reconstruct the original image (page 4, column 1) , and an overall model that can be conditioned on latent embeddings (page 1, column 1) . Regarding claim 18 , van den Oord teaches mapping h , a one-hot encoding, to a spatial representation with a deconvolutional neural network, to obtain a location dependent bias (page 5, column 1) . Lin et al. teaches that though the editing system is sensitive to the number, positions and distribution of mismatches (page 656, column 2), most of the existing methods do not consider the potential relationships between mismatched and matched sites, which affect the off-target activity (page 657, column 1). Doench et al. teaches that several typical scoring systems consider the position and number, but not identity, of mismatches between the RNA and DNA (page 189, column 2) . Therefore, van den Oord et al. provides sufficient motivation for one of ordinary skill in the art to adopt a conditional autoregressive neural network with an autoencoder , in order to effectively model a genetic sequence matrix and consider the relationships of several mismatch properties, taught as pertinent considerations to machine learning off-target prediction by Lin et al. and Doench et al. As such, the model results in a predictable overall improvement to a convolutional neural network. Claims 34, 35, and 9 9 are rejected under 35 U.S.C. 103 as being unpatentable over Hwang et al. ( IDS ref ; NPL; p. 135 , line 1; 2018) and Doench et al. ( IDS ref ; NPL; p. 113 , line 8; 2014) in view of Dandage et al. ( IDS ref ; NPL ; p. 110, line 13; 2018), as applied to claims 1, 6-8, 11-12, 19-23, 36, and 39-40 previously, and in further view of Schatoff et al. ( Methods; p. 164-165; PMCID : PMC6684841 ; 2019 ). Claim 34 is directed to synthesizing an identified guide RNA; Claim 35 is directed to using the identified guide RN and base editing system ; and Claim 99 is directed to editing a target DNA sequence via c ontacting its genome with a guide RNA selected from the method of claim 1. Schatoff et al. teaches key technical aspects of designing and executing BE experiments and provides detailed experimental examples of successful base editing in cell lines and organoids to help guide the effective use of these tools for genome modifications (page 1, column 1) . Regarding claims 34, 35, and 99, Schatoff et al. teaches using chemically stabilized synthetic sgRNAs from Synthego to induce target genome base editing in intestinal organoids (page 5, column 1) that can be delivered by in vitro transcription or plasmid-based techniques (page 5, column 1) . Schatoff et al. further teaches that b ase editing carries the possibility of off-target effects (page 3, column 1) that c a n be reduced or eliminated through careful sgRNA design (page 3, column 1 ) . Therefore, Schatoff et al. acknowledges that a method of carefully design ing and select ing guide RNAs , based on the consideration of off-target effects , should be applied to a base editing system that contacts a synthesized guide RNA with its target sequence. One of ordinary skill in the art would be sufficiently motived to combin e the methods in order to yiel d the predictable result of one or more base edit s . Conclusion No claims are currently allowed. Correspondence Any inquiry concerning this communication or earlier communications from the examiner should be directed to Milana Thompson whose telephone number is FILLIN "Phone number" \* MERGEFORMAT (571)272-8740 . The examiner can normally be reached FILLIN "Work Schedule?" \* MERGEFORMAT Monday - Friday, 9:00-6:00 ET . Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice. If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, FILLIN "SPE Name?" \* MERGEFORMAT Karlheinz Skowronek can be reached at FILLIN "SPE Phone?" \* MERGEFORMAT (571) 272-1113 . The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300. Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center ( EBC ) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000. / M.K.T ./ Examiner, Art Unit 1687 /Karlheinz R. Skowronek/ Supervisory Patent Examiner, Art Unit 1687