DETAILED ACTION
Applicant's response, filed 2/2/2026, has been fully considered. The following rejections and/or objections are either reiterated or newly applied. They constitute the complete set presently being applied to the instant application.
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 .
Continued Examination Under 37 CFR 1.114
A request for continued examination under 37 CFR 1.114, including the fee set forth in 37 CFR 1.17(e), was filed in this application after final rejection. Since this application is eligible for continued examination under 37 CFR 1.114, and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114. Applicant's submission filed on 2/2/2026 has been entered.
Priority
Acknowledgment is made of applicant’s claim for priority. Application claims benefit of U.S. Provisional Application No. 63/070,993 filed on 8/27/2020. As such, the effective filing date of claims 1-2, 4, 9, 22-23, 26-27, 29, 34, 47-48, 51-52, 54, 59, 72-73, and 76-77 is 8/27/2020.
Claim Status
Claims 1, 2, 4, 9, 22, 23, 26, 27, 29, 34, 47, 48, 51,52, 54, 59, 72, 73, 76, and 77 are pending.
Claims 76 and 77 are newly added.
Claims 1, 2, 4, 9, 22, 23, 26, 27, 29, 34, 47, 48, 51,52, 54, 59, 72, 73, 76, and 77 are rejected.
Specification
Response to Amendment
In view of applicant’s amendments to the specification, previous objections to the specification are withdrawn.
Claim Objections
Response to Amendment
In view of applicant’s amendments to the claims, previous objections to the claims are withdrawn.
Claim Rejections - 35 USC § 112
Response to Amendment
In view of applicant’s amendments to the claims 112(b) claim interpretations have been withdrawn.
Claim Rejections - 35 USC § 101
Response to Amendment
In view of applicant’s amendments to the claims previous rejections under 35 U.S.C. 101 were reviewed, updated, and provided below.
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, 2, 4, 9, 22, 23, 26, 27, 29, 34, 47, 48, 51,52, 54, 59, 72, 73, 76, and 77 are rejected under 35 U.S.C. 101 because the claimed invention is directed to abstract ideas without significantly more. The claims recite a method of aligning single cell data with spatial data to create spatial maps of cell types and gene expression at a single cell resolution. The judicial exception is not integrated into a practical application because while the claims attempt to integrate the exception into a practical application, said application are generically recited computer elements that do not add meaningful limitations to the abstract idea as it is simply implementing the abstract idea on a computer. The claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception because the computer elements only store and retrieve information in memory as well as perform basic calculations that are known to be well-understood, routine and conventional computer functions as recognized by the decisions listed in MPEP § 2106.05(d).
Framework with which to Analyze Subject Matter Eligibility:
Step 1: Are the claims directed to a category of statutory subject matter (a process, machine, manufacture, or composition of matter)? [See MPEP § 2106.03]
Claims are directed to statutory subject matter, specifically a method (claims 1-2, 4, 9, 22-23), a system (Claims 26-27, 29, 34, 47-48), and a computer program product (Claims 51-52, 54, 59, 72-73)
Step 2A Prong One: Do the claims recite a judicially recognized exception, i.e., an abstract idea, a law of nature, or a natural phenomenon? [See MPEP § 2106.04(a)]
The claims herein recite abstract ideas, specifically mental processes and mathematical concepts.
With respect to the Step 2A Prong One evaluation, the instant claims are found herein to recite abstract ideas that fall into the grouping of mathematical concepts.
Claims 1, 26, 51: Generating a mapping matrix, and generating a spatial map are processes of comparing/contrasting, and calculating that can be done by hand or in the human mind and are therefore abstract ideas, specifically mental processes. The shared measured features being one of those specified is merely further limiting the data itself which is an abstract idea, specifically a mental process.
Claims 2, 27, 52: The unsupervised deep learning nonlinear optimization program the cell density distribution and measured features using those specified is a verbal articulation of a mathematical process and therefore an abstract idea, specifically a mathematical concept.
Claims 4, 29, 54: Validating the spatial map, applying a first learned spatial map to a second, relating the spatial map to histological and anatomical data, and registering the spatial map on an anatomically annotated common coordinate framework are processes of comparing/contrasting, and calculating that can be done by hand or in the human mind and are therefore abstract ideas, specifically mental processes.
Claims 9, 34, 59: The single cell data comprising expression of more than 10000 genes and expression data comprising more than 100 genes, and the single cell data being multi-modal data are directed to the data itself which is an abstract idea, specifically a mental process.
Claim 23, 48, 73: Determining a quantity of cells in the spatial and single cell data, generating a single cell matrix, generating a mapping matrix, applying a mapping filter, assigning a filter value based on the probability, generating a filter vector, and filtering the single cell matrix and mapping matrix are processes of counting, comparing/contrasting, removing, and calculating that can be done by hand or in the human mind and are therefore abstract ideas, specifically mental processes.
Claim 76: Segmenting cells in the histological data is a process of identifying that can be done by hand or within the human mind and is therefore an abstract idea, specifically a mental process. Deconvoluting the spatial data using deterministic mapping is a verbal articulation of a mathematical process and is therefore an abstract idea, specifically a mathematical concept.
Claim 77: Determining a sectioning depth, and selectively mapping the single cell data are processes of identifying, selecting and calculating that can be done via pen and paper or within the human mind and are therefore abstract ideas, specifically mental processes.
Step 2A Prong Two: If the claims recite a judicial exception under prong one, then is the judicial exception integrated into a practical application? [see MPEP § 2106.04(d)]
Because the claims do recite judicial exceptions, direction under Step 2A Prong Two provides that the claims must be examined further to determine whether they integrate the abstract ideas into a practical application.
The following claims recite the following additional elements in the form of non-abstract elements:
Claims 1, 26, 51: Receiving single cell data and spatial data are insignificant extra solution activities, specifically mere data gathering (See Performing clinical tests on individuals to obtain input for an equation, In re Grams, 888 F.2d 835, 839-40; 12 USPQ2d 1824, 1827-28 (Fed. Cir. 1989) and Determining the level of a biomarker in blood, Mayo, 566 U.S. at 79, 101 USPQ2d at 1968. See also PerkinElmer, Inc. v. Intema Ltd., 496 Fed. App'x 65, 73, 105 USPQ2d 1960, 1966 (Fed. Cir. 2012) (assessing or measuring data derived from an ultrasound scan, to be used in a diagnosis)) [See MPEP § 2106.05(g)]. A system, storage device, processor, non-transitory computer readable medium, and a computer program product are all generic and nonspecific elements of a computer that do not improve the functioning of any computer or technology described herein [See MPEP § 2106.04(d)(1) and MPEP § 2106.05(d)].
Claim 22, 47, 72: Receiving an input of one or more types of genes is an insignificant extra solution activity, specifically mere data gathering (See Performing clinical tests on individuals to obtain input for an equation, In re Grams, 888 F.2d 835, 839-40; 12 USPQ2d 1824, 1827-28 (Fed. Cir. 1989) and Determining the level of a biomarker in blood, Mayo, 566 U.S. at 79, 101 USPQ2d at 1968. See also PerkinElmer, Inc. v. Intema Ltd., 496 Fed. App'x 65, 73, 105 USPQ2d 1960, 1966 (Fed. Cir. 2012) (assessing or measuring data derived from an ultrasound scan, to be used in a diagnosis)) [See MPEP § 2106.05(g)].
Claim 76: Receiving histological data is an insignificant extra solution activity, specifically mere data gathering (See Performing clinical tests on individuals to obtain input for an equation, In re Grams, 888 F.2d 835, 839-40; 12 USPQ2d 1824, 1827-28 (Fed. Cir. 1989) and Determining the level of a biomarker in blood, Mayo, 566 U.S. at 79, 101 USPQ2d at 1968. See also PerkinElmer, Inc. v. Intema Ltd., 496 Fed. App'x 65, 73, 105 USPQ2d 1960, 1966 (Fed. Cir. 2012) (assessing or measuring data derived from an ultrasound scan, to be used in a diagnosis)) [See MPEP § 2106.05(g)].
Step 2B: if the claims do not integrate the judicial exception, do the claims provide an inventive concept? [see MPEP § 2106.05]
Because the additional claim elements do not integrate the abstract idea into a practical application, the claims are further examined under Step 2B, which evaluates whether the additional elements, individually and in combination, amount to significantly more than the judicial exception itself by providing an inventive concept.
The claims do not recite additional elements that are sufficient to amount to significantly more than the judicial exceptions because the claims recite additional elements that are generic, convention, nonspecific or mere data gathering. These additional elements include:
The additional elements of receiving an input of one or more types of genes, receiving histological data, and receiving single cell data and spatial data, are insignificant extra solution activities, specifically mere data gathering that have been determined to be conventional (Receiving or transmitting data over a network, e.g., using the Internet to gather data, Symantec, 838 F.3d at 1321, 120 USPQ2d at 1362 (utilizing an intermediary computer to forward information), Performing repetitive calculations, Flook, 437 U.S. at 594, 198 USPQ2d at 199 (recomputing or readjusting alarm limit values), and Storing and retrieving information in memory, Versata Dev. Group, Inc. v. SAP Am., Inc., 793 F.3d 1306, 1334, 115 USPQ2d 1681, 1701 (Fed. Cir. 2015)) [see MPEP § 2106.5(g)]. Therefore, these additional elements, both taken individually and as a whole do not amount to significantly more than the judicial exception by providing an inventive concept.
The additional elements of a system, storage device, processor, non-transitory computer readable medium, and a computer program product (Receiving or transmitting data over a network, e.g., using the Internet to gather data, Symantec, 838 F.3d at 1321, 120 USPQ2d at 1362 (utilizing an intermediary computer to forward information), Performing repetitive calculations, Flook, 437 U.S. at 594, 198 USPQ2d at 199 (recomputing or readjusting alarm limit values), and Storing and retrieving information in memory, Versata Dev. Group, Inc. v. SAP Am., Inc., 793 F.3d 1306, 1334, 115 USPQ2d 1681, 1701 (Fed. Cir. 2015)) [See MPEP § 2106.05(d)(II)]. Therefore, these additional elements, both taken individually and as a whole do not amount to significantly more than the judicial exception by providing an inventive concept.
Therefore, claims 1, 2, 4, 9, 22, 23, 26, 27, 29, 34, 47, 48, 51,52, 54, 59, 72, 73, 76, and 77, when the limitations are considered individually and as a whole, are rejected under 35 U.S.C. § 101 as being directed to non-statutory subject matter.
Response to Arguments
Applicant's arguments filed 2/2/2026 have been fully considered but they are not persuasive.
Regarding Step 2A Prong One:
Applicant asserts on page 17 of the Remarks filed 2/2/2026 that the specified features of claim 1 are impossible to perform within the human mind by way of example “the claim requires the manipulation of computer-based biological data structures and the output of modified data structures”, and makes comparison to Research Corp. Techs. v. Microsoft Corp., 627 F.3d 859, 868 (Fed. Cir. 2010). However, Examiner reminds applicant that the computer-based biological structures and methods may be performed within a computer do not inherently render either the processes nor the data input/output beyond the capability of the human mind to understand. Furthermore, the case cited was found to not be a mental process as the halftoned digital image was not a mental process that humans are capable, as compared to the count matrix of gene expression data or spatial data, which are merely tables, and a mapping matrix which is matrix that represents a linear transformation, all of which are easily understandable and performable within the human mind.
Applicant asserts on page 18 of the Remarks filed 2/2/2026 that the data is not interpretable by a human mind and cites the mapping of 58,000 snRNA-seq profiles to spatial voxels. However, examiner reminds applicant that this is an argument of scale, not an argument of ability, and this is not commensurate in scope with the claims as written.
Applicant asserts on page 18 of the Remarks filed 2/2/2026 that the data itself is not abstract and cites that the data is representative of measured features. However, examiner reminds applicant that the data itself may be representative of measured features, but the limitation of that data and what it represents through the construction of a reference frame in experimentation is an abstract idea. This additionally back up by the Electric Power Group Decision ELECTRIC POWER GROUP, LLC v. ALSTOM S.A. wherein on page 7 is said - Accordingly, we have treated collecting information, including when limited to particular content (which does not change its character as information), as within the realm of abstract ideas.
Applicant asserts on page 18 of the Remarks filed 2/2/2026 that the alignment of the single cell data to the spatial data using a nonlinear optimization program is not an abstract concept but is rather a specific technical process. Furthermore, applicant asserts similarity to example 39 in terms of the training of a neural network. However, examiner reminds applicant that while it may be a technical process, that does not preclude it from being an abstract concept as alignment is merely a matching of two sequences and the addition of a nonlinear optimization merely further integrates a verbal articulation of a mathematical concept as that is a specific mathematical function, neither of which are additional elements. In regards to similarities with example 39, said example is directed to the neural network, and additional element, here the claim is directed to the process of alignment and how that is accomplished mathematically.
Step 2A Prong Two:
Applicant asserts on page 20 that the amended claims integrate the alleged judicial exception into a practical application. Specifically, applicant asserts that amended claim 1 recites an improvement to technology through the generation of spatial maps at a higher resolution, single-cell resolution, and a new approach to cellular mapping that was not previously possible. However, examiner reminds applicant that according to MPEP 2106.05(a) “It is important to note, the judicial exception alone cannot provide the improvement. The improvement can be provided by one or more additional elements”, and while applicant suggest improvements, said improvements are directed to the improvement of the judicial exception not the additional elements, i.e. the computer data storage itself or processing are not improved, only the process by which the alignment or determination, etc. are being performed are improved. Examples applicant gives are those of the nonlinear optimization that is implemented, this however is directed to a mathematical concept, not an additional element.
Furthermore, applicant asserts a practical application of the claimed invention in the ability to expand from a measured subset of genes to genome wide profiles spatial measurements, mapping the location of cell types, deconvolving low resolution measurements and resolving spatial patterns. However, examiner reminds applicant that these again are directed to abstract concepts not to improvements or practical applications of the additional elements.
Applicant asserts on page 21 of the Remarks filed 2/2/2026 that the claim recites a particular solution to the problem of generating spatial genomics data at single cell resolution. However, examiner reminds applicant that the examples provided “generating a mapping matrix”, “generating the spatial map”, etc. are directed to abstract ideas, not additional elements, and therefore, cannot be particular solutions or improvements.
Step 2B:
Applicant asserts on page 22 that the specific combination of technical features is significantly more than the judicia exception. However, examiner reminds applicant that MPEP 2106.05(b) state “When determining whether a claim integrates a judicial exception, into a practical application in Step 2A Prong Two and whether a claim recites significantly more than a judicial exception in Step 2B, examiners should consider whether the judicial exception is applied with, or by use of, a particular machine” and “Whether its involvement is extra-solution activity or a field-of-use, i.e., the extent to which (or how) the machine or apparatus imposes meaningful limits on the claim. Use of a machine that contributes only nominally or insignificantly to the execution of the claimed method (e.g., in a data gathering step or in a field-of-use limitation) would not integrate a judicial exception or provide significantly more”, with the current claim reciting mere additional elements that are nothing more than mere data gathering and conventional within the art, which can therefore not be significantly more than the judicial exception.
Claim Rejections - 35 USC § 103
Response to Amendment
In view of applicant’s amendments to the claims previous rejections under 35 U.S.C. 103 were reviewed, updated, and provided below.
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.
Claims 1, 9, 22-23, 26, 34, 47-48, 51, 59 and 72-73 are rejected under 35 U.S.C. 103 as being unpatentable over Satija et al. (Nature Biotechnology (2019) 495-502; previously cited), Tian et al. (Nature Machine Intelligence (2019) 191-198; previously cited), and Mahfouz et al. (Methods (2015) 79-89; previously cited).
Claim 1 is directed to a method of aligning single cell data with spatial data to generate spatial maps of cell types and gene expression.
Claim 26 is directed to a system for aligning single cell data with spatial data to generate spatial maps of cell types and gene expression.
Claim 51 is directed to a CRM for aligning single cell data with spatial data to generate spatial maps of cell types and gene expression.
Satija et al. teaches on page 495, column 2, paragraph 2 “We use single-cell RNA-seq to identify thousands of RNAs expressed in each cell and infer its spatial origin computationally”, further on in the same paragraph “Seurat uses a statistical framework to combine cells’ gene expression profiles, as measured by single-cell RNA-seq, with complementary in situ hybridization data for a smaller set of ‘landmark’ genes that guide spatial assignment” and in Figure 1 providing an overview of the method, reading on receiving, by one or more computing devices, single cell data obtained from a first specimen from a specific anatomical region or tissue type, receiving, by one or more computing devices, spatial data obtained from the first specimen or a second specimen from the anatomical region or tissue type, wherein the single cell data and spatial data comprises one or more shared measured features including at least one of gene expression data, chromatin accessibility data, or an epigenetic mark and generating a spatial map wherein the single cell data constitutes the new spatial data.
Satija et al. does not teach aligning the single cell data to the spatial data based on one or more of the shared features using unsupervised deep learning nonlinear optimization.
Tian et al. teaches on page 191 in the abstract “we have developed scDeepCluster, a single-cell model-based deep embedded clustering method, which simultaneously learns feature representation and clustering via explicit modelling of scRNA-seq data generation”, and on page 192, column 1, paragraph 3 “DNNs are a natural choice to parameterize a nonlinear transforming function that maps the original high-dimensional data to a small latent space”, which in view of the teachings in Satija et al., regarding a method for aligning scRNA seq data with spatial data to produce spatial maps, reads on aligning, by the one or more computing devices, the single cell data to the spatial data based on the one or more shared features using an unsupervised deep learning nonlinear optimization program that rewards spatial alignment of the single cell data to the spatial data based on cell density distributions, gene expression or both, and generating, by the one or more computing devices, a spatial map wherein the single cell data constitutes new spatial data such that the spatial map provides likely spatial locations of the cells within the specific anatomical region or tissue type.
Mahfouz et al. teaches on page 80, column 2, paragraph 3 “For genes with more than one sagittal experiment, the maximum correlation value was used. A mask was applied to exclude all non-brain voxels, resulting in a 61,164 x 3241 (voxels x genes) matrix for the coronal experiments and a 27,365 x 3241 matrix for the sagittal experiments”, reading on generating a mapping matrix that maps cells from the single cell data to voxels of the spatial data by aligning.
It would have been obvious at the time of the effective filing date to modify the teachings of Satija et al. for combining scRNA-Seq expression data with in situ hybridization data to generate spatial maps of probabilities, with those of Tian et al. for the use of deep learning models within scRNA-Seq clustering, as the both are within the same field and seeking to maximize the use of scRNA-Seq data and both clustering and spatial mapping are similar in that they are attempting to group cells based upon expression. Furthermore, it would have been obvious to modify the teachings of both Satija et al. and Tian et al. with the teachings of Mahfouz et al. for the use of a mapping matrix for visualizing spatial gene expression as within the abstract the latter details “Based on our observations, we conclude that BH-SNE maps with or without prior dimensionality reduction (based on PCA) provide comprehensive and intuitive insights in both the local and global spatial transcriptome structure of the human and mouse Allen Brain Atlases”. One would have had a reasonable expectation of success given that in the abstract Tian et al. teaches “DNNs are a natural choice to parameterize a nonlinear transforming function that maps the original high-dimensional data to a small latent space”, which is what the invention is directed to. Additionally, one would have had a reasonable expectation of success for combining with Mahfouz et al. given that the method is specifically directed to expression visualization directed to mapping expression (the genes) directly to the voxels (Voxel x gene matrix). Therefore, it would be obvious to one with ordinary skill in the art to incorporate the teachings of each and to be successful.
Claim 9 is directed to the method of claim 1 but further specifies that the shared features comprise gene expression, accessible chromatin, an epigenetic mark, a combination thereof or one or more of a specific list of data derived features.
Claim 34 is directed to the system of claim 26 but further specifies that the shared features comprise gene expression, accessible chromatin, an epigenetic mark, a combination thereof or one or more of a specific list of data derived features.
Claim 59 is directed to the CRM of claim 51 but further specifies that the shared features comprise gene expression, accessible chromatin, an epigenetic mark, a combination thereof or one or more of a specific list of data derived features.
Satija et al. teaches in the Figure 1 text “Seurat uses expression measurements across many correlated genes to ameliorate stochastic noise in individual measurements for landmark genes. As schematized, Seurat learns a model of gene expression for each of the landmark genes based on other variable genes in the data set, reducing the reliance on a single measurement, and mitigating the effect of technical errors”, reading on wherein the shared features comprise gene expression, accessible chromatin, an epigenetic mark, and/or a combination thereof and wherein the single cell data is multi-modal single cell data.
Tian et al. teaches on page 193 in Table 1 the number of genes for each dataset, with scDeepCluster not placing a limitation on the number of genes used and in their example in the paper use up to 24000 genes, but there is no upper/lower limit given so using 100 genes would be within the scope, therefore reading on wherein the single cell data comprises the expression of greater than 10,000 genes and the spatial data comprises spatial expression of less than 100 genes.
Claim 22 is directed to the method of claim 1 but further specifies the receiving of an input of one or more types of genes.
Claim 47 is directed to the system of claim 26 but further specifies the receiving of an input of one or more types of genes.
Claim 72 is directed to the CRM of claim 51 but further specifies the receiving of an input of one or more types of genes.
Satija et al. teaches in the Figure 1 text “Seurat takes single-cell RNA-seq data from dissociated cells, where information about the original spatial context was lost during dissociation”, which reads on comprising receiving an input of one or more types of genes.
Claim 23 is directed to the method of claim 1 but further specifies that both a single cell matrix and a mapping matrix be generated.
Claim 48 is directed to the system of claim 26 but further specifies that both a single cell matrix and a mapping matrix be generated.
Claim 73 is directed to the CRM of claim 51 but further specifies that both a single cell matrix and a mapping matrix be generated.
Satija et al. teaches on page 506, column 1, paragraph 7 “We then constructed a separate matrix of ‘imputed’ measurements for each of our landmark genes across all single cells”, a single cell matrix, and on the same page but in column 2, paragraph 5 “In order to estimate a valid covariance matrix for n landmark genes for an individual bin, we needed data from at least n cells”, a mapping matrix, and on page 495, column 2, paragraph 2 “Seurat uses a statistical framework to combine cells’ gene expression profiles, as measured by single-cell RNA-seq, with complementary in situ hybridization data for a smaller set of ‘landmark’ genes that guide spatial assignment”, which would inherently require knowing the number cells in the single cell data and spatial data. Additionally Figure 1 and paragraphs 1 and 2 of column 1 on page 496 detail the binning of cells, filtering, etc. “(i) the expression profilesof individual dissociated cells and (ii) a spatial reference map of gene expression for a small number of landmark genes. This requires the subdivision of the tissue of interest into discrete spatial domains (hereafter, ‘bins’) of user-defined geometry and size. For the map, landmark genes are defined as either ‘on’ or ‘off’ in each bin, for example, as determined from published in situ stainings. Seurat then uses the single-cell expression levels of the landmark genes to determine in which bins a cell likely originated. Seurat consists of the following steps. (i) It uses co-expression patterns across cells in the single-cell RNA-seq profiles to impute the expression of each landmark gene in each cell…(ii) It relates the continuous imputed RNA-seq expression levels of each landmark gene to the binary spatial expression values using a mixture model constrained by the proportion of cells expressing the gene in the reference map. (iii) For each bin, it constructs a multivariate normal model for the joint expression of the landmark genes based on these mixture models, the binary spatial reference map and an optional quantitative refinement step that estimates covariance parameters between all pairs of genes. (iv) Given these models, it infers the spatial origin of each profiled cell by calculating a posterior probability for each cell-bin pair, allowing determination of the cell’s likely position(s) and confidence in the mapping”, of which binning and generating proportions would inherently require determining the quantity of cells, constraining by proportion of cells in the reference map is a mapping filter, and calculating a posterior is determining a probability and applying a sigmoid function. This reads on generating a single cell matrix and a mapping matrix, wherein the method comprises:- determining a quantity of cells in the single cell data,- determining a quantity of cells in the spatial data; and- applying a mapping filter if the quantity of cells in the single cell data is greater than the quantity of cells in the spatial data, wherein applying the mapping filter comprises:- determining a probability of finding each cell of the single cell data in a voxel of the spatial data - assigning a filter value to each cell of the single cell data based on the determined probability;- generating a filter vector;- applying a sigmoid function to the filter vector; and- filtering the single cell matrix and the mapping matrix.
Mahfouz et al. teaches on page 80, column 2, paragraphs 2-3 “We retrieved all expression energy volumes from using the Allen Brain Atlas application programming interface (API). Expression energy is a measurement combining the expression level (the integrated amount of signal within each voxel) and the expression density (the amount of expressing cells within each voxel)…For each gene, we computed the Spearman’s rank correlation between the corresponding coronal and sagittal experiments and selected genes in the top-three quartiles of correlation (3241 genes)”, of which the use of single cell RNA-seq data would inherently create a single cell matrix of expression and genes, the voxel x gene matrix is a mapping matrix based on quantity of cells in the expression data and the voxels of spatial data, and the mapping matrix by use of expression density incorporates a probability of finding cell expression data within said voxel, reading on generating a single cell matrix based on the quantity of cells in the single cell data and a number of genes; generating the mapping matrix based on the quantity of cells in the single cell data and the voxels of the spatial data, wherein each element within the mapping matrix represents a probability of finding each cell of the single cell data in each voxel of the spatial data.
Claim 47 is directed to the system of claim 26 but further specifies the receiving of an input of one or more types of genes.
Claim 72 is directed to the CRM of claim 51 but further specifies the receiving of an input of one or more types of genes.
Satija et al. teaches in the Figure 1 text “Seurat takes single-cell RNA-seq data from dissociated cells, where information about the original spatial context was lost during dissociation”, which reads on comprising receiving an input of one or more types of genes.
Claims 2, 27, and 52 are rejected under 35 U.S.C. 103 as being unpatentable over Satija et al. (Nature Biotechnology (2019) 495-502; previously cited), Tian et al. (Nature Machine Intelligence (2019) 191-198; previously cited), and Mahfouz et al. (Methods (2015) 79-89; previously cited) as applied to claims 1, 9, 22-23, 26, 34, 47-48, 51, 59 and 72-73 above, and further in view of Cheng et al. (Nucleic Acids Research (2019) 1-14; newly cited).
Claim 2 is directed to the method of claim 1 but further specifies that the nonlinear optimization function uses one or more similarity functions the one or more similarity functions comprise the Kullback-Leibler divergence and the cosine similarity.
Claim 27 is directed to the System of claim 26 but further specifies that the nonlinear optimization function uses one or more similarity functions the one or more similarity functions comprise the Kullback-Leibler divergence and the cosine similarity.
Claim 52 is directed to the CRM of claim 51 but further specifies that the nonlinear optimization function uses one or more similarity functions the one or more similarity functions comprise the Kullback-Leibler divergence and the cosine similarity.
Satija et al., Tian et al., and Mahfouz et al. teach the method, system and CRM of claims 1, 26, and 51 as previously described.
Tian et al. teaches on page 192, column 1, paragraph 4 “In the proposed framework, the nonlinear function mapping the read count matrix of scRNA-seq data to a low-dimensional latent representation is learned by the ZINB model-based autoencoder, while the clustering task on latent space is performed by clustering with Kullback–Leibler (KL) divergence, as described in the ‘deep embedded clustering’ (DEC) algorithm”.
Cheng et al. teaches on page 3, column 1, paragraph 3 “We classify unknown cells X’ efficiently by projecting them into space spanned by the informative LC states learned from a representative sample: where Al is calculated from the original A by removing those rows and columns that are not associated with the informative LC states, and Gl is calculated from G by removing those columns that are not associated with the informative LC states. We can then calculate the cosine similarity between S’ and an ‘average’ cell from each cluster and find the cluster with the maximum similarity for each unknown cell”, which in view of Tian et al. reads on wherein the unsupervised deep learning nonlinear optimization program assesses the cell density distribution using Kullback-Leibler (KL) divergence and assesses the one or more shared measured features using cosine similarity.
It would have been obvious at the time of first filing to have modified the teaching of Satija et al. and Tian et al. for the method, system and CRM of claims 1, 26, and 51, with the teachings of Cheng et al. for the use of a cosine similarity in the assessment of shared features given that the latter describes in the abstract “We show that LCA is robust, accurate, and powerful by comparison with multiple state-of-the-art computational methods when applied to large-scale real and simulated scRNA-seq data. Importantly, the ability of LCA to learn from representative subsets of the data provides scalability, thereby addressing a significant challenge posed by growing sample sizes in scRNAseq data analysis”. One would have had a reasonable expectation of success given that the method is directed to assessment of shared features in RNA-seq, expression, data and would merely be exchanging one known method for another. Therefore, it would have been obvious at the time of first filing to have modified the teachings of each and to be successful.
Claims 4, 29, and 54 are rejected under 35 U.S.C. 103 as being unpatentable over Satija et al. (Nature Biotechnology (2019) 495-502; previously cited), Tian et al. (Nature Machine Intelligence (2019) 191-198; previously cited), and Mahfouz et al. (Methods (2015) 79-89; previously cited) as applied to claims 1, 9, 22-23, 26, 34, 47-48, 51, 59 and 72-73 above, and further in view of Allen Institute (2005; previously cited).
Claim 4 is directed to the method of claim 1 but further specifies validating the spatial map, applying a first spatial map to a second to increase the resolution, relating the map to histological and anatomical data, and registering the map on an anatomically annotated common coordinate framework.
Claim 29 is directed to the system of claim 26 but further specifies validating the spatial map, applying a first spatial map to a second to increase the resolution, relating the map to histological and anatomical data, and registering the map on an anatomically annotated common coordinate framework.
Claim 54 is directed to the CRM of claim 51 but further specifies validating the spatial map, applying a first spatial map to a second to increase the resolution, relating the map to histological and anatomical data, and registering the map on an anatomically annotated common coordinate framework.
Satija et al., Tian et al., and Mahfouz et al. teach the method, system and CRM of claims 1, 26, and 51 as previously described.
Satija et al. further teaches in the Figure 1 “Build expression models for the landmark genes in each region of the tissue using the binary expression maps and bimodal mixture models of RNA-seq data”, in the Figure 1 text “As schematized, Seurat learns a model of gene expression for each of the landmark genes based on other variable genes in the data set, reducing the reliance on a single measurement, and mitigating the effect of technical errors. Seurat then builds statistical models of gene expression in each bin by relating the bimodal expression patterns of the RNA-seq estimates to the binarized in situ data. Shown are probability distributions for genes X, Y and Z for three different embryonic bins. Finally, Seurat uses these models to infer the cell’s original spatial location, assigning posterior probability of origin (depicted in shades of purple) to each bin”, on page 499, column 2, paragraph 1 “Seurat’s spatial inferences can be combined with unsupervised analysis of single-cell RNA-seq data to define and characterize both known and previously unidentified, rare subpopulations of cells within complex tissues”, reading on validating the spatial map by predicting the expression of one or more holdout genes, applying a first learned spatial map to a second spatial map to increase the resolution of the second map, and relating the spatial map to histological and anatomical data for the first specimen or the second specimen.
Satija et al., Tian et al., and Mahfouz et al. do not teach registering the spatial map on an anatomically annotated common coordinate framework.
Allen Institute teaches on page 1 paragraph 2 “An essential tool to understand the structure and function of the mouse brain at molecular, cellular, system and behavioral levels, it has been successfully used for large-scale data mapping, quantification, presentation, and analysis and has evolved through the creation of multiple versions. The first version (in 2005) of the CCF (col. 1) was created to support the product goals of the Allen Mouse Brain Atlas”, reading on registering the spatial map on an anatomically annotated common coordinate framework.
It would have been obvious at the time of the effective filing date to modify the teachings of Satija et al. and Tian et al. for the method of claims 1, 26, and 51 with the teachings of the Allen institute for the first CCF for storing expression data relating to mouse brain tissue expression maps, as one is merely developing the maps while the other is aggregating them into an atlas, thus the latter would not need to rely upon the existing maps but could effectively generate their own. One would have had a reasonable expectation of success given that the Allen Institute has been collecting and aggregating these expression maps into atlases for decades with increasing precision. Therefore, it would be obvious to one with ordinary skill in the art to incorporate the teachings of each and to be successful.
Claims 76 and 77 are rejected under 35 U.S.C. 103 as being unpatentable over Satija et al. (Nature Biotechnology (2019) 495-502; previously cited), Tian et al. (Nature Machine Intelligence (2019) 191-198; previously cited), and Mahfouz et al. (Methods (2015) 79-89; previously cited) as applied to claims 1, 9, 22-23, 26, 34, 47-48, 51, 59 and 72-73 above, and further in view of Mignardi et al. (Proceedings of the IEEE (2016) 530-541; newly cited).
Claim 76 is directed to the method of claim 1 but further specifies using histological data to segment and estimate the number of cells in each voxel and deconvolute the spatial data.
Satija et al., Tian et al., and Mahfouz et al. teach the method, system and CRM of claims 1, 26, and 51 as previously described.
Mignardi et al. teaches in the abstract “Also, information on histology in the form of tissue architecture and organization is lost in the process. Recently, new transcriptome-wide analysis technologies have enabled the study of RNA molecules directly in tissue samples, thus maintaining spatial resolution and complementing histological information with molecular information”, on page 531, column 2, paragraphs 3-4 “Another way to provide spatial resolution to scRNAseq analyses is the reverse approach where the localization of single cells within a tissue of origin is inferred by the analysis of their gene expression… In situ techniques maintain spatial information by performing molecular reactions directly in fixed cells and tissue sections. The resolution of these techniques depends on the size and density of the signals generated and the resolution of the imaging system… These highly parallel in situ assays are ideal tools to study molecular composition of cells and tissues in direct relation to morphology and histology”, reading on wherein the spatial data has a resolution that is lower than the single cell resolution, and wherein generating, by the one or more computing devices, based on the mapping matrix, the spatial map indicating the likely spatial locations of the cells within the anatomical region or tissue type at the single cell resolution further comprises: receiving histological data for the first specimen or the second specimen; segmenting cells in the histological data to estimate a number of cells in each voxel of the spatial data; and deconvoluting the spatial data to the single cell resolution using a deterministic mapping that assigns the number of cells to each voxel.
It would have been obvious at the time of first filing to have modified the teachings of Satija et al., Tian et al. and Mahfouz et al. for the method, system, and CRM of claims 1, 26, and 51 with the teachings of Mignardi et al. for the use of histological sectioning and basing resolution on the imaging system, as Mignardi et al. is a review of the current methods for resolving gene expression in specific tissue environments based on spatial resolution, specifically stating in the abstract “At the first level, biochemical techniques are used to generate signals that can be imaged by different means of fluorescence microscopy. At the second level, images are subject to digital image processing and analysis in order to detect and identify the aforementioned signals. At the third level, the collected data are analyzed and transformed into interpretable information by statistical methods and visualization techniques relating them to each other, to spatial distribution, and to tissue morphology. In this review, we describe state-of-the-art techniques used at all three levels of analysis”. One would have had a reasonable expectation of success given that the latter is a review and providing state of the art information and walkthroughs on the implementation of various techniques. Therefore, it would have been obvious at the time of first filing to have modified the teachings of each and to be successful.
Claim 77 is directed to the method of claim 1 but further specifies determining a sectioning depth and mapping single cell data based at that depth.
Satija et al., Tian et al., and Mahfouz et al. teach the method, system and CRM of claims 1, 26, and 51 as previously described.
Mignardi et al. teaches in the abstract “Also, information on histology in the form of tissue architecture and organization is lost in the process. Recently, new transcriptome-wide analysis technologies have enabled the study of RNA molecules directly in tissue samples, thus maintaining spatial resolution and complementing histological information with molecular information”, on page 531, column 2, paragraphs 3-4 “Another way to provide spatial resolution to scRNAseq analyses is the reverse approach where the localization of single cells within a tissue of origin is inferred by the analysis of their gene expression… In situ techniques maintain spatial information by performing molecular reactions directly in fixed cells and tissue sections. The resolution of these techniques depends on the size and density of the signals generated and the resolution of the imaging system… These highly parallel in situ assays are ideal tools to study molecular composition of cells and tissues in direct relation to morphology and histology”, reading on wherein generating the mapping matrix that maps cells from the single cell data to voxels of the spatial data further comprises: determining a sectioning depth of the first specimen based on a histological image of the first specimen; and selectively mapping the single cell data collected at the sectioning depth of the first specimen.
Response to Arguments
Applicant’s arguments with respect to claims 1, 2, 4, 9, 22, 23, 26, 27, 29, 34, 47, 48, 51,52, 54, 59, 72, 73, 76, and 77 have been considered but are not persuasive.
Applicant asserts on page 25 of the Remarks filed 2/2/2026 that the aligning of the single cell data to the spatial data based on shared features was not taught by Tian et al. in the referenced pages of 191 and 192. However, examiner reminds applicant that the spatial data in combination with scRNA-seq data was already a part of the Satija et al. reference and Tian et al. was merely presenting a method that used a feature representation along with clustering of scRNA-seq data and that this combination of the two would lead to an obviousness for feature representation of scRNA-seq data with the spatial in-situ hybridization data of Satija et al. for deriving spatial maps using an unsupervised deep learning nonlinear optimization such as that described in Tian et al. for scDeepCluster.
Applicant asserts that the rationale for combining the two references conflicts with the evidence of record, i.e. that spatial mapping and clustering are similar in attempting to group cells based upon expression. However, this conflicts with the very purpose of the use of expression data for spatial mapping, i.e. the invention uses expression to spatially map cells, which is the same goal as Satija et al., and clustering is definitionally grouping cells based upon a specified metric, expression in this case. The placing of single cell data into space and rearranging them to match spatial data is different from the more general clustering of expression profiles but ignores the similarity and more importantly the elements behind the methods, specifically the unsupervised deep learning nonlinear optimization.
Applicant asserts that the references do not read on the amended claim material. Applicant is correct, however newly cited references have been provided to rectify said deficiencies.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to KEENAN NEIL ANDERSON-FEARS whose telephone number is (571)272-0108. The examiner can normally be reached M-Th, alternate F, 8-5.
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, Karlheinz Skowronek can be reached at 571-272-9047. 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.
/K.N.A./Examiner, Art Unit 1687
/OLIVIA M. WISE/Supervisory Patent Examiner, Art Unit 1685