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 Objections
Claim 1 is objected to because of the following informalities. The phrase “obtaining” should be corrected to “obtained” to maintain proper grammatical structure.
Claim 9 is objected to because of the following informalities. The phrase “fist recommendation” should be corrected to “first recommendation”.
Appropriate correction is required.
Claim Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claims 16-20 are rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as failing to set forth the subject matter which the inventor or a joint inventor, or for applications subject to pre-AIA 35 U.S.C. 112, the applicant regards as the invention.
Claim 16 recites “The computer-implemented method of claim 8, further comprising: selecting, by the system, the recommended mass spectrometry action by the neural network from an action database of available mass spectrometry actions, available mass spectrometry actions are capable of being performed using one or more mass spectrometry devices communicatively coupled to the system.”
The claim contains two independent clauses separated only by a comma. The second clause has its own subject and verb and it is not grammatically tied to the preceding clause. The claim could be corrected by incorporating the second sentence into the first using a transitional phrase such as “wherein”. For the purpose of examination, claim 16 is interpreted as if amended to include “wherein”.
Claim 17 recites “ … generate, by the processor, an input dataset, comprising one or more acquisition metrics and one or more associated scores that are associated with the one or more acquisition metrics, corresponding to a type of compound, …”.
It is not clear whether the dataset corresponds to 1. a type of compound or 2. The metrics or 3. The scores. For purpose of examination, it is assumed that the claim is intended to recite acquisition metrics associated with experiments performed on the type of compound.
Claims 18-20 are also rejected because they inherit the issues of claim 17 which they depend from.
The following is a quotation of 35 U.S.C. 112(d):
(d) REFERENCE IN DEPENDENT FORMS. —Subject to subsection (e), a claim in dependent form shall contain a reference to a claim previously set forth and then specify a further limitation of the subject matter claimed. A claim in dependent form shall be construed to incorporate by reference all the limitations of the claim to which it refers.
The following is a quotation of pre-AIA 35 U.S.C. 112, fourth paragraph:
Subject to the following paragraph [i.e., the fifth paragraph of pre-AIA 35 U.S.C. 112], a claim in dependent form shall contain a reference to a claim previously set forth and then specify a further limitation of the subject matter claimed. A claim in dependent form shall be construed to incorporate by reference all the limitations of the claim to which it refers.
Claims 18-20 are rejected under 35 U.S.C. 112(d) or pre-AIA 35 U.S.C. 112, 4th paragraph, as being of improper dependent form for failing to further limit the subject matter of the claim upon which it depends, or for failing to include all the limitations of the claim upon which it depends. Applicant may cancel the claim(s), amend the claim(s) to place the claim(s) in proper dependent form, rewrite the claim(s) in independent form, or present a sufficient showing that the dependent claim(s) complies with the statutory requirements.
Claims 18-20 are the computer program product claims and each depend from claim 15, however claim 15 is directed to a computer-implemented method. Each of claims 18-20 recites limitations that are not necessarily related to or dependent upon the “fragmented acquisition” limitation of claim 15. If applicant amends claim 18-20 to depend from claim 17 instead of from claim 15 the dependency issue would be resolved, and for purpose of examination, claim 18-20 are considered herein as dependent from claim 17.
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 1-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more. The claims recite an abstract idea as discussed below. The claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception for the reasons discussed below.
Step 1 of the 2019 Guidance requires the examiner to determine if the claims are to one of the statutory categories of invention. Applied to the present application, the claims belong to one of the statutory classes of a process or product as a computer implemented method or a computer system/product.
Step 2A of the 2019 Guidance is divided into two Prongs. Prong 1 requires the examiner to determine if the claims recite an abstract idea, and further requires that the abstract idea belongs to one of three enumerated groupings: mathematical concepts, mental processes, and certain methods of organizing human activity.
Claim 1-20 are copied below, with the limitations belonging to an abstract idea being
underlined.
1. A system, comprising: a memory that stores computer executable components; and a processor that executes the computer executable components stored in the memory, wherein the computer executable components comprise: an acquisition component that acquires data for a compound, the data defining a first mass spectrometry spectrum for the compound; an evaluation component that, based on the data, and employing a neural network that is trained on an input dataset comprising an acquisition metric, and employing an associated score that is associated with the acquisition metric, generates a recommendation to perform a mass spectrometry action for the compound; and an execution component that, based on the recommendation, directs execution of the mass spectrometry action at a mass spectrometer and obtaining a mass spectrum result.
2. The system of claim 1, wherein the evaluation component further specifies an amount of a resource to employ for the mass spectrometry action.
3. The system of claim 1, wherein the evaluation component further specifies the compound, or a fragmented compound, from the first mass spectrometry spectrum, as a target of the mass spectrometry action.
4. The system of claim 1, wherein the evaluation component further specifies a plurality of targets corresponding to the compound, to be fragmented from the compound, or corresponding to one or more fragmented compounds from the first mass spectrometry spectrum, for which to obtain additional data by performance of the mass spectrometry action.
5. The system of claim 1, further comprising: a metric component that obtains a metric of interest for the compound, wherein the recommendation to perform the mass spectrometry action is at least partially based on the metric of interest.
6. The system of claim 1, further comprising: a reward component that generates a reward indicator resulting from the recommendation of the mass spectrometry action, or from an execution of the mass spectrometry action recommended; and an updating component that updates one or more weights employed by the NN or performs an adjustment to the acquisition metric, based on the reward indicator.
7. The system of claim 1, further comprising: an updating component that updates the neural network according to a set of reward indicators amortized over time and obtained based on a plurality of recommendations of generations by the evaluation component, including the recommendation to perform the mass spectrometry action.
8. A computer-implemented method, comprising: comparing, by a system operatively coupled to a processor, first data for a compound to an input data set comprising one or more acquisition metrics and one or more associated scores that are associated with the one or more acquisition metrics, for the compound; and based on the comparison, and on an obtained metric of interest associated with the compound, and employing a neural network trained on the input dataset, generating, by the system, a first recommendation to perform a recommended mass spectrometry action for the compound, wherein the recommended mass spectrometry action comprises use of a mass spectrometry device to obtain acquisition of the compound.
9. The computer-implemented method of claim 8, further comprising: identifying, by the system, the compound and a second compound from a mass spectrometry spectrum defined by the first data; and generating in parallel, by the system, the fist recommendation and a second recommendation for the second compound.
10. The computer-implemented method of claim 8, wherein the first recommendation is based on historical data defining one or more results of acquisition of the compound caused by mass spectrometry analysis or other separation analysis of the compound.
11. The computer-implemented method of claim 8, further comprising: generating, by the system, a dataset matrix for the compound, based on the input dataset, and comprising the one or more associated scores, wherein the one or more associated scores define probabilities that one or more thresholds corresponding to the one or more acquisition metrics will be satisfied by one or more additional mass spectrometry actions performed for the compound.
12. The computer-implemented method of claim 8, further comprising: evaluating, by the system, the recommended mass spectrometry action, resulting in a reward indicator obtained by the system, wherein the reward indicator is employed to update one or more weights employed by the neural network.
13. The computer-implemented method of claim 8, further comprising: generating, by the system, an associated score as a number between 0 and 1.
14. The computer-implemented method of claim 8, further comprising: prior to the employing of the neural network to generate the recommended mass spectrometry action, correlating, by the system, the input dataset to the metric of interest.
15. The computer-implemented method of claim 8, further comprising: selecting, by the system, a recommended mass spectrometry action comprising a fragmented acquisition that acquires the compound.
16. The computer-implemented method of claim 8, further comprising: selecting, by the system, the recommended mass spectrometry action by the neural network from an action database of available mass spectrometry actions, available mass spectrometry actions are capable of being performed using one or more mass spectrometry devices communicatively coupled to the system.
17. A computer program product facilitating a process for reinforcement learning-based mass spectrometry control, the computer program product comprising a computer readable storage medium having program instructions embodied therewith, and the program instructions executable by a processor to cause the processor to: generate, by the processor, an input dataset, comprising one or more acquisition metrics and one or more associated scores that are associated with the one or more acquisition metrics, corresponding to a type of compound, wherein the one or more associated scores define probabilities that one or more thresholds corresponding to the one or more acquisition metrics will be satisfied by a mass spectrometry action performed for the type of compound; train, by the processor, a neural network on the input dataset; and generate, by the processor, a recommended mass spectrometry action for obtaining second data on a compound, of the type of compound, by employing the neural network to compare first data for the compound to the input dataset associated with the type of compound.
18. The computer program product of claim 15, wherein the generating of the recommended mass spectrometry action is further based on a metric of interest for the type of compound, which metric of interest is provided by a user entity to tailor functioning of the neural network.
19. The computer program product of claim 15, wherein the generating of the recommended mass spectrometry action further comprises specifying, by the processor, an amount of a resource to employ for the recommended mass spectrometry action, based on historical data defining acquisition of the compound caused by mass spectrometry analysis of the type of compound.
20. The computer program product of claim 15, wherein the recommended mass spectrometry action comprises a fragmented acquisition that is a mass spectrometry/mass spectrometry (MS2) acquisition for the compound or a mass spectrometry/mass spectrometry/mass spectrometry (MS3) acquisition for the compound.
The limitation underlined in Claim 1 recites data collection, mathematical analysis, and decision making which falls within mental process. Claim 2 further recites determining a parameter value based on data analysis. This is still mathematical optimization, decision-making and evaluation of information. Claim 3 recites logical determinations derived from data. That is a form of mental process implemented by computer.
Claim 4 further recites data-analysis and decision making. Claim 5 under BRI recites Collecting desired objectives and applying them to guide decision-making which are mathematical concepts. Claim 6 and claim 7 recite mathematical concepts and mental process, namely reinforcement learning, reward calculation, and weight updating. Updating weights based on rewards and amortizing rewards over time are classical mathematical operations used in training machine learning models. Claims 8-16 recite mathematical processing and evaluation of data (scores, metrics, comparisons, NN outputs) and decision-making process (selecting a recommended action) applied to spectral data. All of these falls within mathematical concepts and mental processes. Claim 17 recite mathematical modeling, statistical scoring, and data analysis (probabilities, comparisons, neural network outputs) as well as decision-making (selecting a recommended action). Claims 18-20 merely refine how the abstract model selects and parametrizes actions using known data, preferences, or acquisition types. These falls within mathematical concepts and mental processes
Claims 1-20 recite mathematical concepts and mental process which are abstract ideas at prong 1 of the 101 analyses.
In Step 2A prong 2: examiner needs to determine if the claim(s) recite additional elements that integrate the exception into a practical application of the exception. The additional elements in the claim have been left in normal font.
Claim 1: The additional elements, memory and processor, provide the generic computing environment to execute the claimed data processing and analysis steps. The acquisition component acquires mass spectral data from MS system and the execution component directs the mass spectrometer to perform an action and obtain result. While these elements reference a mass spectrometer and involve obtaining spectral data, they do not meaningfully limit the abstract idea. The claim does not specify any particular improvement in mass spectrometry technology, any specific control technique that improves instrument operation, or any particular hardware configuration that transforms the abstract idea into a practical technological application. Instead, the mass spectrometer is used merely as a source of data and recipient of the recommended action, while the core of the invention remains the abstract idea of analyzing data using a neural network and generating a recommendation based on that analysis.
Claim 2: The only added limitation to claim 1 under BRI is “allocating more or less of something” which is a routine optimization step. Claim 2 doesn’t add any new technological structure beyond claim 1.
Claim 3: The added limitation only requires that the system identify which compound or fragment should be targeted. It doesn’t require a particular fragmentation technique or a technical modification to the instrument. It doesn’t define how fragmentation improves the instrument’s operation.
Claim 4: The claim merely requires selecting multiple targets for additional data acquisition. The claim does not define how the plurality improves instrument performance structurally.
Claim 5: Under BRI The claim only requires that the recommendation be “at least partially based on” the metric. It doesn’t require particular algorithmic implementation or a change to MS hardware configuration. The claim doesn’t require how the metric structurally modifies the NN.
Claim 6: The additional elements do not integrate the abstract idea into a practical application. Because, the reward generation and updating are generic reinforcement learning (RL) training steps performed on data. The NN used only as a tool to implement the abstract optimization and the claim doesn’t require any improvement to the MS hardware or control mechanism itself.
Claim 7: The updating step is generic RL training operation applied to data and the NN and the processor are used only as tools to implement the mathematical learning process.
Claim 8: the additional elements merely perform generic data acquisition and execution functions and do not impose meaningful limits on the abstract idea. The use of processor and neural network to process data and generate recommendation represent well understood, routine, and conventional computer functionality. No additional element or ordered combination of elements improves the functioning of the computer or the mass spectrometer itself.
Dependent claims 9-16 additional elements, similarly to claim 8, do not improve the functioning of the mass spectrometer or the computer itself, nor do they effect a particular technological improvement in mass spectrometry. The additional elements represent further data analysis, evaluation, and decision-making operations performed by a computing system.
Claim 17: The additional elements, including a processor, computer-readable storage medium merely perform generic data processing, training and comparison operations and they represent well-understood, routine, and conventional computer functionality. The claim therefore amounts to using a generic computer system to analyze data and generate recommendation.
Dependent claims 18-20 additional elements, similarly to claim 17, do not integrate the judicial exceptions into a practical application. The additional elements provide further details regarding the dataset, metrics, scores, or recommendation generation used in the RL-based mass spectrometry control process. The additional elements remain within the scope of the abstract idea in claim 17.
At Step 2B of § 101 analyses examiner needs to determine whether the claim as a whole amount to significantly more than the judicial exception.
Claims 1-20 do not include additional elements that amount to significantly more than the judicial exception. The additional elements recited in the claims, including processors, memory or computer-readable storage media, neural networks, and mass spectrometry system components, are described at a high level of generality and perform generic data collection, data processing, training, comparison, and recommendation functions. These elements represent well-understood, routine, and conventional activities commonly performed by generic computer systems when analyzing data and generating outputs based on that analysis.
When considered individually and as an ordered combination, the additional elements merely implement the abstract idea using generic computing technology and do not improve the functioning of the computer, the neural network, or the mass spectrometer itself. Accordingly, the claims do not recite an inventive concept sufficient to transform the judicial exception into patent-eligible subject matter, and therefore do not amount to significantly more than the abstract idea.
Claim Rejections - 35 USC § 103
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.
Claims 1-3 and 5-20 are rejected under 35 U.S.C. 103 as being unpatentable over Gioioso (US 20220084802) in view of Bazargan (20210074533)
Regarding claim 1 Gioioso (US 20220084802) teaches a system (fig 10; 702, fig 7), comprising: a memory (1010, fig 10; 710, fig 7) that stores computer executable components; and a processor (1012, fig 10; 706, fig 7) that executes the computer executable components (AI Algorithm 726 and AI model 730, fig 7) stored in the memory,
wherein the computer executable components comprise: an acquisition component (The AI model 730 trained on data 712 derived from experimental MS data 708 and associated performance metrics such as intensity 722, and peak shape 724 (¶ [102])) that acquires data for a compound, the data defining a first mass spectrometry spectrum for the compound (P1 in fig 3; ¶ [0067]-¶ [0069]);
an evaluation component (AI model 730 or tuning device 128) that, based on the data (training data 712), and employing an AI model (730) that is trained on an input dataset (712) comprising an acquisition metric (AI model 730 or AI model 132 that can be trained via reinforcement learning (¶ [0021])), and employing an associated score that is associated with the acquisition metric ((probability distributions)134 that assign likelihood (fitness score ¶[92])(¶ [53])), generates a recommendation to perform a mass spectrometry action (system determines a “recommended configuration” of MS operating parameters for the MS apparatus (¶ [0108])) for the compound; and an execution component that, based on the recommendation, directs execution of the mass spectrometry action at a mass spectrometer and obtaining a mass spectrum result (transmitting the selected parameter values to the MS apparatus 812, running the MS apparatus using those parameters, and obtaining resulting spectrometry data 814, fig 8).
Gioioso describes a general AI/ML model and discloses that the other types of AI/ML techniques may be used (¶ [0103] & ¶ [0109]). Gioioso does not explicitly identify the model as a “neural network” (NN).
Bazargan (us 20210074533 A1) teaches the use of ML models, including “neural network” ¶ [0166], for determining or optimizing operating parameters of analytical instruments (including mass spectrometry systems) based on measured data and performance metrics.
It would have been obvious to one ordinary skill in the art at the time of the invention to implement the reinforcement learning framework of Gioioso using a neural network structure as taught by Bazargan to create a control system capable of optimizing hardware parameters in real time for highly dynamic and transient samples, where manual tuning by a human expert is not feasible.
Regarding claim 2, Gioioso in view of Bazargan further teaches wherein the evaluation component (an AI system 702 with AI model 730) further specifies an amount of a resource (selects values for parameters ¶ [6] such as (ion energy, steering voltage, etc.) ¶ [51], ¶[77]) to employ for the mass spectrometry action (to be used during the data gathering process or the AI model 730 generates an output structure 734 to run an experiment ¶ [108] & ¶ [111]).
Regarding claim 3, Gioioso in view of Bazargan further teaches wherein the evaluation component further specifies the compound (ions associated with precursors with low energy spectra¶ [45]), or a fragmented compound (ions primarily from fragmented precursors ¶ [48]), from the first mass spectrometry spectrum (¶ [46]), as a target of the mass spectrometry action (The system is configured to perform a “pre-selection (or targeting) of the targeted precursor”. This pre-selection occurs when a desired mass reaches a specified intensity value in the first (low-energy) spectrum, at which point the system triggers a transition to obtain a fragmentation result ¶ [46]. A person of ordinary skill in the art would understand that selecting a specific mass-to-charge (m/z) value for isolation and fragmentation as taught by Gioioso’s pre-selection protocol, is the functional equivalent of “specifying the compound... as a target” for the MS action.)
Regarding claim 5, Gioioso in view of Bazargan further teaches further comprising: a metric component (AI/ML model 702 or tuning device 128) that obtains a metric of interest (peak shape 724 ¶ [11]) for the compound, wherein the recommendation to perform the mass spectrometry action is at least partially based on the metric of interest (the AI algorithm 726 generates an output structure 734, which is a recommendation for parameter values. This selection is explicitly based on peak shape 724 ¶ [102-108]).
Regarding claim 6, Gioioso in view of Bazargan further teaches further comprising: a reward component (the tuning device 128 and its internal evaluation logic 132 which operates within RL framework (¶ [56], ¶ [104])) that generates a reward indicator (The system is configured to assign a score (such as fitness score, (likelihood that a selected value optimizes the goal)) ¶ [14]. The tuning device 128 deploys AI model 132 that includes probability distributions 134 ¶ [52] & ¶ [53]. Moreover, the system may penalize/reinforce configurations ¶ [118] according to the scores (probability distributions) assigned to the configuration) resulting from the recommendation of the mass spectrometry action, or from an execution of the mass spectrometry action recommended (tuning device 128 evaluates the “result of a recommended configuration” after it is run on the MS apparatus ¶ [108].);
and an updating component (tuning device 128, that is responsible for executing the logic described in updating step 818) that updates (updates NN by the AI based on the scores assigned to the configurations (fig. 8, block 818)) one or more weights employed by the NN (the “amount of adjustment may be weighted based on the score” ¶ [122]) or performs an adjustment to the acquisition metric, based on the reward indicator (the system determines if values for parameters reduced data variability and, based on that determination, updates the model to select new sets of values, which constitutes an “adjustment to the acquisition metric” logic, fig8 blocks 814, 816 and 818).
Regarding claim 7, Gioioso in view of Bazargan further teaches further comprising: an updating component (the tuning device 128) that updates the neural network according to a set of reward indicators (fig. 8, block 818) amortized over time (the system considers the standard deviation of fitness scores across the last 10 maximum scores to determine if the model has truly improved ¶ [98]) and obtained based on a plurality of recommendations of generations by the evaluation component, including the recommendation to perform the mass spectrometry action (The system covers a plurality of recommendations through its evolutionary algorithms which generate and evaluate multiple “candidate solutions” or “configurations” in iterative batches called generations ¶ [71]. The system can select one or more candidate configurations to be tested by the MS apparatus ¶ [131]).
Regarding claim 8, Gioioso in view of Bazargan teaches a computer-implemented method and an AI-based tuning system for a mass spectrometry apparatus ¶ [23] & ¶ [123] comprising: comparing (new MS results against the model-derived metrics and scores to determine whether a configuration reduces data variability (¶ [0068], ¶ [0102])), by a system (computing device 702) operatively coupled to a processor (706, fig 7 and 1012 fig 10), first data (under BRI the first data is data that is associated with first spectrum as disclosed in claim 1) for a compound (precursor ion ¶[42]) to an input data set (716, fig. 7) comprising one or more acquisition metrics (peak shape 724) and one or more associated scores (fitness value ¶[33]) that are associated with the one or more acquisition metrics (fitness value of a peak shape 724), for the compound; and based on the comparison , and on an obtained metric of interest associated with the compound (The parameter configurations are generated based on the comparison of measured MS data to the metrics and scores, and may be tailored according to a user-specified metric of interest ( peak shape 724) (¶ [0102], ¶ [0053])),
and employing neural network trained on the input dataset (AI model 730 that is trained on 712), generating (output structure 734), by the system, a first recommendation to perform a recommended mass spectrometry action (¶ [108]) for the compound, wherein the recommended mass spectrometry action comprises use of a mass spectrometry device to obtain acquisition of the compound (tuning routine 800, fig. 8).
Regarding claim 9, Gioioso in view of Bazargan further teaches further comprising: identifying, by the system, the compound and a second compound (precursors) from a mass spectrometry spectrum defined by the first data (low energy spectra ¶[45] contain precursors which are examined by the software in real time ); and generating (The system is configured to identify when “a precursor ion” reaches a specified intensity value to trigger a transition to a high-energy state ¶[46]) in parallel (Gioioso explicitly describes “leveraging the parallel operation of various mass analyzers within the instrument” to achieve high acquisition rates ¶ [43]), by the system, the first recommendation and a second recommendation for the second compound.
The AI model 730 is designed to search a multi-dimensional space to select a “set of values for the parameters” (plural) predicted to reduce data variability [fig. 8]. Gioioso further teaches that parameters for a “plurality of MS apparatuses” can be “tuned as a group” ¶ [16].
This disclosure of managing a "set of values" and operating analyzers in "parallel" to optimize the "group" renders the limitation of generating recommendations “in parallel” for multiple compounds obvious to a person of ordinary skill in the art.
Regarding claim 10, Gioioso in view of Bazargan further teaches wherein the first recommendation (AI model 730 uses the data 708 to generate output structure with recommended configuration) is based on historical data (the AI model may make use of experimental data 708 which comprises “pre-existing data” ¶ [101]) defining one or more results of acquisition of the compound caused by mass spectrometry analysis (the historical data 708 is taught to include an “identification of a configuration of an MS apparatus” ¶ [101]) or other separation analysis (the system integrates results from a liquid chromatograph 104 which performs the separation analysis) of the compound.
Regarding claim 11, Gioioso in view of Bazargan further teaches further comprising: generating, by the system, a dataset matrix (model 132, probability distributions 134 and covariance matrix 136 which are multidimensional representations of parameters and their relationships) for the compound, based on the input dataset, and comprising the one or more associated scores, wherein the one or more associated scores ( system populates these structures with “probability distributions and likelihoods ” ¶ [53-55]) define probabilities that one or more thresholds corresponding to the one or more acquisition metrics will be satisfied by one or more additional mass spectrometry actions performed for the compound (further system uses these scores to predict and evaluate whether the experimental results will “satisfy predetermined thresholds” ¶ [121].).
Regarding claim 12, claim 12 depends on claim 8 which was rejected above. Claim 12 is rejected for the same reasons as set forth with respect to claim 6.
Regarding claim 13, Gioioso in view of Bazargan further teaches further comprising: generating, by the system, an associated score (fitness score ¶ [88]) as a number between 0 and 1 (a normalization process used to convert raw experimental metrics (like resolution or intensity) into a standardized score range of 0 to 1 ¶ [0085]).
Regarding claim 14, Gioioso in view of Bazargan further teaches further comprising: prior to the employing of the neural network (model 730) to generate the recommended mass spectrometry action (output structure 734 with recommended configurations), correlating (Gioioso explicitly builds correlations between parameters and data variability metrics during model construction block 204, block 206 and ¶ [59- 64].), by the system (700), the input dataset ( experimental 708) to the metric of interest (peak shape 724).
Regarding claim 15, Gioioso in view of Bazargan further teaches further comprising: selecting (system evaluates and selects ((tuning routine 800, fig 8) configurations based on criteria such as “fragmentation ratio” and “fragmentation efficiency”. By recommending specific parameter values to optimize these fragmentation metrics, the system functionally selecting “fragmented acquisition” ¶[69].), by the system (model 730), a recommended mass spectrometry action (output structure 734) comprising a fragmented acquisition that acquires the compound (the MS system includes a collision cell 118 that fragments precursor ions and generates fragment spectra using high energy acquisition modes, thereby performing a fragmented acquisition of a compound ¶ [45]).
Regarding claim 16, Gioioso in view of Bazargan further teaches further comprising: selecting (block 810), by the system, the recommended mass spectrometry action by the neural network (model 730) from an action database of available mass spectrometry actions (search space 804 (all the possible recommended configurations for MS apparatus ¶ [67])),wherein available mass spectrometry actions are capable of being performed using one or more mass spectrometry devices communicatively coupled to the system (the selected values are transmitted to the MS apparatus (block 812) which is communicatively coupled to the tuning device ( figure 1) and the MS apparatus then runs an experiment using these values, demonstrating that the actions were capable of being performed by the coupled hardware ¶[68]).
Regarding claim 17, Gioioso in view of Bazargan teaches a computer program product and an AI-based tuning system for a mass spectrometry apparatus ¶ [23] & ¶ [123], facilitating a process for reinforcement learning-based (AI model can be trained using Reinforcement Learning ¶ [21]) mass spectrometry control, the computer program product comprising a computer readable storage medium having program instructions embodied therewith, and the program instructions executable by a processor to cause the processor (tuning system and AI model logic may be implemented as “instructions stored on a non-transitory computer readable storage medium” and embodied on an apparatus with a processor ¶ [139]) to: generate, by the processor (1012, fig 10), an input dataset (experimental data 708 and training data 712), comprising one or more acquisition metrics (peak shape 724) and one or more associated scores (((probability distributions)134 that assign likelihood (fitness score ¶[92])(¶ [53]))) that are associated with the one or more acquisition metrics (peak shape, fig 6) , corresponding to a type of compound (the tuning construct can be “specific to a particular type of sample” ¶ [115]), wherein the one or more associated scores define probabilities that one or more thresholds corresponding to the one or more acquisition metrics will be satisfied by a mass spectrometry action performed (the AI model utilizes an experimental cumulative distribution function (CDF) to determine fitness scores ¶ [85]. These CDF-based scores are calculated for a particular type of sample, the resulting scores define the probability that a mass spectrometry action (tuning) will satisfy the performance thresholds ¶ [121] specifically for that category/type of compound.) for the type of compound (the models and probability distributions can be application specific or sample specific ¶[115].); train (AI model 730 using training data 712), by the processor ( processor circuit 706), a neural network (AI model 730) on the input dataset (experimental data 708 and training data (712)); and generate, by the processor (processor circuit 706), a recommended mass spectrometry action (output structure 734) for obtaining second data (Block 820 and ¶ [67-69]) on a compound, of the type of compound, by employing the neural network to compare (¶ [0068], ¶ [0102]), first data (the low energy spectrum) for the compound to the input dataset( experimental data 708 and training data 712) associated with the type of compound (When the system receives the initial results, it evaluates them against the “type specific” model and the second data are for that specific type of sample. The system can tune on a “sample by sample basis” ¶ [13]).
Regarding claim 18, Gioioso in view of Bazargan further teaches wherein the generating of the recommended mass spectrometry action (output structure 734 ¶ [108]) is further based on a metric of interest (peak shape 724) for the type of compound (tunning may be tailored to particular sample type ¶ [115]), which metric of interest is provided by a user entity (that the system may be configured based on user-specified goals and tradeoffs ¶ [13, 137]. The AI search is not random but is directed or narrowed based on the user’s desired outcomes, the system may adjust the model based on the user’s criteria ¶ [14]) to tailor functioning of the neural network.
Regarding claim 19, Gioioso in view of Bazargan further teaches wherein the generating of the recommended mass spectrometry action (output structure 734 ¶[108]) further comprises specifying, by the processor (processor circuit 706), an amount of a resource (selects values for parameters ¶ [6] such as (ion energy, steering voltage, etc.) ¶ [51], ¶[77]) to employ for the recommended mass spectrometry action (to be used during the data gathering process; or the AI model 730 generates an output structure 734 to run an experiment ¶ [108] & ¶ [111]), based on historical data (experimental data 708 and training data 712) defining acquisition of the compound caused by mass spectrometry analysis of the type of compound (tunning may be tailored to particular sample type ¶ [115]).
Regarding claim 20, Gioioso in view of Bazargan further teaches wherein the recommended mass spectrometry action (output structure 734 ¶ [108]) comprises a fragmented acquisition (the collision cell 118 that performs fragmentation of the precursor. Moreover, the system is configured to analyze both precursor and fragmented ions simultaneously ¶ [41-42]) that is a mass spectrometry/mass spectrometry (MS2) acquisition for the compound (The high-energy spectra ¶ [42]).
Claim 4 is rejected under 35 U.S.C. 103 as being unpatentable over Gioioso (US 20220084802) in view of Bazargan (20210074533) as applied to claims 1-3 and 5-19 above, and further in view of Erikson (J. Proteome Res. 2019, 18, 1299−1306).
Regarding claim 4, Gioioso in view of Bazargan teaches the system of claim 1 including an evaluation component (AI model 730).
However, Gioioso in view of Bazargan does not teach a plurality of targets corresponding to the compound, to be fragmented from the compound, or corresponding to one or more fragmented compounds from the first mass spectrometry spectrum, for which to obtain additional data by performance of the mass spectrometry action.
Erikson further specifies a plurality of targets (Synchronous Precursor Selection (SPS) workflow where the evaluation logic selects "10 fragment ions" as targets for the subsequent scan(page 1301, section MS analysis)) corresponding to the compound (the peptide)(page 1299, Abstract), to be fragmented from the compound (It specifically selects targets that "match a b- or y-type ion" from the identified peptide (page 1301, section MS analysis))), or corresponding to one or more fragmented compounds (the algorithm evaluates the ITMS2 spectrum in real time to identify fragments (figure 2, (C))) from the first mass spectrometry spectrum (ITMS2 scan), for which to obtain additional data (quantitative profiling, reporter ion intensities and SNR ratios (page 1299, Abstract & page 1301 section “proteome informatic”)) by performance (the client application instructs the instrument to collect the scan) of the mass spectrometry action (acquisition of the SPS-MS3 scan (page 1299, Introduction))).
It would have been obvious to a person of ordinary skill in the art to incorporate the real-time multi target fragmentation techniques of Erikson into the AI driven mass spectrometry system of Gioioso to improve analytical output.
Relevant Prior Art
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Cirovic (TrAc Trends in Analytical chemistry, Volume 16, issue 3, March 1997, Pages 148-155) discloses that artificial neural networks are ideally suited for modeling complex non-linear relationships where explicit mathematical equations are unknown or unsolvable.
Erikson (J. Proteome Res. 2019, 18, 1299−1306) discloses a Real-Time Search (RTS)-MS3 methodology that optimizes quantitative workflows by selectively triggering MS3 scans only upon confident peptide identifications.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to SAEEDE NAFOOSHE whose telephone number is (571)272-8629. The examiner can normally be reached Monday-Friday 7:00 am -3:00pm.
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, Andrew Schecter can be reached at 571-272-2302. 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.
/SAEEDE NAFOOSHE/Examiner, Art Unit 2857
/ANDREW SCHECHTER/Supervisory Patent Examiner, Art Unit 2857