DETAILED ACTION
Status of the Claims
The following is a non-final Office Action in response to claims filed 15 September 2024.
Claims 1-16 are pending.
Claims 1-16 have been examined.
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Information Disclosure Statement
The information disclosure statement (IDS) submitted on 15 September 2024 are being considered by the Examiner.
Priority
Applicant’s claim for the benefit of a prior-filed application(s) 371 of PCT/KR2023/003364 filed 03/13/2023 under 35 U.S.C. 119(e) or under 35 U.S.C. 120, 121, or 365(c) is acknowledged.
Claim Rejections - 35 USC § 102
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 the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention.
Claim(s) 1-5, 8-13 and 16 is/are rejected under 35 U.S.C. 102(a)(1) as being anticipated by Song et al. (US PG Pub. 2021/0374536).
As per claims 1 and 9, Song discloses a retrosynthetic translation method of predicting a reactant for a product using a neural machine translation (NMT) model based on transformer architecture, and retrosynthetic translation apparatus that predicts a reactant for a product using a neural machine translation (NMT) model based on transformer architecture, comprising: a control unit configured to control the NMT model; a communication unit configured to communicate with an external server; a memory unit; a display unit; and an input unit configured to receive a user input, the method comprising (neural network apparatus, ¶44; processor, memory, user interface, ¶45; transformer, neural network, ¶70) (Examiner notes the neural network apparatus using a transformer model as the equivalent to the neural machine translation model):
preparing an input sequence and an output sequence of the model, in which the input sequence and the output sequence represent molecules as a list of fragments, each fragment constituting the list of fragments is a fragment expressed based on an atom environment (AE), and the product and the reactant are converted into a sequence expressed as the AE and prepared as the input sequence and the output sequence, respectively (input character string sequence, Song ¶62; based upon central atom, ¶70; input sequence can be converted, ¶91; molecules, ¶136) (Examiner notes the molecules as the equivalent to the fragments, as discussed in the instant specification pages 4-5);
training the model using the input sequence and the output sequence (The processor 110 may train the retrosynthesis prediction model based on a result of decoding the second character string information. That the processor 110 trains the retrosynthesis prediction model may mean the training of hidden states of the encoder and the decoder so that the cross-attention matrix catches the atom-mapping information. In addition, that the processor 110 trains the retrosynthesis prediction model may mean the training of information on an edge representing a relationship between a node and a node in the graph information of each test product and test reactant. Further, that the processor 110 trains the retrosynthesis prediction model may mean that the accuracy of the prediction result of the neural network apparatus 100 is increased. According to an embodiment, that the processor 110 trains a retrosynthesis prediction model may mean the training of atom-mapping information indicating a relationship between an atom included in a product and an atom included in a reactant, Song ¶64); and
predicting the reactant by retro-synthesizing the product through the trained model, wherein a new product is converted into a sequence represented by the AE and input as an input sequence of the model, an output sequence is output through the model, and the predicted reactant is detected based on the output sequence (The processor 110 may train a retrosynthesis prediction model through a test data set, and predict at least one combination of reactants from a target product based on the trained retrosynthesis prediction model. The retrosynthesis prediction may denote a prediction of a combination of reactant molecules for synthesizing a target product by searching for an inverse reaction pathway of the target product. The retrosynthesis prediction model may be implemented by using a transformer model. As the neural network apparatus 100 uses the transformer model, parallel processing and rapid computation of data may be performed, Song ¶48-¶49),
wherein the AE is a fragment composed of a central atom having a predetermined radius and its covalent neighbors, and the predetermined radius is a maximum allowable topological distance between the central atom and all covalent atoms (with respect to central atom, Song ¶70; ).
As per claims 2 and 10, Song discloses as shown above. Song further discloses wherein the predetermined radius is the number of bonds on a shortest pathway between atoms (distance, Song ¶10-¶11).
As per claims 3 and 11, Song discloses as shown above. Song further discloses wherein the fragment is expressed as one of a set of AEs (AE0) with a predetermined radius of 0 and a set of AEs (AE2) with a predetermined radius of 1 (The generating of the first mask matrix may include setting any one of nodes included in the received first graph information, as a reference node, expressing the reference node and adjacent nodes present at a distance separated by the preset reference distance from the reference node, among the nodes, as “1”, and expressing remaining nodes, among the nodes, as “0”, Song ¶11).
As per claims 4 and 12, Song discloses as shown above. Song further discloses wherein the fragment is expressed as a combination of a set of AEs (AE0) with a predetermined radius of 0 and a set of AEs (AE2) with a predetermined radius of 1 (The generating of the first mask matrix may include setting any one of nodes included in the received first graph information, as a reference node, expressing the reference node and adjacent nodes present at a distance separated by the preset reference distance from the reference node, among the nodes, as “1”, and expressing remaining nodes, among the nodes, as “0”, Song ¶11).
As per claims 5 and 13, Song discloses as shown above. Song further discloses wherein the AE is expressed as a simplified molecular-input line-entry system arbitrary target specification (SMARTS) pattern (the processor 110 may be able to receive an input of a chemical structure of a product and a reactant in a smiles arbitrary target specification (SMARTS) format or an international chemical identifier (InChi) format, Song ¶72).
As per claims 8 and 16, Song discloses as shown above. Song further discloses wherein the model uses an encoder unit and a decoder unit, and applies a multi-head attention mechanism to each unit to translate the input sequence and the output sequence (encoder, decoder, Song ¶64 and ¶78).
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claim(s) 6-7 and 14-15 is/are rejected under 35 U.S.C. 103 as being unpatentable over Song et al. (US PG Pub. 2021/0374536) and further in view of Ristoski et al. (US PG Pub. 2021/030762).
As per claims 6 and 14, Song discloses as shown above but does not expressly disclose wherein the SMARTS pattern for each of the AEs is associated with a unique integer value.
However, Ristoski teaches wherein the SMARTS pattern for each of the AEs is associated with a unique integer value (Molecular Fingerprints: Molecular fingerprints are representations of molecule structures originally designed to assist in chemical database substructure search, but later used for various, analysis tasks and machine learning applications. The molecular fingerprints follow the idea of applying graph kernels on an input graph in order to generate a feature vector for the graph. An embodiment of the methodology can use the Extended-Connectivity Fingerprints (ECFPs), which are a novel class of topological fingerprints for molecular characterization. ECFPs are circular fingerprints that can be rapidly calculated and can represent an infinite number of different molecular features. The features in the ECFPs represent the presence of particular molecular substructures. The ECFPs can be calculated in 3 steps. First, each atom is assigned a unique integer identifier. Second, each atom identifier is updated with the identifiers of its neighbors. A hash function is used to reduce the concatenated identifiers back into a new, single-integer identifier for each atom. All the atoms are being relabeled with the new identifiers, and the steps are being repeated n times, where n is the radius. In the final step, all duplicate identifiers are removed, and the remaining identifiers define the ECFP fingerprint. The fingerprint represents a bit vector, which can be used for data analysis and machine learning. For example, the vector can be used to calculate similarities between molecules (e.g., using cosine similarity), and as a feature vector for classification, Ristoski ¶46-¶47).
Both the Ristoski and the Song references are analogous in that both are directed towards/concerned with design and discovery of new material with prediction models. Before the effective filing date of the claimed invention, it would have been obvious to a person of ordinary skill in the art to use Ristoski’s method of ECFPs in Song’s system to improve the system and method with reasonable expectation that this would result in a system that can more rapidly calculate a near infinite different molecular features. 0020
The motivation being that a challenge in existing approaches is that these approaches tend to generate very large numbers of candidates, often exceeding tens of millions. While the polymers proposed may all meet the material properties desired, there are many additional practical constraints that such approaches fail to consider, for example, synthetic viability, robust polymerization, compliance with internal and external regulations, availability of feed-stock, and/or others. These constraints represent the institutional knowledge, or “chemical common sense” that a human chemist brings to the problem. As a result, a gap exists that separates computer generated candidates from what can actually be realized in an industrial setting (Ristoski ¶3).
As per claims 7 and 15, Song discloses as shown above but does not expressly disclose wherein the AE is generated by an Extended Circular FingerPrint (ECFP) algorithm (Molecular Fingerprints: Molecular fingerprints are representations of molecule structures originally designed to assist in chemical database substructure search, but later used for various, analysis tasks and machine learning applications. The molecular fingerprints follow the idea of applying graph kernels on an input graph in order to generate a feature vector for the graph. An embodiment of the methodology can use the Extended-Connectivity Fingerprints (ECFPs), which are a novel class of topological fingerprints for molecular characterization. ECFPs are circular fingerprints that can be rapidly calculated and can represent an infinite number of different molecular features. The features in the ECFPs represent the presence of particular molecular substructures. The ECFPs can be calculated in 3 steps. First, each atom is assigned a unique integer identifier. Second, each atom identifier is updated with the identifiers of its neighbors. A hash function is used to reduce the concatenated identifiers back into a new, single-integer identifier for each atom. All the atoms are being relabeled with the new identifiers, and the steps are being repeated n times, where n is the radius. In the final step, all duplicate identifiers are removed, and the remaining identifiers define the ECFP fingerprint. The fingerprint represents a bit vector, which can be used for data analysis and machine learning. For example, the vector can be used to calculate similarities between molecules (e.g., using cosine similarity), and as a feature vector for classification, Ristoski ¶46-¶47).
Both the Ristoski and the Song references are analogous in that both are directed towards/concerned with design and discovery of new material with prediction models. Before the effective filing date of the claimed invention, it would have been obvious to a person of ordinary skill in the art to use Ristoski’s method of ECFPs in Song’s system to improve the system and method with reasonable expectation that this would result in a system that can more rapidly calculate a near infinite different molecular features. 0020
The motivation being that a challenge in existing approaches is that these approaches tend to generate very large numbers of candidates, often exceeding tens of millions. While the polymers proposed may all meet the material properties desired, there are many additional practical constraints that such approaches fail to consider, for example, synthetic viability, robust polymerization, compliance with internal and external regulations, availability of feed-stock, and/or others. These constraints represent the institutional knowledge, or “chemical common sense” that a human chemist brings to the problem. As a result, a gap exists that separates computer generated candidates from what can actually be realized in an industrial setting (Ristoski ¶3).
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant’s disclosure (additional art can be located on the PTO-892):
Sun et al. (US PG Pub. 2023/0223112) Retrosynthesis using neural networks.
Zagribelnyy et al. (US PG Pub. 2023/0154572) Retrosynthesis-related synthetic accessibility.
Manica et al. (US PG Pub. 2022/0359045) Prediction of enzymatically catalyzed chemical reactions.
Any inquiry concerning this communication or earlier communications from the Examiner should be directed to ANDREW B WHITAKER whose telephone number is (571)270-7563. The examiner can normally be reached on M-F, 8am-5pm, EST.
If attempts to reach the examiner by telephone are unsuccessful, the Examiner’s supervisor, Lynda Jasmin can be reached on (571) 272-6782. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of an application may be obtained from Patent Center. Status information for published applications may be obtained from Patent Center. Status information for unpublished applications is available through Patent Center for authorized users only. Should you have questions about access to Patent Center, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free).
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) Form at https://www.uspto.gov/patents/uspto-
automated- interview-request-air-form
/ANDREW B WHITAKER/Primary Examiner, Art Unit 3629