DETAILED ACTION
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 .
Status of Claims
Claims 1, 6, 8, 13, 15, 18 and 20 are amended, and claims 5, 12 and 19 are canceled.
Claims 1-4, 6-11, 13-18 and 20 are pending.
Claims 1-4, 6-11, 13-18 and 20 are rejected (Final Rejection).
Response to Amendments
Applicant’s amendments to the drawings and claims (dated: 24 October 2025) obviate the prior drawing objections, claim objections and 35 U.S.C. § 112(b) rejections.
For these reasons, the previous drawing objections, claim objections and 35 U.S.C. § 112(b) rejections have been withdrawn.
Response to Arguments
Applicant’s arguments filed 24 October 2025, at Page 5, with respect to the rejections under 35 U.S.C. § 101 have been fully considered and they are found persuasive. Accordingly, the prior 35 U.S.C. 101 rejections have been withdrawn.
The arguments regarding the rejections under 35 U.S.C. § 103 challenge certain limitations. These limitations are newly added and were therefore not addressed in the previous rejection; therefore, the arguments are moot. The amendments are newly addressed by the new grounds of rejection under 35 U.S.C. § 103.
Claim Rejections - 35 U.S.C. § 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.
Claims 1-4, 6-11, 13-18 and 20 are rejected under 35 U.S.C. § 102(a)(1) as being unpatentable over FAN et al. (U.S. Patent Publication No. 2018/0341754 A1), which was cited in Applicant’s 03/24/2025 IDS, in view of ZHANG et al. (U.S. Patent Publication No. 2021/0027860 A1), and further in view of WEI et al. (U.S. Patent Publication No. 2019/0304568 A1).
Regarding claim 1, FAN discloses a computer-implemented method for predicting molecular conformation (predicting a conformation of a ligand docked into a protein, Abstract of FAN), the method comprising: determining a data structure representing chemical identities of atoms in a molecule (system may derive quantized information on the input based on its chemical formula, a chemical name, a high-resolution image of the crystal structure, a chemical drawing, data about the molecule, data about the atoms comprising the molecule, data about atom interactions, Para. [0077] of FAN; See also extracted features comprise features of a first atom and features of the interaction between the first atom and a second atom … the features of the first atom comprise one or more of an atom type, a radius of the atom, a number of rings in which the atom is included, a size of the ring in which the atom is included, whether the first atom is part of an aromatic ring, and the pairwise potential of the first atom, Para. [0017] of FAN; See also Table 1 provides a list of exemplary atom features that may comprise a feature vector for atom a, Paras. [0079] & [0080] of FAN; [at least a feature vector corresponds to a data structure]); training an energy potential model using a training set comprising the data structure, true conformations of the molecule, and false conformations of the molecule (in step 1502, the processor obtains the training data … the processor may obtain correct ligand conformations based on ligand-receptor complexes in PDB files … the processor may also generate incorrect conformations used for the training … in step 1504, the processor extracts the features related to each conformation … in step 1506, the processor uses the extracted features to construct a feature vector for each conformation … in step 1508, the processor trains a classification model or a ranking model to search for the weight vector {right arrow over (W)}, Para. [0127] of FAN; See also the processor uses the classification model or ranking model trained in steps 1502-1508 to calculate the energy scores of the conformations, Para. [0129]; See also the scoring function H(distance, anglescore) may be used to calculate a pairwise potential energy score of the interaction of both atoms, Para. [0101] of FAN; See also after the distances and angle scores are determined, the atom pairwise potential energy may be determined, Para. [0099] of FAN; See also the machine-learning algorithm for determining the weight vector is trained on real-world protein structure data … the training of the machine-learning algorithm comprises determining a weight vector {right arrow over (W)}=(w1, w2, w3, . . . , wn) for the real-world protein structure data, where, when the feature vector for the correct ligand conformation equals (x1, x2, x3, . . . , xn) and the feature vector for an incorrect ligand conformation is (y1, y2, y3, . . . , yn), Para. [0021] of FAN; See also Paras. [0100], [0125] & [0126] of FAN); determining, using the trained energy potential model, a potential function associated with the molecule (method 1500 uses the feature vectors and weight vectors to construct implicit energy terms and use a machine-learning algorithm to derive the correct energy scoring functions, Para. [0130] of FAN; See also the scoring function H(distance, anglescore) may be used to calculate a pairwise potential energy score of the interaction of both atoms, Para. [0101] of FAN; See also ranking model trained in steps 1502-1508 to calculate the energy scores of the conformations, Para. [0129] of FAN; See also in step 1520, the processor outputs the energy scores, Para. [0129] of FAN; See also Para. [0100] of FAN), wherein determining the potential function associated with the molecule comprises: converting the data structure into a neural network based (a scoring function H(x) may be applied to a ligand atom and a receptor atom to obtain a score representing the interaction for that pair of atoms … all interactions between the ligand atom and potentially many receptor atoms are scored using the H(x) function … the sum of these scores is the pairwise potential of that ligand atom … the H(x) function may be developed by machine learning algorithms, such as the H(distance, anglescore) function described below, Para. [0092] of FAN; See also in the neural network model, the pairwise potential score of an atom may be used as one feature for each atom, as shown in the last row of Table 1, above, Para. [0093] of FAN; See also an atom's pairwise potential relates to forces between that atom and atoms of the receptor protein, such as van der Waals force and electrostatic force, Para. [0094] of FAN; See also para [0095], [0099]-[0102] and [0129] of FAN); and determining a conformation of the molecule based on potential function (the processor may predict the conformations of the ligand, or portion of the ligand, based on the energy scores associated with the conformations, Para. [0129] of FAN; See also method 1500 can accurately predict the ligand or ligand portion conformations, Para. [0130] of FAN).
Although (as discussed above) FAN discloses a van der Waals related potential energy function, FAN does not appear to explicitly disclose the potential function is a smooth function and wherein the neural network based smooth potential function comprises a polynomial-like function that transforms embeddings of atom types associated with the atoms through a logarithmic scale transformation and an exponential transformation. ZHANG, however, is in the same field of complex protein design modeling (Para. [0010] of ZHANG) and discloses a Van der Waals related potential function is a smooth function (the methods provided herein include smoothing the input data, such as the data of a gridded box or local structural information … the smoothing of the input data may include applying three dimensional truncated Gaussian functions with standard deviation ax=σy=σz=r, wherein r is equal to ±0.5 angstrom of a Van der Waals radius of an atom in a voxel of the gridded box, Para. [0045] of ZHANG).
It would have been obvious for one of ordinary skill in the art before the effective filing date of the invention to modify the potential function having the Van der Waals function of FAN with the smooth Van der Waals function of ZHANG [to arrive at the claimed features] for the purpose of improving performance (Para. [0041] of ZHANG).
FAN as modified by ZHANG does not appear to explicitly disclose wherein the neural network based smooth potential function comprises a polynomial-like function that transforms embeddings of atom types associated with the atoms through logarithmic scale transformation and an exponential transformation.
WEI, however, is in the field of predicting molecules/biomolecules using machine learning/neural network algorithms (Abstract of WEI) and discloses wherein the neural network based (predicting structure-function relationships is associated with potential function, Para. [0004] of WEI; See also neural networks, Para. [0052] of WEI) comprises a polynomial-like function that transforms embeddings of atom types associated with the atoms (the physical features used as inputs to machine learning algorithms includes atomic type, atomic charge, Para. [0052] of WEI) through logarithmic scale transformation (data 1002, corresponding to log(P), and data 1004, corresponding to log(S) may be calculated for the biomolecular complex, Para. [0152] of WEI) and an exponential transformation (exponential function, Para. [0181] of WEI; See also exponential kernel function of Para. [0095], where X denotes a type of heavy atoms in the protein (Pro) and Y denotes a type of heavy atoms in the ligand (LIG), Para. [0096] of WEI).
It would have been obvious for one of ordinary skill in the art before the effective filing date of the invention to modify the smooth potential function of FAN as modified by ZHANG with the logarithmic and exponential transformations of WEI [to arrive at the claimed features] for the purpose of screening and predicting the activity of classes of biomolecules (Para. [0003] of WEI).
Regarding claim 2, FAN as modified by ZHANG and WEI teaches the method according to claim 1, wherein determining the data structure representing the chemical identities of the atoms in the molecule comprises: training a neural network to determine embeddings of atom types associated with the atoms (the machine-learning algorithm for determining the weight vector is trained on real-world protein structure data … the training of the machine-learning algorithm comprises determining a weight vector {right arrow over (W)}=(w1, w2, w3, . . . , wn) for the real-world protein structure data, where, when the feature vector for the correct ligand conformation equals (x1, x2, x3, . . . , xn) and the feature vector for an incorrect ligand conformation is (y1, y2, y3, . . . , yn), Para. [0021]; See also in step 1502, the processor obtains the training data … the processor may obtain correct ligand conformations based on ligand-receptor complexes in PDB files … the processor may also generate incorrect conformations used for the training … in step 1504, the processor extracts the features related to each conformation … in step 1506, the processor uses the extracted features to construct a feature vector for each conformation … in step 1508, the processor trains a classification model or a ranking model to search for the weight vector {right arrow over (W)}, Para. [0127]; See also Table 1 provides a list of exemplary atom features that may comprise a feature vector for atom a, Paras. [0079] & [0080]; See also for each pair of atoms in a certain molecular environment, there may be a unique function H(distance,anglescore) based on atom types and molecular environments of both atoms … the unique H(distance,anglescore) for the pair of atoms may be trained using machine learning algorithms, Para. [0099]; See also Para. [0020]; Examiner’s Note: Applicant’s specification, at Para. [0040], supports a vector corresponding to an embedding]).
Regarding claim 3, FAN as modified by ZHANG and WEI teaches the method according to claim 2, wherein input to the neural network comprises a graphic representation of the molecule (system may derive quantized information on the input based on … a high-resolution image of the crystal structure … data about the atoms comprising the molecule, data about atom interactions, Para. [0077] of FAN; See also the deep neural network based ranking model was trained using a two-class classification setting … high resolution X-ray crystallography structures of receptor-ligand complexes deposited in the Research Collaboratory for Structural Bioinformatics (RCSB) protein database (PDB) were taken as true examples, Para. [0143] of FAN; See also Paras. [0100] & [0144]-[0147] of FAN).
Regarding claim 4, FAN as modified by ZHANG and WEI teaches the method according to claim 2, further comprising: extracting features of the atoms from an input file of the neural network (protein structural information used in the disclosed embodiments may be extracted from the PDB data, which may be organized in various file formats, such as PDB file format, Extensible Markup Language (XML) file format, or macromolecular Crystallographic Information File (mmCIF) format, Para. [0076]; See also method for generating the atom types includes extract information regarding the bond environment of each atom in the amino acids of a protein, Para. [0082] of FAN; See also the extracted features comprise features of a first atom and features of the interaction between the first atom and a second atom, Para. [0017] of FAN; See also FIG. 1 is a flowchart of an example of a neural network architecture according to one embodiment, Para. [0038] of FAN; See also data regarding the protein environment may be extracted from a PDB file and include the conformations and sequences of other amino acids surrounding the ligand or ligand portion to be predicted, Para. [0128] of FAN), wherein the features includes one or more of: an element type of a first atom in the molecule, an electrostatic charge of the first atom, a van der Wells radius or a covalent radius of the first atom, information indicating whether the first atom is part of a ring, information indicating whether the first atom is part of an aromatic ring, information indicating whether the first atom forms a single, double, triple, or aromatic bond, or information indicating whether the first atom and a second atom of the molecule are in a same ring (the features of the first atom comprise one or more of an atom type, a radius of the atom, a number of rings in which the atom is included, a size of the ring in which the atom is included, whether the first atom is part of an aromatic ring, and the pairwise potential of the first atom, Para. [0017] of FAN; See also van der Waals radius and covalent radius of atom, Para. [0080] of FAN; See also Table 1 (“Atom Feature”) of Para. [0080] of FAN).
Regarding claim 6, FAN as modified by ZHANG and WEI teaches the method according to claim 1, wherein the smooth potential function comprises potential terms representing one or more of: a bonded potential, an angle potential, a dihedral potential, an out of plane potential, an unbonded pairwise potential, an unbonded angle potential, or an unbonded dihedral potential. (different terms of the atom pairwise potential may be merged … for example, if the atom pairwise potential includes a term F1 expressed in F1(distance), a term F2 expressed in F2 (distance), then a new term F may be defined according to: F(distance F1(distance)+F2 (distance) … therefore, any number of explicit pairwise energy functions can be merged to a single implicit scoring function H(x), which may be the H(distance,anglescore) function introduced below … this way, the pairwise potential is described by implicit potential terms instead of explicit potential terms, Para. [0095 of FAN]; See also angle score; dihedral potential, Paras. [0096]-[0097] and [0117]-[0120] of FAN).
Regarding claim 7, FAN as modified by ZHANG and WEI teaches the method according to claim 1, wherein the molecule is an amino acid side chain or a ligand docked in another molecule. (amino acid side chain, Paras. [0067] & [0082] of FAN; See also atom types are essential for ranking the potential energies of the possible side chain conformations, Para. [0080]; See also atoms found in the 20 common amino acids are classified into 23 atom types, Para. [0086]; See also the conformations of the sampled anchors for a particular ligand with the higher energy scores are more appropriate to act as anchors, Para. [0129] of FAN; See also Paras. [0021], [0024], [0128] and [0130] of FAN).
Regarding claim 8, FAN teaches an apparatus for predicting molecular conformation (predicting a conformation of a ligand docked into a protein, Abstract of FAN; See also FIG. 13 is a block diagram of a device 1300 for predicting ligand conformations, Para. [0171] of FAN), the apparatus comprising: at least one memory for storing instructions (memory 1320, Para. [0175]); and at least one processor configured to execute the instructions to cause the apparatus to (memory 1320 is configured to store various types of data and/or instructions to support the operation of device 1300 … memory 1320 may include a non-transitory computer-readable storage medium including instructions for applications or methods operated on device 1300, executable by the one or more processors of device 1300, Para. [0175] of FAN) perform: determining a data structure representing chemical identities of atoms in a molecule (system may derive quantized information on the input based on its chemical formula, a chemical name, a high-resolution image of the crystal structure, a chemical drawing, data about the molecule, data about the atoms comprising the molecule, data about atom interactions, Para. [0077] of FAN; See also extracted features comprise features of a first atom and features of the interaction between the first atom and a second atom … the features of the first atom comprise one or more of an atom type, a radius of the atom, a number of rings in which the atom is included, a size of the ring in which the atom is included, whether the first atom is part of an aromatic ring, and the pairwise potential of the first atom, Para. [0017] of FAN; See also Table 1 provides a list of exemplary atom features that may comprise a feature vector for atom a, Paras. [0079] & [0080] of FAN; [at least a feature vector corresponds to a data structure]); training an energy potential model using a training set comprising the data structure, true conformations of the molecule, and false conformations of the molecule (in step 1502, the processor obtains the training data … the processor may obtain correct ligand conformations based on ligand-receptor complexes in PDB files … the processor may also generate incorrect conformations used for the training … in step 1504, the processor extracts the features related to each conformation … in step 1506, the processor uses the extracted features to construct a feature vector for each conformation … in step 1508, the processor trains a classification model or a ranking model to search for the weight vector {right arrow over (W)}, Para. [0127] of FAN; See also the processor uses the classification model or ranking model trained in steps 1502-1508 to calculate the energy scores of the conformations, Para. [0129]; See also the scoring function H(distance, anglescore) may be used to calculate a pairwise potential energy score of the interaction of both atoms, Para. [0101] of FAN; See also after the distances and angle scores are determined, the atom pairwise potential energy may be determined, Para. [0099] of FAN; See also the machine-learning algorithm for determining the weight vector is trained on real-world protein structure data … the training of the machine-learning algorithm comprises determining a weight vector {right arrow over (W)}=(w1, w2, w3, . . . , wn) for the real-world protein structure data, where, when the feature vector for the correct ligand conformation equals (x1, x2, x3, . . . , xn) and the feature vector for an incorrect ligand conformation is (y1, y2, y3, . . . , yn), Para. [0021] of FAN; See also Paras. [0100], [0125] & [0126] of FAN); determining, using the trained energy potential model, a potential function associated with the molecule (method 1500 uses the feature vectors and weight vectors to construct implicit energy terms and use a machine-learning algorithm to derive the correct energy scoring functions, Para. [0130] of FAN; See also the scoring function H(distance, anglescore) may be used to calculate a pairwise potential energy score of the interaction of both atoms, Para. [0101] of FAN; See also ranking model trained in steps 1502-1508 to calculate the energy scores of the conformations, Para. [0129] of FAN; See also in step 1520, the processor outputs the energy scores, Para. [0129] of FAN; See also Para. [0100] of FAN) wherein determining the potential function associated with the molecule comprises: converting the data structure into a neural network based smooth potential function (a scoring function H(x) may be applied to a ligand atom and a receptor atom to obtain a score representing the interaction for that pair of atoms … all interactions between the ligand atom and potentially many receptor atoms are scored using the H(x) function … the sum of these scores is the pairwise potential of that ligand atom … the H(x) function may be developed by machine learning algorithms, such as the H(distance, anglescore) function described below, Para. [0092] of FAN; See also in the neural network model, the pairwise potential score of an atom may be used as one feature for each atom, as shown in the last row of Table 1, above, Para. [0093] of FAN; See also an atom's pairwise potential relates to forces between that atom and atoms of the receptor protein, such as van der Waals force and electrostatic force, Para. [0094] of FAN; See also para [0095], [0099]-[0102] and [0129] of FAN); and determining a conformation of the molecule based on potential function (the processor may predict the conformations of the ligand, or portion of the ligand, based on the energy scores associated with the conformations, Para. [0129] of FAN; See also method 1500 can accurately predict the ligand or ligand portion conformations, Para. [0130] of FAN).
Although (as discussed above) FAN discloses a van der Waals related potential energy function, FAN does not appear to explicitly disclose the potential function is a smooth function and wherein the neural network based smooth potential function comprises a polynomial-like function that transforms embeddings of atom types associated with the atoms through a logarithmic scale transformation and an exponential transformation. ZHANG, however, is in the same field of complex protein design modeling (Para. [0010] of ZHANG) and discloses a Van der Waals related potential function is a smooth function (the methods provided herein include smoothing the input data, such as the data of a gridded box or local structural information … the smoothing of the input data may include applying three dimensional truncated Gaussian functions with standard deviation ax=σy=σz=r, wherein r is equal to ±0.5 angstrom of a Van der Waals radius of an atom in a voxel of the gridded box, Para. [0045] of ZHANG).
It would have been obvious for one of ordinary skill in the art before the effective filing date of the invention to modify the potential function having the Van der Waals function of FAN with the smooth Van der Waals function of ZHANG [to arrive at the claimed features] for the purpose of improving performance (Para. [0041] of ZHANG).
FAN as modified by ZHANG does not appear to explicitly disclose wherein the neural network based smooth potential function comprises a polynomial-like function that transforms embeddings of atom types associated with the atoms through logarithmic scale transformation and an exponential transformation.
WEI, however, is in the field of predicting molecules/biomolecules using machine learning/neural network algorithms (Abstract of WEI) and discloses wherein the neural network based (predicting structure-function relationships is associated with potential function, Para. [0004] of WEI; See also neural networks, Para. [0052] of WEI) comprises a polynomial-like function that transforms embeddings of atom types associated with the atoms (the physical features used as inputs to machine learning algorithms includes atomic type, atomic charge, Para. [0052] of WEI) through logarithmic scale transformation (data 1002, corresponding to log(P), and data 1004, corresponding to log(S) may be calculated for the biomolecular complex, Para. [0152] of WEI) and an exponential transformation (exponential function, Para. [0181] of WEI; See also exponential kernel function of Para. [0095], where X denotes a type of heavy atoms in the protein (Pro) and Y denotes a type of heavy atoms in the ligand (LIG), Para. [0096] of WEI).
It would have been obvious for one of ordinary skill in the art before the effective filing date of the invention to modify the smooth potential function of FAN as modified by ZHANG with the logarithmic and exponential transformations of WEI [to arrive at the claimed features] for the purpose of screening and predicting the activity of classes of biomolecules (Para. [0003] of WEI).
Regarding claim 15, FAN teaches a non-transitory computer readable storage medium storing a set of instructions that are executable by one or more processing devices to cause an apparatus to perform a method (memory 1320 is configured to store various types of data and/or instructions to support the operation of device 1300 … memory 1320 may include a non-transitory computer-readable storage medium including instructions for applications or methods operated on device 1300, executable by the one or more processors of device 1300, Para. [0175] of FAN) comprising: determining a data structure representing chemical identities of atoms in a molecule (system may derive quantized information on the input based on its chemical formula, a chemical name, a high-resolution image of the crystal structure, a chemical drawing, data about the molecule, data about the atoms comprising the molecule, data about atom interactions, Para. [0077] of FAN; See also extracted features comprise features of a first atom and features of the interaction between the first atom and a second atom … the features of the first atom comprise one or more of an atom type, a radius of the atom, a number of rings in which the atom is included, a size of the ring in which the atom is included, whether the first atom is part of an aromatic ring, and the pairwise potential of the first atom, Para. [0017] of FAN; See also Table 1 provides a list of exemplary atom features that may comprise a feature vector for atom a, Paras. [0079] & [0080] of FAN; [at least a feature vector corresponds to a data structure]); training an energy potential model using a training set comprising the data structure, true conformations of the molecule, and false conformations of the molecule ((in step 1502, the processor obtains the training data … the processor may obtain correct ligand conformations based on ligand-receptor complexes in PDB files … the processor may also generate incorrect conformations used for the training … in step 1504, the processor extracts the features related to each conformation … in step 1506, the processor uses the extracted features to construct a feature vector for each conformation … in step 1508, the processor trains a classification model or a ranking model to search for the weight vector {right arrow over (W)}, Para. [0127] of FAN; See also the processor uses the classification model or ranking model trained in steps 1502-1508 to calculate the energy scores of the conformations, Para. [0129]; See also the scoring function H(distance, anglescore) may be used to calculate a pairwise potential energy score of the interaction of both atoms, Para. [0101] of FAN; See also after the distances and angle scores are determined, the atom pairwise potential energy may be determined, Para. [0099] of FAN; See also the machine-learning algorithm for determining the weight vector is trained on real-world protein structure data … the training of the machine-learning algorithm comprises determining a weight vector {right arrow over (W)}=(w1, w2, w3, . . . , wn) for the real-world protein structure data, where, when the feature vector for the correct ligand conformation equals (x1, x2, x3, . . . , xn) and the feature vector for an incorrect ligand conformation is (y1, y2, y3, . . . , yn), Para. [0021] of FAN; See also Paras. [0100], [0125] & [0126] of FAN); determining, using the trained energy potential model, a potential function associated with the molecule (method 1500 uses the feature vectors and weight vectors to construct implicit energy terms and use a machine-learning algorithm to derive the correct energy scoring functions, Para. [0130] of FAN; See also the scoring function H(distance, anglescore) may be used to calculate a pairwise potential energy score of the interaction of both atoms, Para. [0101] of FAN; See also ranking model trained in steps 1502-1508 to calculate the energy scores of the conformations, Para. [0129] of FAN; See also in step 1520, the processor outputs the energy scores, Para. [0129] of FAN; See also Para. [0100] of FAN) wherein determining the potential function associated with the molecule comprises: converting the data structure into a neural network based (a scoring function H(x) may be applied to a ligand atom and a receptor atom to obtain a score representing the interaction for that pair of atoms … all interactions between the ligand atom and potentially many receptor atoms are scored using the H(x) function … the sum of these scores is the pairwise potential of that ligand atom … the H(x) function may be developed by machine learning algorithms, such as the H(distance, anglescore) function described below, Para. [0092]; See also in the neural network model, the pairwise potential score of an atom may be used as one feature for each atom, as shown in the last row of Table 1, above, Para. [0093]; See also an atom's pairwise potential relates to forces between that atom and atoms of the receptor protein, such as van der Waals force and electrostatic force, Para. [0094]; See also para [0095], [0099]-[0102] and [0129]); and determining a conformation of the molecule based on potential function (the processor may predict the conformations of the ligand, or portion of the ligand, based on the energy scores associated with the conformations, Para. [0129] of FAN; See also method 1500 can accurately predict the ligand or ligand portion conformations, Para. [0130] of FAN).
Although (as discussed above) FAN discloses a van der Waals related potential energy function, FAN does not appear to explicitly disclose the potential function is a smooth function and wherein the neural network based smooth potential function comprises a polynomial-like function that transforms embeddings of atom types associated with the atoms through a logarithmic scale transformation and an exponential transformation.
ZHANG, however, is in the same field of complex protein design modeling (Para. [0010] of ZHANG) and discloses a Van der Waals related potential function is a smooth function (the methods provided herein include smoothing the input data, such as the data of a gridded box or local structural information … the smoothing of the input data may include applying three dimensional truncated Gaussian functions with standard deviation ax=σy=σz=r, wherein r is equal to ±0.5 angstrom of a Van der Waals radius of an atom in a voxel of the gridded box, Para. [0045] of ZHANG).
It would have been obvious for one of ordinary skill in the art before the effective filing date of the invention to modify the potential function having the Van der Waals function of FAN with the smooth Van der Waals function of ZHANG [to arrive at the claimed features] for the purpose of improving performance (Para. [0041] of ZHANG).
FAN as modified by ZHANG does not appear to explicitly disclose wherein the neural network based smooth potential function comprises a polynomial-like function that transforms embeddings of atom types associated with the atoms through logarithmic scale transformation and an exponential transformation.
WEI, however, is in the field of predicting molecules/biomolecules using machine learning/neural network algorithms (Abstract of WEI) and discloses wherein the neural network based (predicting structure-function relationships is associated with potential function, Para. [0004] of WEI; See also neural networks, Para. [0052] of WEI) comprises a polynomial-like function that transforms embeddings of atom types associated with the atoms (the physical features used as inputs to machine learning algorithms includes atomic type, atomic charge, Para. [0052] of WEI) through logarithmic scale transformation (data 1002, corresponding to log(P), and data 1004, corresponding to log(S) may be calculated for the biomolecular complex, Para. [0152] of WEI) and an exponential transformation (exponential function, Para. [0181] of WEI; See also exponential kernel function of Para. [0095], where X denotes a type of heavy atoms in the protein (Pro) and Y denotes a type of heavy atoms in the ligand (LIG), Para. [0096] of WEI).
It would have been obvious for one of ordinary skill in the art before the effective filing date of the invention to modify the smooth potential function of FAN as modified by ZHANG with the logarithmic and exponential transformations of WEI [to arrive at the claimed features] for the purpose of screening and predicting the activity of classes of biomolecules (Para. [0003] of WEI).
Claims 9 and 16 have substantially similar limitations as recited in claim 2; therefore, they are rejected under 35 U.S.C. § 102 for the same reasons.
Claims 10 and 17 have substantially similar limitations as recited in claim 3; therefore, they are rejected under 35 U.S.C. § 102 for the same reasons.
Claims 11 and 18 have substantially similar limitations as recited in claim 4; therefore, they are rejected under 35 U.S.C. § 102 for the same reasons.
Claims 13 and 20 have substantially similar limitations as recited in claim 6; therefore, they are rejected under 35 U.S.C. § 103 for the same reasons.
Claim 14 has substantially similar limitations as recited in claim 7; therefore, it is rejected under 35 U.S.C. § 102 for the same reasons.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure: CHAPMAN et al. (US 5,526,281 A1) published June 11, 1996. See, e.g., Col. 17, Lines 9-17 teaches “[c]onsider a molecule in a particular conformation at a particular location and orientation in space … [t]his mathematically defines the pose of the molecule … [f]rom each pose p of a molecule m, we generate a high-dimensional vector of features V(m,p) for purposes of activity prediction … each element of the feature vector characterizes a portion of the smoothed van der Wall's surface of the molecule” (emphasis added). Examiner’s Note: Chapman is cited because it appears to suggest that a van der Wall function/surface is smoothed (inherently) or is at least known to be smoothed.
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to JOHN P HOCKER whose telephone number is (571)272-0501. The examiner can normally be reached Monday-Friday 9:00 AM - 5:00 PM EST.
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, Rehana Perveen can be reached on (571)272-3676. 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.
/JOHN P HOCKER/Examiner, Art Unit 2189
JOHN P. HOCKER
Examiner
Art Unit 2189
/REHANA PERVEEN/Supervisory Patent Examiner, Art Unit 2189