CLIAMS:We claim:

1. A method for assessing input data comprising at least one of one or more chemical pathways, one or more reactions and one or more molecules with respect to chemical knowledgebase, comprising the steps of:
processing the input data;
extracting one or more features from the processed input data;
obtaining one or more coefficients of one or more association measures computed using the chemical knowledgebase, wherein the chemical knowledgebase comprises at least one of one or more chemical pathways, one or more reactions and one or more molecules;
analyzing the input data by scoring the one or more features extracted from the input data based on the one or more association measures; and
assessing the input data by at least one of a ranking and filtering based on the analysis.

2. The method as claimed in claim 1, further comprising computing one or more coefficients of one or more association measures using the chemical knowledgebase in absence of the computed one or more coefficients of one or more association measures.

3. The method as claimed in claim 2, wherein computing the one or more coefficients of the one or more association measures using the chemical knowledgebase comprises
processing the chemical knowledgebase;
extracting the one or more features from the processed chemical knowledgebase; and
computing the one or more coefficients of the one or more association measures based on the one or more extracted features.

4. The method as claimed in claim 3, further comprising storing at least one of the one or more features extracted from the chemical knowledgebase and the one or more coefficients of one or more association measures computed using the one or more features extracted from the chemical knowledgebase.

5. The method as claimed in claim 1, wherein processing at least one of the input data and the chemical knowledgebase comprises at least one of:
identifying one or more reaction transformations for one or more reactions and classifying one or more reactions based on one or more reaction transformation types;
identifying one or more molecular transformations for one or more molecules and classifying one or more molecules based on one or more molecular transformation types; and
identifying one or more molecular substructures comprising one or more molecules,
wherein the identified one or more transformation types for the one or more reactions and the one or more molecules, and one or more molecular substructures comprising the one or more molecules are tabulated.

6. The method as claimed in claim 5, wherein the molecular transformation comprises of one or more molecular substructures residing on a molecule and at least one of one or more bond changes, one or more bond rearrangements and one or more chemical state changes they undergo during a reaction process.

7. The method as claimed in claim 5, wherein the reaction transformation comprises of set of molecular transformations of one or more molecules participating in a reaction.

8. The method as claimed in claim 1, wherein the extracted feature comprises one of
association of the one or more molecular substructures;
association of the one or more molecular substructures classified with respect to one or more molecular transformation types; and
association of one or more reaction transformation types.

9. The method as claimed in claim 8, wherein the association of the one or more molecular substructures is derived from at least one of
occurrences of the one or more molecular substructures, and
co-occurrences of the one or more molecular substructures,
wherein the co-occurrence is recorded at least one of with and without relative distance between the molecular substructures.

10. The method as claimed in claim 8, wherein the association of the one or more reaction transformation types is derived from at least one of
occurrences of one or more reaction transformation types, and
co-occurrences of one or more reaction transformation types.

11. The method as claimed in claims 1 and 2, wherein computing the one or more coefficients for the one or more association measures is performed based on the extracted one or more features comprising the one or more associations of the one or more molecular substructures.

12. The method as claimed in claim 1 and 2, wherein computing the one or more coefficients for the one or more association measures is performed based on the extracted one or more features comprising the one or more associations of one or more molecular substructures classified with respect to one or more molecular transformation types.

13. The method as claimed in claim 1 and 2, wherein computing the one or more coefficients for the one or more association measures is performed based on the extracted one or more features comprising the one or more associations of one or more reaction transformation types.

14. The method as claimed in claim 1, wherein analyzing the input data by scoring the one or more features extracted from the input data based on at least one of the association measures comprises computing one or more composite scores for at least one of the one or more chemical pathways, one or more reactions and one or more molecules present in the input data based on the one or more scores of the one or more extracted features.

15. The method as claimed in claim 1, wherein the ranking comprises assigning ranks to at least one of the one of one or more chemical pathways, one or more reactions and one or more molecules in the input data according to one or more composite scores of the one or more chemical pathways, one or more reactions and one or more molecules in the input data.

16. The method as claimed in claim 1, wherein the filtering comprises selecting at least one of the one or more chemical pathways, one or more reactions and one or more molecules in the input data meeting defined threshold of one or more composite scores of the one or more chemical pathways, one or more reactions and one or more molecules in the input data.

17. A device for assessing input data comprising at least one of one or more chemical pathways, one or more reactions and one or more molecules with respect to chemical knowledgebase, comprising
a memory; and
one or more processors operatively coupled to the memory, the one or more processors are configured to perform the steps of:
processing the input data;
extracting one or more features from the processed input data;
obtaining one or more coefficients of one or more association measures computed using the chemical knowledgebase, wherein the chemical knowledgebase comprises at least one of one or more chemical pathways, one or more reactions and one or more molecules;
analyzing the input data by scoring the one or more features extracted from the input data based on the one or more association measures; and
assessing the input data by at least one of a ranking and filtering based on the analysis.

18. The device as claimed in claim 17, further comprising performing step of computing one or more coefficients of one or more association measures using the chemical knowledgebase in absence of the computed one or more coefficients of one or more association measures.

Dated this the 31st day of December 2014

Signature

SANTOSH VIKRAM SINGH
Patent Agent
Agent for the Applicant
,TagSPECI:FORM 2

THE PATENTS ACT, 1970
(39 OF 1970)
&
THE PATENTS RULES, 2003
COMPLETE SPECIFICATION
(Section 10 and Rule 13)

METHOD AND DEVICE FOR ASSESSING INPUT DATA WITH RESPECT TO CHEMICAL KNOWLEDGEBASE

SAMSUNG R&D INSTITUTE INDIA – BANGALORE PRIVATE LIMITED
# 2870, ORION Building, Bagmane Constellation Business Park,
Outer Ring Road, Doddanakundi Circle,
Marathahalli Post, Bangalore-560 037

An Indian Company

The following Specification particularly describes the invention and the manner in which it is to be performed

FIELD OF THE INVENTION
The present invention relates to assessing chemical pathway(s), reaction(s) and molecule(s) in terms of their feasibility. More particularly, it relates to assessing chemical pathway(s) or reaction(s) or molecule(s) or combination thereof with respect to a chemical knowledgebase so as to select suitable chemical pathway(s) or reaction(s) or molecule(s).
BACKGROUND OF THE INVENTION
Possibilities of synthesizing a molecule through organic design are numerous. As known in chemical retro-synthetic approaches a molecule can be synthesized from multiple precursors. As we increase the number of synthesis steps, possibilities of potential start compounds (precursors) increase exponentially. Figure 1 illustrates a schema for chemical retro-synthesis. Possible precursors for target molecule A are listed. In the figure A1, A2 and so on represent precursors that can form target molecule A in one step of chemical synthesis. Similarly A11, A12 and so on represent precursors that can form target molecule A in two steps via precursor A1. Pathways for synthesis of target molecule A are constructed using the listed precursors. With increase in number of chemical synthesis steps there is also an exponential rise in possible pathways for synthesizing the target molecule A. It is often challenging to assess experimentally all synthesis pathways reported using methods such as retro-synthesis.
Wherefore, it is required for the purpose of effective assessment of pathways to have an efficient method that would enable appropriate selection of the most relevant pathway(s).

SUMMARY OF THE INVENTION
The present invention provides a method and device for assessing input data with respect to chemical knowledgebase.
An embodiment of the present invention describes a method for assessing input data including chemical pathway(s) or reaction(s) or molecule(s) or combination thereof with respect to chemical knowledgebase including chemical pathway(s) or reaction(s) or molecule(s) or combination thereof. The method includes the steps of processing the input data, extracting feature(s) from the processed input data, obtaining coefficient(s) of association measure(s) computed using the chemical knowledgebase, analyzing the input data by scoring the feature(s) extracted from the input data based on the association measure(s), and assessing the input data by ranking and/or filtering based on the analysis.
A further embodiment of the present invention describes a method for assessing the input data with respect to the chemical knowledgebase where the computed coefficient(s) of association measure(s) are not available. The method includes computing coefficient(s) of the association measures using the chemical knowledgebase in absence of the computed coefficient(s) of association measure(s).
Another further embodiment of the present invention provides for a method for assessing the input data with respect to the chemical knowledgebase, where the feature(s) extracted from the chemical knowledgebase or the coefficient(s) of the association measure(s) computed using the one or more extracted features or combination thereof are stored for future use.
Another embodiment of the present invention describes a device for assessing input data including chemical pathway(s) or reaction(s) or molecule(s) or combination thereof with respect to chemical knowledgebase including chemical pathway(s) or reaction(s) or molecule(s) or combination thereof. The device includes a memory and processor(s) operatively coupled to the memory, where the processor(s) is/are configured to perform the steps of processing the input data, extracting feature(s) from the processed input data, obtaining coefficient(s) of association measure(s) computed using the chemical knowledgebase, analyzing the input data by scoring the feature(s) extracted from the input data based on the association measure(s), and assessing the input data by ranking and/or filtering based on the analysis.
Yet another further embodiment of the present invention provides a device for assessing the input data with respect to the chemical knowledgebase where the computed coefficient(s) of association measure(s) are not available. The device is further enabled to compute coefficient(s) of the association measure(s) using the chemical knowledgebase in absence of the computed coefficient(s) of association measure(s).
BRIEF DESCRIPTION OF FIGURES ACCOMPANYING THE PRESENT INVENTION
The aforementioned aspects and other features of the present invention will be explained in the following description, taken in conjunction with the accompanying drawings, wherein:
Figure 1 illustrates chemical retro-synthesis of chemical pathways involving multiple steps, according to prior art.
Figure 2 depicts a flow diagram for a method for assessing input data with respect to chemical knowledgebase, according to an embodiment.
Figure 3 depicts a flow diagram for a method for assessing input data with respect to chemical knowledgebase, where the computed coefficient(s) of the association measure(s) is/are not available, according to another embodiment.
Figure 4 depicts a flow diagram for a method of computing coefficients of association measure(s) based on features including associations of reaction transformation types, according another embodiment.
Figure 5 is a block level diagram of a device for assessing the input data with respect to the chemical knowledgebase, according to yet another embodiment.
Figure 6 is a block level diagram of a device for assessing the input data with respect to the chemical knowledgebase, where the computed coefficient(s) of the association measure(s) is/are not available, according to yet another further embodiment.
The drawings described herein are for illustration purposes only and are not intended to limit the scope of the present disclosure in any way.

DETAILED DESCRIPTION OF THE INVENTION
The embodiments of the present invention will now be described in detail with reference to the accompanying drawings. However, the present invention is not limited to the embodiments. The present invention can be modified in various forms. Thus, the embodiments of the present invention are only provided to explain more clearly the present invention to the ordinarily skilled in the art of the present invention. In the accompanying drawings, like reference numerals are used to indicate like components.
The specification may refer to “an”, “one” or “some” embodiment(s) in several locations. This does not necessarily imply that each such reference is to the same embodiment(s), or that the feature only applies to a single embodiment. Single features of different embodiments may also be combined to provide other embodiments.
As used herein, the singular forms “a”, “an” and “the” are intended to include the plural forms as well, unless expressly stated otherwise. It will be further understood that the terms “includes”, “comprises”, “including” and/or “comprising” when used in this specification, specify the presence of stated features, integers, steps, operations, elements and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof. As used herein, the term “and/or” includes any and all combinations and arrangements of one or more of the associated listed items.
Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure pertains. It will be further understood that terms, such as those defined in commonly used dictionaries, should be interpreted as having a meaning that is consistent with their meaning in the context of the relevant art and will not be interpreted in an idealized or overly formal sense unless expressly so defined herein.
The present invention provides method and device for assessing input data including chemical pathway(s) and/or reaction(s) and/or molecule(s) by ranking and/or filtering the input data. The assessment is performed with reference to a chemical knowledgebase. The said chemical knowledgebase includes chemical pathway(s) and/or reaction(s) and/or molecule(s). In brief, a set of features (representing structural and chemical properties) are extracted from the input data and assessed based on the similar features extracted from chemical knowledgebase to score chemical pathway(s) and/or reaction(s) and/or molecule(s) present in the input data.
The chemical knowledgebase contains known chemical pathway(s) and/or reaction(s) and/or molecule(s). Aim of the assessment method is to prioritize features extracted from the input data based on the propensity of these features in the chemical knowledgebase. It facilitates selection of pathway(s) and/or reaction(s) and/or molecule(s) present in the input data that represent transformation(s) and/or structure(s) that are likely to be more feasible.
Figure 2 provides the steps of the present method of assessing the input data with respect to the chemical knowledgebase, according to one embodiment. The input data is processed to identify structural and/or chemical attribute(s) such as, but not limited to, reaction and molecular transformation(s), transformation type(s) and molecular substructure(s) at step 202. The feature(s) from the processed input data is/are extracted at step 204. Coefficient(s) of association measure(s), computed using the chemical knowledgebase, is/are obtained at step 206. The step 206 is performed independent of steps 202 and 204. The input data is analyzed by scoring feature(s) extracted from the input data based on the association measure(s) at step 208. Finally, the input data is assessed based on the analysis performed in the previous step (208) at step 210. The assessment involves ranking and/or filtering the chemical pathway(s) and/or reaction(s) and/or molecule(s) in the input data.

Processing the input data (step 202)
The chemical pathway(s) and/or reaction(s) and/or molecule(s) in the input data is/are processed to identify and tabulate structural and/or chemical attribute(s). The input data is processed to identify reaction transformation(s) and subsequently reaction transformation type(s) for the reaction(s), and/or molecular transformation(s) and subsequently molecular transformation type(s) for the molecule(s) participating in the reaction(s) and/or substructure(s) contained in the molecule(s). The processed information created and/or stored into table(s) at this stage is used for extracting association information in next step (step 204).
The molecular substructure in the context of the present invention represents a connected structure whose atoms are subset of that of the molecule. The molecular transformation in the context of the present invention comprises of substructure(s) residing on a molecule and bond change(s) and/or bond rearrangement(s) and/or chemical state change(s) it/they undergo during a reaction process. Set of molecular transformations with common characteristics define a molecular transformation type. The molecule(s) in reaction(s) is/are classified with respect to molecular transformation type(s).
The reaction transformation in the context of the present invention comprises set of molecular transformations of the molecule(s) participating in a reaction. Further, the set of reaction transformations with common characteristics define the reaction transformation type. The reaction(s) in the input data is/are classified with respect to reaction transformation type(s).
Extraction of the features from the processed input data (step 204)
The structural and/or chemical attribute(s) identified by processing the input data in step 202 is/are used for extraction of feature(s). The extracted feature(s) include association(s) of the molecular substructure(s) and/or association(s) of the molecular substructure(s) classified with respect to the molecular transformation type(s) and/or association(s) of the reaction transformation type(s).
The feature comprising of association of the molecular substructure(s) relates to the propensity of existence and/or coexistence of structural features on a molecule. It is derived from occurrence(s) and/or co-occurrence(s) of the molecular substructure(s) of the molecule(s) present in the input data. The co-occurrence(s) is/are recorded with or without relative distance(s) between the molecular substructures. Furthermore, the association(s) of the molecular substructure(s) is/are also classified with respect to molecular transformation type(s). The feature comprising of association of the reaction transformation type(s) relates to the propensity of existence and/or coexistence of reaction transformation type(s) in a pathway of reactions. It is derived from occurrences and/or co-occurrences of the reaction transformation type(s) of the reactions forming pathway(s).
Obtaining the coefficient(s) of association measure(s) computed using the chemical knowledgebase (step 206)
The coefficient(s) of association measure(s) represent aggregated structural and/or chemical attributes of the chemical knowledgebase. The coefficient(s) is/are computed based on the feature(s) extracted from the chemical knowledgebase. The steps for computing the coefficient(s) of association measure(s) are detailed in later part of the specification (steps 302, 304 and 306 of Figure 3). The coefficient(s) is/are retrieved at step 206 and is/are used for analysis of input data through association measure(s) (step 208).

Analyzing the input data by scoring feature(s) extracted from the input data based on the association measure(s) (step 208)
The feature(s) extracted from the input data (step 204) and the coefficient(s) of association measure(s) determined based on the chemical knowledgebase (step 206) are together used for the analysis of the input data. The feature(s) extracted from the input data is/are scored using the association measure(s). Further, composite score(s) for the chemical pathway(s) and/or reaction(s) and/or molecule(s) in input data are computed based on score(s) of the corresponding extracted feature(s).
Assessing the input data (step 210)
The assessment of the input data is performed by ranking and/or filtering the chemical pathway(s) and/or reaction(s) and/or molecule(s) based on the score(s) assigned in step 208.
The chemical pathway(s) and/or reaction(s) and/or molecule(s) present in the input data are ranked based on the composite scores computed for the chemical pathway(s) and/or reaction(s) and/or molecule(s) in the input data.
Similarly, the filtering is performed by selecting chemical pathway(s) and/or reaction(s) and/or molecule(s) in the input data having the assigned composite score(s) meeting defined thresholds.
Computing the coefficients of the association measure(s) based on the features extracted from the chemical knowledgebase
Figure 3 illustrates another embodiment of the present invention where the coefficient(s) of association measure(s) is/are not available and therefore the same is/are computed using the chemical knowledgebase. The steps 202, 204, 206, 208 and 210 are already described above (Figure 2). Further, the steps 302 and 304 are performed on the chemical knowledgebase in the same manner as the steps 202 and 204 detailed for the input data.
The coefficient(s) of association measure(s) is/are computed using the chemical knowledgebase at step 306. The coefficient(s) is/are computed based on feature(s) extracted from the chemical knowledgebase using methods such as, but not limiting to, conditional probability, joint probability, and Bayesian statistics.
In the present embodiment, the coefficient(s) of association measure(s) is/are computed using the feature(s) comprising of association(s) of molecular substructure(s) and/or feature(s) comprising of associations(s) of molecular substructure(s) classified with respect to the molecular transformation type(s) and/or feature(s) comprising of association(s) of the reaction transformation type(s).
Figure 4 describes an exemplary embodiment to compute coefficient(s) of association measure(s) based on feature(s) comprising of association(s) of reaction transformation type(s). First, the reaction(s) contained in the chemical knowledgebase is/are processed to identify reaction transformation(s) and further classified into reaction transformation type(s). The processing is performed as described in step 302. The figure describes computation of two sets of coefficients of association measure(s). The computation of the first set is performed based on the occurrences of reaction transformation type(s) and is defined as the probability of occurrence of a specified reaction transformation type. The computation of second set is performed based on the co-occurrences of a reaction transformation type pair in a reaction pathway. It is defined as the joint probability between the reaction transformation types of two reactions in a reaction pathway.
In another embodiment of the present invention, the feature(s) extracted from the chemical knowledgebase at step 304 and/or the coefficient(s) of association measure(s) computed at step 306 are stored in memory device(s) for future use, enabling user(s) to retrieve the stored data as and when desired.
An exemplary embodiment that assesses an input data containing chemical pathways is described here. As the first step, the input data is processed to identify reaction transformations and subsequently reaction transformation types for each reaction as per the step 202 of Figure 2. Next, the features that comprise of the occurrences and co-occurrences of reaction transformation types within individual pathways are extracted (step 204). Coefficients of association measures computed using features extracted from the chemical knowledgebase comprising the associations of reaction transformation types are retrieved (step 206). The computation of these coefficients is previously described (Figure 4). Features corresponding to each pair of reaction transformation types of reactions co-occurring in a pathway are scored by association measure based on coefficients of joint probability of co-occurrence (Figure 4). To score individual pathways, a composite score based on the average joint probability of co-occurrence is used. Finally, based on the scores the input chemical pathways are assessed.
The present invention also provides a device for assessing the input data with respect to the chemical knowledgebase. Figure 5 is a block level diagram of the device in accordance with an embodiment of the present invention. The device is configured to assess the input data with respect to the chemical knowledgebase.
The device 500 includes processor(s) 504, and memory 502 coupled to the processor(s) 504.
The processor(s) 504, as used herein, means any type of computational circuit, such as, but not limited to, a microprocessor, a microcontroller, a complex instruction set computing microprocessor, a reduced instruction set computing microprocessor, a very long instruction word microprocessor, an explicitly parallel instruction computing microprocessor, a digital signal processor, or any other type of processing circuit, or a combination thereof.
The memory 502 includes a plurality of modules stored in the form of executable program which instructs the processor 504 to perform the method steps illustrated in Figure 2. The memory 502 has following modules: processing module (for input data and/or chemical knowledgebase) 508, feature extraction module (from processed input data and/or chemical knowledgebase) 510, the coefficient(s) of the association measure(s) obtaining module 512, analyzing module 514, and assessment module 516.
Computer memory elements may include any suitable memory device(s) for storing data and executable program, such as read only memory, random access memory, erasable programmable read only memory, electrically erasable programmable read only memory, hard drive, removable media drive for handling memory cards and the like. Embodiments of the present subject matter may be implemented in conjunction with program modules, including functions, procedures, data structures, and application programs, for performing tasks, or defining abstract data types or low-level hardware contexts. Executable program stored on any of the above-mentioned storage media may be executable by the processor(s) 504.
The processing module 508 instructs the processor(s) 504 to perform the step 202 (Figure 2) and/or step 302 (Figure 3).
The feature extraction module 510 instructs the processor(s) 504 to perform the step 204 (Figure 2) and/or step 304 (Figure 3).
The coefficient(s) of the association measure(s) obtaining module 512 instructs the processor(s) 504 to perform the step 206 (Figure 2).
The analyzing module 514 instructs the processor(s) 504 to perform the step 208 (Figure 2).
The assessment module 516 instructs the processor(s) 504 to perform the step 210 (Figure 2).
Figure 6 is a block level diagram of the device for assessing the input data with respect to the knowledgebase and computing the coefficient(s) of the association measure(s) in accordance with another embodiment of the present invention. The device is configured to perform the assessment of the input data with respect to the knowledgebase in the same way as detailed for Figure 5. However, the additional feature of computing the coefficient(s) of the association measure(s) in absence of the computed coefficient(s) of the association measure(s) is performed by coefficient computation module 602, which instructs the processor(s) to perform steps 302, 304 and 306 (Figure 3).
The present embodiments have been described with reference to specific example embodiments; it will be evident that various modifications and changes may be made to these embodiments without departing from the broader spirit and scope of the various embodiments. Furthermore, the various devices, modules, and the like described herein may be enabled and operated using hardware circuitry, for example, complementary metal oxide semiconductor based logic circuitry, firmware, software and/or any combination of hardware, firmware, and/or software embodied in a machine readable medium.