Claims:We Claim:
1. A method for mutation prioritization for personalized therapy of one or more patients, comprising:
acquiring mutation information of one or more patients to be treated, wherein the mutation information comprises information associated with at least one of one or more diseases, one or more genes, and one or more alterations of genomic DNA;

mapping the acquired mutation information with disease knowledgebase, wherein the disease knowledgebase comprises:
data of associations of one or more data points indicative of at least one of the one or more diseases, one or more genes, and one or more alterations of genomic DNA derived from one or more knowledge sources falling under one or more categories with one or more data points indicative of parameters of clinical relevance from one or more knowledge sources falling under the one or more categories,
the one or more categories of the one or more knowledge sources are ranked based on one of user input and in any pre-defined priority, and

the data of associations of the one or more data points in the one or more categories of the one or more knowledge sources are classified into a plurality of classes predefined for each of the category of the one or more knowledge sources with pre-assigned precedence;

identifying at least one of the one of one or more diseases, one or more genes, and one or more alterations of genomic DNA provided in the acquired mutation information of the one or more patients mapped with the data of associations of the one or more data points, thereby forming mapped mutation information;

generating one or more frequency tables for the mapped mutation information category-wise and subsequently respective class-wise; and

prioritizing the mapped mutation information in the one or more frequency tables based on a prioritization scheme.

2. The method as claimed in claim 1, wherein the one or more categories of the one or more knowledge sources are Clinical Trials, Therapies and Publications.

3. The method as claimed in claim 1, wherein each of the one or more frequency tables comprises
a plurality of columns, where each column is populated with number of occurrences of the data of associations of the one or more data points belonging to a particular class of a category of the one or more knowledge sources, and
a plurality of rows, where each row is populated with the number of occurrences of the data of associations of the one or more data points linked with a particular alteration of genomic DNA.

4. The method as claimed in claim 1, wherein the prioritization scheme provides for
filtering the frequency table based on one category selected from the one or more categories of the one or more knowledge sources;
populating the filtered frequency table with data points of the one or more categories of the one or more knowledge sources, not selected in previous step, linked with one or more data points associated with the selected category of the one or more knowledge sources; and
sorting the frequency table based on number of occurrence of the one or more data points vis a vis ranking of the one or more categories of the one or more knowledge sources and pre-assigned precedence of the one or more classes of the association of the data points.

5. The method as claimed in claim 1, wherein the prioritization scheme provides for
arranging linked one or more data points in the frequency table category -wise and subsequently respective class-wise; and
sorting the frequency table for the mapped mutation information based on number of occurrence of the one or more data points vis a vis ranking of the one or more categories of the one or more knowledge sources and pre-assigned precedence of the one or more classes of the association of the data points.

6. The method as claimed in claims 4 and 5, wherein one or more categories of the one or more knowledge sources are ranked based on one of user input or and in a pre-defined priority.

7. A device for mutation prioritization for personalized therapy of one or more patients, 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:
acquiring mutation information of one or more patients to be treated, wherein the mutation information comprises information associated with at least one of one or more alterations of genomic DNA, one or more genes and one or more diseases;
mapping the acquired mutation information with disease knowledgebase, wherein the disease knowledgebase comprises
data of associations of one or more data points indicative of at least one of the one or more alterations of genomic DNA, the one or more genes, and the one or more diseases derived from one or more knowledge sources falling under one or more categories with one or more data points indicative of parameters of clinical relevance from one or more knowledge sources falling under the one or more categories,
the one or more categories of the one or more knowledge sources are ranked based on one of user input and in any pre-defined priority, and
the data of associations of the one or more data points in the one or more categories of the one or more knowledge sources are classified into a plurality of classes predefined for each of the category of the one or more knowledge sources with pre-assigned precedence;
identifying the one or more alterations of genomic DNA provided in the acquired mutation information of the one or more patients mapped with the data of associations of one or more data points, thereby forming mapped mutation information;
generating one or more frequency tables for the mapped mutation information category-wise and subsequently respective class-wise; and
prioritizing the mutation information in the frequency table by sorting based on a prioritization scheme.

8. The method as claimed in claim 7, wherein the one or more categories of the one or more knowledge sources are Clinical Trials, Therapies and Publications.

9. A method for generating a disease knowledgebase, comprising
obtaining information pertaining to at least one of one or more diseases, one or more genes, and one or more alterations of genomic DNA and one or more parameters of clinical relevance from one or more knowledge sources falling under one or more categories;
curating the obtained information to extract one or more data points indicative of at least one of the one or more diseases, one or more genes, and one or more alterations of genomic DNA and one or more parameters of clinical relevance from the one or more knowledge sources falling under one or more categories;
identifying association of the one or more data points indicative of at least one of the one or more alterations of genomic DNA, the one or more genes, and the one or more diseases with the one or more data points indicative of parameters of clinical relevance, thereby forming data of associations of the one or more data points;
classifying the association of the one or more data points for their linkage with one or more diseases, one or more genes, and one or more alterations of genomic DNA in the one or more categories of the one or more knowledge sources into one or more classes,
wherein the one or more classes are assigned to each of the data point for its linkage with the one or more diseases, one or more genes, and one or more alterations of genomic DNA; and
generating the disease knowledgebase based on the classified data of associations of one or more data points in the one or more categories.

10. The method as claimed in claim 9, wherein the disease knowledgebase comprises:
category-wise and subsequently respective class-wise arrangement of data of associations of one or more data points indicative of the one or more alterations of genomic DNA, the one or more genes, and the one or more diseases derived from one or more knowledge sources with one or more data points indicative of parameters of clinical relevance from one or more knowledge sources.

11. The method as claimed in claims 9 and 10, wherein the plurality of the classes of each of the one or more categories have pre-assigned precedence.

12. The method as claimed in claims 9 and 10, wherein the one or more categories are ranked based on one of user input and in any pre-defined priority.

13. A disease knowledgebase generated through the method of claim 9.

14. A device for generating a disease 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:
obtaining information pertaining to at least one of one or more diseases, one or more genes, and one or more alterations of genomic DNA and one or more parameters of clinical relevance from one or more knowledge sources falling under one or more categories;
curating the obtained information to extract one or more data points indicative of at least one of the one or more alterations of genomic DNA, one or more genes, one or more diseases and one or more parameters of clinical relevance from one or more knowledge sources falling under the one or more categories;
identifying associations of the one or more data points indicative of at least one of the one or more alterations of genomic DNA, the one or more genes, and the one or more diseases with the one or more data points indicative of parameters of clinical relevance, thereby forming data of associations of the one or more data points in the one or more categories of the one or more knowledge sources
wherein one or more classes are assigned to a data point for its linkage with one or more data points from a category of knowledge source;
classifying the association of the one or more data points for their linkage with one or more diseases, one or more genes, and one or more alterations of genomic DNA in the one or more categories of the one or more knowledge sources into a plurality of classes; and
generating the disease knowledgebase based on the classified one or more data points in the one or more categories.

15. The device as claimed in claim 14, wherein the one or more categories of the one or more knowledge sources are Clinical Trials, Therapies and Publications.

Dated this the 12th day of August 2015

Signature

KEERTHI JS
Patent Agent
Agent for the Applicant , Description:FORM 2
THE PATENTS ACT, 1970
[39 of 1970]
&
THE PATENTS RULES, 2003
COMPLETE SPECIFICATION
(Section 10; Rule 13)

METHOD AND DEVICE FOR MUTATION PRIORITIZATION FOR PERSONALIZED THERAPY OF ONE OR MORE PATIENTS

SAMSUNG R&D INSTITUTE INDIA – BANGALORE PRIVATE LIMITED,
#2870, Orion Building, Bagmane Constellation Business Park,
Outer Ring Road, Doddanekundi Circle,
Marathahalli Post, Bangalore – 560037,
Karnataka, India.
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 generally relates to the field of clinical genomics, and more particularly relates to a method and device for mutation prioritization for therapy personalization.

BACKGROUND OF THE INVENTION

Next generation sequencing (NGS) based personalized diagnostics holds great potential as a valuable tool for clinical decision making in healthcare. Its market is currently estimated to be $393m and is expected to grow at a fast pace in coming years. The emphasis of personalized diagnostics has been on genetic disorders, especially on cancer. With ~1m cancer cases being diagnosed annually in US alone and poor response rates (25%) to generic treatments, NGS-based diagnostics can have a significant impact on prescribing effective treatment to an individual.

Such personalization is based on the set of mutations obtained from analyzing an individual’s DNA data through a NGS analysis pipeline. These mutations that characterize the individual’s disease help clinicians in tailoring therapy to it. Although very promising, several challenges need to be addressed before the mutation data becomes useful for therapy personalization. A key issue is to organize often unstructured data such as mutation-disease association or cancer-specific targeted therapy information into a structured format for automated analysis. Systematic organization of relevant information plays a vital role in data-driven approaches that leverage existing knowledge to recommend therapies options to clinicians and researchers.
Existing approaches often focus on therapies and on prioritizing them. The evidence used in these approaches is extracted and curated from similar knowledge as used in the present specification. These can include among other sources, clinical trials and publications backing up a particular therapy. In addition biomarker data is also used. In some other approaches, mutations are classified using evidences from sources such as publications into different classes based on the publication evidence.

Thus, there exists a need for a method that considers knowledgebase specified by user, obtains mutations of patient’s, prioritize mutations based on data gathered as knowledgebase, assists in deciding treatment options based on the information gathered regarding one or more mutations in question.

SUMMARY OF THE INVENTION

The various embodiments herein disclose a method and device for mutation prioritization for personalized therapy of patient(s).

An embodiment of the present invention provides a method for mutation prioritization for personalized therapy of patient(s). The method steps include acquiring mutation information of patient(s) to be treated, wherein the mutation information comprises information associated with alteration(s) of genomic DNA and/or gene(s) and disease(s); mapping the acquired mutation information with disease knowledgebase; identifying at least one of the disease(s), gene(s), and the alteration(s) of genomic DNA provided in the acquired mutation information of the patient(s) mapped with the data of associations of the data point(s), thereby forming mapped mutation information; generating frequency table(s) for the mapped mutation information category (of the knowledge source)-wise and subsequently respective class-wise; and prioritizing the mapped mutation information in the frequency table(s) based on a prioritization scheme. The disease knowledgebase includes data of associations of data point(s) indicative of the alteration(s) of genomic DNA, the gene(s), and the disease(s) derived from knowledge source(s) falling under one or more categories with data point(s) indicative of parameters of clinical relevance from knowledge source(s) falling under one or more categories. The category(ies) of the knowledge sources are ranked based on one of user input and in any pre-defined priority. The data point(s) in the category of the knowledge source(s) are classified into a plurality of classes predefined for each of the category of the knowledge source(s) with pre-assigned precedence.
Another embodiment of the present invention describes a device for mutation prioritization for personalized therapy of patient(s). The device includes a memory; and processor(s) operatively coupled to the memory. The processor(s) is/are configured to perform the steps of acquiring mutation information of patient(s) to be treated; mapping the acquired mutation information with disease knowledgebase; identifying the alteration(s) of genomic DNA provided in the acquired mutation information of the patient(s)` mapped with the data of associations of data point(s), thereby forming mapped mutation information; generating frequency table(s) for the mapped mutation information category -wise and subsequently respective class-wise; and prioritizing the mutation information in the frequency table by sorting based on a prioritization scheme.
Further embodiment of the present invention provides a method for generating a disease knowledgebase. The method steps include obtaining information pertaining to alteration(s) of genomic DNA, gene(s), disease(s) and parameter(s) of clinical relevance from knowledge source(s) falling under one or more categories; curating the obtained information to extract data point(s) indicative of the alteration(s) of genomic DNA/gene(s)/disease(s) and parameter(s) of clinical relevance from knowledge sources falling under one or more categories; identifying data of associations of data point(s) indicative of the alteration(s) of genomic DNA, the gene(s), and the disease(s) with data point(s) indicative of parameters of clinical relevance; classifying the association of the data point(s) for their linkage with disease(s), gene(s), and alteration(s) of genomic DNA in the category(ies) of the knowledge source(s) into class(es); and generating the disease knowledgebase based on the classified association of data point(s) present in the category of the knowledge sources.

Yet another embodiment of the present invention provides a device for generating a disease knowledgebase. The device includes a memory and processor(s) operatively coupled to the memory. The processor(s) are configured to perform the steps of obtaining information pertaining to alteration(s) of genomic DNA/gene(s)/disease(s) and parameter(s) of clinical relevance from knowledge source(s) falling under one or more categories; curating the obtained information to extract data point(s) indicative of the alteration(s) of genomic DNA, gene(s), disease(s) and parameter(s) of clinical relevance from knowledge sources falling under one or more categories; identifying data of associations of data point(s) indicative of the alteration(s) of genomic DNA, the gene(s), and the disease(s) with data point(s) indicative of parameters of clinical relevance; classifying the association of the data point(s) for their linkage with disease(s), gene(s), and alteration(s) of genomic DNA in the category(ies) of the knowledge source(s) into class(es); and generating the disease knowledgebase based on the classified association data point(s) present in the category of the knowledge sources.

BRIEF DESCRIPTION OF THE DRAWINGS

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 is a schematic flow representation of method for mutation prioritization for personalized therapy of one or more patients, according to one embodiment.

Figure 2 is a schematic flow representation of obtaining patient’s variation data (e.g. VCF file) and generating frequency table, according to an embodiment of the present invention.

Figure 3 is a schematic diagram illustrating generation of frequency table from a patient’s variation data (e.g. VCF file) where the disease knowledgebase included knowledge sources related to clinical trials and therapy, according to an exemplary embodiment of the present invention.

Figures 4a and 4b are schematic diagrams illustrating two prioritization schemes, according to an embodiment of the present invention.

Figure 5 is schematic diagram illustrating sorting of mutations based on higher clinical trial evidence value than therapy evidence value, according to an embodiment of the present invention.

Figure 6 is schematic diagram illustrating sorting of mutations based on higher therapy evidence value than clinical trial evidence value, according to an embodiment of the present invention.

Figure 7 is block level diagram illustrating a device for mutation prioritization for personalized therapy of one or more patients, according to an embodiment of the present invention.

Figure 8 is a schematic flow representation of a method for generating a disease knowledgebase, according to one embodiment.

Figure 9 is a schematic presentation of step of obtaining/aggregating the data from the plurality of knowledge sources, aggregating, and curating the data to obtain data points and classifying the data points, according to one embodiment.

Figure 10 is block level diagram illustrating a device for generating a disease knowledgebase, according to one embodiment.

DETAILED DESCRIPTION OF THE INVENTION

The present invention provides a method and device for mutation prioritization for personalized therapy of patients to be treated. In the following detailed description of the embodiments of the invention, reference is made to the accompanying drawings that form a part hereof, and in which are shown by way of illustration specific embodiments in which the invention may be practiced. These embodiments are described in sufficient detail to enable those skilled in the art to practice the invention, and it is to be understood that other embodiments may be utilized and that changes may be made without departing from the scope of the present invention. The following detailed description is, therefore, not to be taken in a limiting sense, and the scope of the present invention is defined only by the appended claims.

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 term “alteration in genomic DNA” as used herein includes all sorts of mutations such as, but not limited to, substitutions, insertions, deletions, frameshifts and likes. The term “alteration of genomic DNA” and “mutation” are used synonymously in the context of the invention.

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. For example, the various electrical structure and methods may be embodied using transistors, logic gates, and electrical circuits, such as application specific integrated circuit.

Although the embodiments herein are described with various specific embodiments, it will be obvious for a person skilled in the art to practice the invention with modifications. However, all such modifications are deemed to be within the scope of the claims. It is also to be understood that the following claims are intended to cover all of the generic and specific features of the embodiments described herein and all the statements of the scope of the embodiments which as a matter of language might be said to fall there between.
Mutation prioritization for personalized therapy of patient(s)

The present invention provides for a method and device for mutation prioritization which is helpful in application of personalized therapy to patient(s). In other words, the present invention enables the doctors to personalize the therapy to be given to a patient. Mutation map of a patient suffering with cancer generally indicates tens to thousands of alterations of genomic DNA. During the process of treatment of the said cancer patient it is challenging to identify alterations of genomic DNA that are useful for targeted personalized therapy. The present invention addresses the problem of identifying the most relevant alteration(s) of genomic DNA. Thereby, it provides a decision support system that sorts alterations of genomic DNA in the patient based on supporting evidence from data gathered from various categories of knowledge sources such as, but not limited to, clinical trials, therapy linkages and publications. The knowledge source(s) for clinical trial include, but not limited to, ClinicalTrials.gov; for Therapy it could be including, but not limited to, Drugs@FDA® and DrugBank®; and for publication including, but not limited to, PubMed® etc. The sorted alterations of genomic DNA in the patient gives a fair idea to doctor/caregiver/researcher to identify the most relevant mutations that are helpful to clinicians/researchers in making informed decisions.
The flow diagram as given in Figure 1 provides the detailed steps of the present method, according to one embodiment.
Mutation information of patient(s) to be treated is acquired at step 102. The mutation information includes information associated with disease(s) and/or gene(s) and/or alteration(s) of genomic DNA (refer to Figure 2). Generation of mutation information of patient is performed by methods known in the art. For example, the patient’s genome is sequenced and analyzed to identify relevant mutations and generating patient variation data (e.g. VCF file) containing the identified mutations using a standard NGS pipeline.
The acquired mutation information is mapped with disease knowledgebase at step 104 so as to find out information related to the acquired mutation information available in the disease knowledgebase. The mapping helps in finding the relevant information available in the disease knowledgebase regarding the acquired mutation information.
The disease knowledgebase is created beforehand by gathering data from the one or more knowledge sources falling under the one or more categories. The disease knowledgebase comprises data of associations or linkages of data point(s) indicative of the alteration(s) of genomic DNA, gene(s), and disease(s) derived from the category(ies) of the knowledge source(s) with data point(s) indicative of parameters of clinical relevance from the category(ies) of the knowledge source(s). The parameters of clinical relevance used herein include disease stage, disease type and sub-type.
The category(ies) of the knowledge source(s) are ranked based either on user input or in any pre-defined priority. Further, the data of associations of the data point(s) in the category(ies) of the knowledge source(s) are classified into class(es) predefined for each of the category of the knowledge source(s) with pre-assigned precedence.
Creation of the disease knowledgebase involves curating the gathered data for specific information of ( also represented as linkages or triad), classifying the every triad (linked data) identified during curation, and identifying points from the knowledge source(s) linked to the triad in the knowledge base (detailed in the later part of the specification). As mentioned, the categories of the knowledge sources (clinical trials, therapy linkages and publications) are ranked either based on one of user input or in any pre-defined priority. Therefore, as per ranking assigned to the categories of the knowledge sources, the specific data points from the knowledge source are displayed/presented in the disease knowledgebase. For example, often times, doctors and caregivers are more interested in specific mutations that may be present in a patient. Treatment options are often decided based on these mutations. Therefore, the user preference could be Therapies>Clinical Trials>Publications.
The data point(s) falling under any of the three categories (clinical trials, therapies, and publications) of the knowledge source(s) is/are further classified into a plurality of classes predefined for each of the category of the knowledge source(s), where the classes have pre-assigned precedence. For example, where
(A) Clinical trials are selected as one of the categories of the knowledge sources, and associations are identified with a set (data point).The clinical trial is assigned a specific class for a given based on its relevance to that . Additionally, the same clinical trial could also be associated to a different set and is classified based on its relevance to the said (different) set. Further, a set could be associated with multiple data points from a given category of the knowledge source. Further, a set could be associated with multiple clinical trials. Table 1 illustrates a manner of representation of the data of associations for clinical trials forming a disease knowledgebase or part of the disease knowledgebase. Therefore, the classification is always relative to and same holds true for other classification of other categories of knowledge sources. ClinicalTrials.gov provides a unique id/registry number for each clinical trial called the NCTID which is an 8 digit number preceded by the letters ‘NCT’. However, Table 1 is provided with dummy unique id/registry numbers for the sake of understanding the representation of data and classification. Every class signifies extent of relevance of a given gene and mutation to a clinical trial for a disease. Classes are labelled as CT0, CT1, CT2 and CT3, where the CT0 signifies most relevant class and CT3 least relevant class.
Tumor (Disease) Gene Mutation NCTID Class
Breast ERBB2 S310F NCT01827267 CT1
Breast ERBB2 S310F NCT01670877 CT1
Breast ERBB2 S310F NCT01953926 CT1
Breast ERBB2 S310F NCT00730925 CT1
Breast ERBB2 S310F NCT01288261 CT2
Breast ERBB2 S310F NCT00580333 CT2
Breast ERBB2 S310F NCT01271725 CT3
Breast ERBB2 S310F NCT01441596 CT3
Breast ERBB2 S310F NCT01531764 CT3

Table 1
(B) Therapies are selected as the category of the knowledge source, every association of to drug or drug action through for example, curation of published studies. Classification of therapies is performed based on the patient mutation and disease information. Classes are labelled as T0, T1, T2 and T3, where T0 signifies most relevant class and T3 least relevant class.
(C) Publications are selected as category of the knowledge source, for a given , relevant publications are identified. The identified {, publication} sets are classified into relevant classes based on the clinical, pre-clinical status of the studies discussed in the publication. Classes are labelled as P0, P1, P2 and P3, where P0 signifies most relevant class and P3 least relevant class. A low Class number indicates higher relevance and vice versa, i.e., P0 has high relevance than P3 and is likewise applicable for the clinical trial (CT) and therapy (T).
The alteration(s) of genomic DNA provided in the acquired mutation information of the patient(s), which were mapped with the data of associations or linkages of the data point(s) are identified at step 106. The output of mapping step is mapped mutation information which indicates relevant data points present in the knowledgebase with respect to the acquired mutation information (Figures 2 and 3).
Frequency table(s) for the mapped mutation information is generated category (of the knowledge source) -wise and subsequently respective class-wise at step 108. The frequency table includes a plurality of columns, where each column is populated with number of occurrences of the data of associations or linkages of the data point(s) belonging to a particular class of a category of the knowledge source(s), and a plurality of rows, where each row is populated with the number of occurrences of data of associations or linkages of the data point(s) linked with a particular alteration of genomic DNA. Figure 2 illustrates schematic flow representation of obtaining patient’s variation data (such as including, but not limited to, VCF file) and generating frequency table, according to an embodiment of the present invention. The numeric values populating the columns of the knowledge sources and its subsequent classes in the frequency table indicate the number of occurrences of the mapped mutation information found that particular class of the knowledge source. For example, for the gene-mutation ATP6AP2-K205E the column CT1 under clinical trials shows value ‘1’, which signifies that the ATP6AP2-K205E has been mapped once in the Class 1 of the clinical trials. Similarly, column CT0 under clinical trials shows value ‘0’, which indicates that the ATP6AP2-K205E is not mapped under the category of Class 0 of clinical trials.
An embodiment for the step of generation of frequency table is illustrated in Figure 3. The Table 1 is presented in the Figure 3 for the sake of understanding as how the frequency table is generated from the data of associations for clinical trials forming a disease knowledgebase or part of the disease knowledgebase. In this case, two knowledge sources – clinical trials and therapies are used for generating the frequency table. The gene-mutation ERBB2-S310F is run against the clinical trial knowledgebase (part of the disease knowledgebase), and is found to be mapping four times in class CT1, two times in class CT2 and three times in class CT3 and thereafter corresponding entries are made in the frequency table against the gene-mutation ERBB2-S310F under the respective columns of the classes. Further, the gene-mutation ERBB2-S310F is also run against the therapy knowledgebase (part of the disease knowledgebase) and is found to be mapping once in class T1, once class T2 and once in class T3 thereafter corresponding entries are made in the frequency table against the gene-mutation ERBB2-S310F under the respective columns of the classes. Likewise, the other gene-mutation identified from patient’s VCF is mapped one by one and the frequency table is created. In an alternative embodiment, all the gene-mutations identified from patient’s VCF are taken together for mapping for the purpose of generating the frequency table.
The mapped mutation information in the frequency table(s) is prioritized based on a prioritization scheme at step 110. There can be various prioritization schemes designed on the basis of user requirement for sorting the frequency table. In one of the embodiments, a strict criterion is chosen for selecting data based on the preferred category of the knowledge source as primary filter. This scheme exploits linkages present between various data sources (Figure 4a). The scheme provides for:
(a) filtering the frequency table based on one category of the knowledge source selected from the one or more categories of the knowledge sources;
(b) populating the filtered frequency table with data points of the category(ies) of the knowledge source(s), not selected in previous step, linked with data point(s) associated with the selected category of the knowledge source; and
(c) sorting the frequency table based on the number of occurrence of the data point(s) viz a viz ranking of the category of the knowledge source(s) and pre-assigned precedence of the class(es) of the data point(s) present in respective category of the knowledge sources.
In an exemplary embodiment of the prioritization scheme, the clinical trials (category of the knowledge source) are chosen as primary filter. The frequency table generated (at step 108) is filtered based on clinical trials so as to list out only such mutation information of the patient which are showing corresponding entries in any of the class of the clinical trial section of the frequency table. In the next step, only such data points of the other categories of the knowledge sources (namely therapies and publications) which are related/linked with the identified data points of clinical trials in the previous step are selected and the frequency table is populated accordingly. In the final step, the entries in the frequency table are sorted by giving higher ranking to the gene-mutations indicating higher entries under the corresponding classes. The ranking of the gene-mutations is performed while taking into consideration the ranks assigned to the knowledge sources and the precedence assigned to the classes falling under those knowledge sources (Figure 4a).
In another embodiment, the prioritization scheme provides for sorting while considering all the evidence present (in the frequency table) for a given mutation independently (Figure 4b):
(a) arranging the linked data point(s) in the frequency table category (of the knowledge source)-wise and subsequently respective class-wise; and
(b) sorting the frequency table for the mapped mutation information based on the number of occurrence of the data point(s) viz a viz ranking of the category of the knowledge sources and pre-assigned precedence of the class(es) of the data points present in respective category of the knowledge sources.
In one of the embodiment the sorting technique uses multilevel sort and below is a representation of such sorting (Figure 5). Each mutation is assigned a score S(m)
S(m)= F({CT0, CT1, CT2, CT3}, {T0, T1, T2, T3}, {P0, P1, P2, P3})
Score for a mutation is represented by S(m) and is computed as

Where,
k:number of category(ies) of the knowledge sources
c: Total number of classes for each category of the knowledge sources (for example c= 4: Class 0 to 3)
Nij : Represents the number of data points belonging to category of the knowledge sources i and Class j
t: t is chosen so that 10t represents the maximum number of data per class
In an exemplary embodiment of the prioritization scheme, the sorting is performed while considering the clinical trials and the therapies independently. The data filtration is performed independently on both the clinical trials and the therapies. However, it is to be appreciated that this can be extended over any number of the category of the knowledge sources. The present case, the clinical trials are ranked higher as compared to the therapies (Clinical trial>Therapies). The frequency table generated after the step 108 is sorted as per the prioritization scheme (Figure 5). After the sorting, the row nos. 6 and 7 are presented at top of the sorted frequency table, where
Row #6 has evidence scores 4, 2, 3, 4, 5, 4, 3, 2
Row #7 has evidence scores 5, 2, 3, 4, 5, 2, 3, 1
Thereafter, based on a simple sorting mechanism the sorted order for these two entries would be Row #7 and Row #6
In alternative embodiment of the prioritization scheme, the therapies are ranked higher as compared to the clinical trials (Therapies > Clinical trials) (Figure 6). After the sorting, the row nos. 6 and 7 are presented at the top of the sorted table, where
Row #6 has evidence scores for clinical trials = 4, 2, 3, 4 and therapies = 5, 4, 3, 2
Row #7 has evidence scores for clinical trials = 5, 2, 3, 4 and therapies = 5, 2, 3, 1

In this scenario since the higher priority is given to therapy classes. Hence the evidence order would be
Row #6 has evidence scores 5, 4, 3, 2, 4, 2, 3, 4
Row #7 has evidence scores 5, 2, 3, 1, 5, 2, 3, 4

In this case, based on a simple sorting mechanism the sorted order for these two entries would be Row #6 and Row #7

It is to be understood that the ranking of the different knowledge sources used in the present invention depends on the requirement of user. Once the sorted frequency table is created, it becomes easy for the doctors to choose the correct way to personalise a therapy for the patient based on the evidences/information being made available.

The present invention also provides for a device for mutation prioritization for personalized therapy of patient(s). Figure 7 depicts a block level diagram of the device for the mutation prioritization in accordance with an exemplary embodiment of the present invention. The device is configured to prioritize the mapped mutation information and thereby to generate a list of prioritized mutation for the perusal of doctors/care givers.
The device 700 includes processor(s) 706, and memory 702 coupled to the processor(s) 706.
The processor(s) 706, 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 702 includes a plurality of modules stored in the form of executable program which instructs the processor 706 to perform the method steps illustrated in Figure 1. The memory 702 has following modules: mutation information acquisition module 708, mapping module 710, identification module 712, frequency table generation module 714, and prioritization module 716. The memory 702 may also have the disease knowledgebase. Alternatively, the disease knowledgebase is communicatively coupled to the device through any means of communication.
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) 706.
The mutation information acquisition module 708 instructs the processor(s) 706 to perform the step 102 (Figure 1).
The mapping module 710 instructs the processor(s) 706 to perform the step 104 (Figure 1).
The identification module 712 instructs the processor(s) 706 to perform the step 106 (Figure 1).
The frequency table generation module 714 instructs the processor(s) 706 to perform the step 108 (Figure 1).
The prioritization module 716 instructs the processor(s) 706 to perform the step 110 (Figure 1).

The Disease Knowledgebase and Method of generating the same
The present invention also provides a method for generating the Disease Knowledgebase. Method steps broadly include aggregation of raw data from various public data sources in to a local repository. Cleaning and curating aggregated data collecting specific information ( and data of clinical relevance) called as data points and identifying associations between the data points. Such curated information/data point associations is/are then classified according to classification rules to create the disease knowledgebase. Therefore, the disease knowledgebase includes various knowledge sources linked to three primary categories: clinical trials, therapies and publications. The clinical trials, therapies and publication knowledge sources are independently curated and classified in the context of the present invention. Further, the classification rules for the respective categories (clinical trials, therapies and publications) of the knowledge sources are designed as per the requirement of the user. Hence, there can be differences between classification of data points falling under clinical trials as compared to the therapies or publications.

The flow diagram as given in Figure 8 provides the detailed steps of the present method for generating the disease knowledgebase, according to one embodiment. Information pertaining to alteration(s) of genomic DNA, gene(s), disease(s) and the parameter(s) of clinical relevance from various knowledge sources are obtained at step 802.
The obtained information is curated to extract data point(s) indicative of the alteration(s) of genomic DNA, gene(s), disease(s) and parameter(s) of clinical relevance from knowledge source(s) at step 804. Therefore, after curation broadly two sets of data points are created, i.e., one set for data point(s) indicative of the alteration(s) of genomic DNA, gene(s), disease(s), while another set for parameter(s) of clinical relevance.
The data of associations of the data point(s) indicative of the alteration(s) of genomic DNA, gene(s), disease(s) with the data point(s) indicative of parameters of clinical relevance are identified at step 806. For example, a data point of Breast tumour:ERBB2: S310F can find a match with a clinical trial where the inclusion criteria covers the breast cancer related to gene ERBB2 and associated mutation S310F.
The data point(s) in the knowledge source(s) associated with are classified into a plurality of classes at step 808. The step involves classifying the association of the data point(s) for their linkage with disease(s), gene(s), and alteration(s) of genomic DNA in the category(ies) of the knowledge source(s) into class(es). The class(es) are assigned to each of the data point for its linkage with the disease(s), gene(s), and alteration(s) of genomic DNA. Therefore, this classification is relative to each set. So if a given data point is associated with multiple data points ( sets) from a given knowledge source, then it could have a different classification for each (as discussed previously under the section dealing with Mutation prioritization for personalized therapy of patient). The classes are predefined for each of the category(ies) of the knowledge source(s) with pre-assigned precedence. Further, the category of the knowledge sources are also ranked based on one of user input and in any pre-defined priority. The classification of the data points falling under the three primary categories: clinical trials, therapies and publications of the various knowledge sources is done on the similar way as explained earlier in earlier part of the specification (Figure 1).

For brevity, the step of obtaining/aggregating the data from the plurality of categories of the knowledge sources and curating the data to obtain data points and classifying the data points is illustrated in Figure 9.

Finally the disease knowledgebase is generated based on the classified one or more data points in the one or more category(ies) of the knowledge sources at step 810. The disease knowledgebase so generated includes - category (of the knowledge source)-wise and subsequently respective class-wise arrangement of data of associations of data point(s) indicative of the alteration(s) of genomic DNA, the gene(s), and the disease(s) derived from knowledge source(s) with data point(s) indicative of parameters of clinical relevance from knowledge source(s).

The classification rules for classifying data points falling under clinical trials, therapies, and publications for each scenario has been detailed below:

(A) Clinical Trials
The set (the data point) identified from a clinical trial is assigned a specific class, where every class signifies extent of relevance of a given gene and mutation to a clinical trial for a disease. Classes are labelled as CT0, CT1, CT2 and CT3 and the precedence assigned to the classes makes CT0 most relevant class while CT3 as least relevant class. Classification rules for clinical trials are listed in Table-2. It is to be understood that the definition of the classes (provided in Table-2 are examples of indicative of the parameters of clinical relevance). For example, a data points indicating information for is included in class CT0 and likewise.

Table 2

(B) Therapies
The association of to drug or drug action is done through curation of published studies. Approval status (on-label/off-label) of a given drug is obtained using US FDA drug label information. The classification of therapies is performed based on the acquired patient mutation and disease information. As therapy classification is dependent on patient specific information, it is performed while processing patients’ data. Classes are labelled as T0, T1, T2 and T3 and the precedence assigned to the classes makes T0 most relevant class while T3 as least relevant class. It is to be understood that the definition of the classes (provided in Table-3 are examples indicative of the parameters of clinical relevance). The classification rules for therapy are listed in Table 3. For example, a data point indicating approved therapy for in a given patient’s cancer type is put under the class T0, and likewise.

Table 3

(C) Publications
For a given , relevant publications are identified. The identified {, publication} sets are classified into relevant classes based on the clinical, pre-clinical status of the studies discussed in the publication. It is to be understood that the definition of the classes (provided in Table-4 are examples indicative of the parameters of clinical relevance). The classes are labelled as P0, P1, P2 and P3 and the precedence assigned to the classes makes P0 most relevant class while P3 as least relevant class. The Classification rules for publications are listed in Table 4.

Table 4
As per the rules, a data point indicating that the pre-clinical and clinical studies are in agreement on the use of a therapy for a given the same is put under the class P0 and likewise.
The present invention further provides for additional classification apart from the above discussed so as to have a fine tuned classification of the category of the knowledge sources.

Apart from the main classifications provided in previous text, below criteria could be used for additional classification of category of the knowledge sources and fine grained prioritization.

(a) Location based classification for clinical trials
In an embodiment of the present invention, relevance is assigned to a clinical trial based on geographic location of the clinical trial, where the various relevant geographical locations for the patient to be treated are given precedence based on the user input (for example, 1st preference, 2nd preference, 3rd preference and so on).

(b) Drug action based classification for therapies
In an embodiment of the present invention, Drug action such as “Sensitive, Resistant, No Effect” on a given gene or mutation is used as an additional filter to sort the frequency table.

The present invention also provides for a device for generating disease knowledgebase.

Figure 10 depicts a block level diagram of the device for generating disease knowledgebase in accordance with an exemplary embodiment of the present invention. The device is configured to generating the disease knowledgebase based on the acquired raw data.
The device 1000 includes processor(s) 1006, and memory 1002 coupled to the processor(s) 1006.
The processor(s) 1006, 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 1002 includes a plurality of modules stored in the form of executable program which instructs the processor 1006 to perform the method steps illustrated in Figure 8. The memory 1002 has following modules: raw information acquisition module 1008, curating module 1010, identification module 1012, classification module 1014, and generation module 1016.
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) 1006.
The raw information acquisition module 1008 instructs the processor(s) 1006 to perform the step 802 (Figure 8).
The curating module 1010 instructs the processor(s) 1006 to perform the step 804 (Figure 8).
The identification module 1012 instructs the processor(s) 1006 to perform the step 806 (Figure 8).
The classification module 1014 instructs the processor(s) 1006 to perform the step 808 (Figure 8).
The generation module 1016 instructs the processor(s) 1006 to perform the step 810 (Figure 8).