Text influenced molecular indexing system and computer-implemented and/or computer-assisted method for same
AssigneeMerck & Co., Inc.
InventorsRichard D. Hull, Eugene M. Fluder, Jr., Suresh B. Singh
Life of this patent
Estimating the term…
Abstract
An extension of the vector space model for computing chemical similarity using textual and chemical descriptors is described. The method uses a chemical and/or textual description of a molecule/chemical and a decomposes a molecule/chemical descriptor matrix by a suitable technique such as singular value decomposition to create a low dimensional representation of the original descriptor space. Similarities between a user probe and the textual and/or chemical descriptors are then computed and ranked.
Claims64
A method of calculating similarity or substantial similarity between a first chemical descriptor and at least one other chemical descriptor in a matrix representing a plurality of chemical and textual descriptors, comprising the steps of:
- (a) creating at least one chemical descriptor and at least one textual descriptor for a plurality of compounds in a collection of compounds;
- (b) preparing a descriptor matrix X, wherein the descriptor matrix comprises:
- at least one text source containing textual and chemical descriptions, and;
- at least one descriptor associated with each respective text source,
- wherein the entries in the descriptor matrix indicate a relevancy of a descriptor to a respective text source;
- (c) performing a singular value decomposition (SVD) of the descriptor matrix to produce resultant matrices;
- (d) using at least one of the resultant matrices to compute the similarity between the first chemical descriptor di and the at least one other chemical descriptor dj, ;
- and
- e) outputting at least a subset of the at least one other chemical descriptor ranked in order of similarity to the first chemical descriptor.
A method of calculating similarity or substantial similarity between a first document Vi and at least one other document Vj representing a plurality of chemical and textual descriptors, comprising the steps of:
- (a) creating at least one chemical descriptor and at least one text descriptor for a plurality of compounds in each document;
- (b) preparing a descriptor template, wherein the descriptor template comprises:
- at least one text source containing textual and chemical descriptions, and;
- at least one descriptor associated with each respective text source,
- wherein the entries in the descriptor template indicate the the relevance of a descriptor with respect to a text source;
- (c) performing a decomposition operation on the descriptor template to produce result;
- (d) using the result to compute the similarity between the first document and the at least one other document;
- and
- e) outputting at least a subset of the at least one other document ranked in order of similarity to the first document.
A method of calculating similarity or substantial similarity between a chemical descriptor dj and at least one document Vi characterizing one or more of a plurality of chemical and textual descriptors, comprising the steps of:
- (a) creating at least one chemical descriptor and at least one text descriptor for a plurality of compounds in each document;
- (b) preparing a descriptor template, wherein the descriptor template comprises:
- at least one text source containing textual and chemical descriptions, and;
- at least one descriptor associated with each respective text source,
- wherein the entries in the descriptor template indicate the the relevance of a descriptor with respect to a text source;
- (c) performing a decomposition operation of the descriptor template to produce a result;
- (d) using the result to compute the similarity between at least one of the at least one document Vi and the chemical descriptor dj;
- and
- e) outputting at least a subset of the at least one document ranked in order of similarity to the chemical descriptor.
A method of calculating similarity or substantial similarity between a textual descriptor dj and at least one document Vi characterizing a plurality of chemical and textual descriptors, comprising the steps of:
- (a) creating at least one chemical descriptor and at least one textual descriptor for a plurality of compounds in each document;
- (b) preparing a descriptor, wherein the descriptor comprises:
- at least one text source containing textual and chemical descriptions, and;
- at least one descriptor associated with each respective text source,
- wherein the entries in the descriptor indicate the the relevancy of a descriptor with respect to a text source;
- (c) performing an operation on the descriptor matrix to produce at least one result;
- (d) using the at least one result to determine the similarity between at least one of the at least one document Vi and textual descriptor dj;
- and
- e) outputting at least a subset of the at least one document with respect to the similarity to the chemical descriptor.
A computer readable medium including instructions being executable by a computer, the instructions instructing the computer to generate a searchable representation of chemical structures, the instructions comprising:
- (a) creating at least one chemical descriptor and at least one text descriptor for each compound in a collection of compounds;
- (b) preparing a descriptor, wherein the descriptor comprises
- at least one text source containing textual and chemical descriptions, and;
- at least one descriptor associated with each respective text source,
- wherein the entries in the descriptor indicate the relevancy of a descriptor with respect to a text source;
- (c) performing an operation on the descriptor to produce at least one result;
- (d) using the at least one result to determine the similarity between the first chemical descriptor di and the at least one other chemical descriptor dj;
- and
- e) outputting at least a subset of the at least one other chemical descriptor with respect to the similarity to the first chemical descriptor.
A computer readable medium for calculating the similarity between a first text source and at least one other text source from a plurality of chemical and textual descriptors, comprising the steps of:
- (a) creating at least one chemical descriptor and at least one text descriptor for each compound in each text source;
- (b) preparing a descriptor, wherein the descriptor comprises:
- at least one text source containing textual and chemical descriptions, and;
- at least one descriptor associated with each respective text source,
- wherein the descriptor indicates the relevance of a corresponding descriptor with respect to a text source;
- (c) performing an operation on the descriptor to produce at least one result;
- (d) using the at least one result to determine the similarity between the first text source Vi and the at least one other text source Vj;
- and
- e) outputting at least a subset of the at least one other test source with respect to the similarity to the first text source.
A computer readable medium for determining the similarity between a chemical descriptor dj and at least one text source Vi from a plurality of chemical and textual descriptors, comprising the steps of:
- (a) creating at least one chemical descriptor and at least one text descriptor for each compound in each text source;
- (b) preparing a descriptor, wherein the descriptor comprises:
- at least one text source containing textual and chemical descriptions, and;
- at least one descriptor associated with each respective text source,
- wherein the entries in the descriptor indicate the relevance of a corresponding descriptor with respect to a text source;
- (c) processing the descriptor to produce at least one result;
- (d) using the at least one result to determine the similarity between at least one of the at least one text source Vi and the chemical descriptor dj;
- and
- e) outputting at least a subset of the at least one text source with respect to the similarity to the chemical descriptor.
A computer readable medium for determining the similarity between at least one textual descriptor dj and at least one text source Vi from a plurality of chemical and textual descriptors, comprising the steps of:
- (a) creating at least one chemical descriptor and at least one textual descriptor for each compound in each text source;
- (b) preparing a descriptor, wherein the descriptor comprises
- at least one text source containing textual and chemical descriptions, and;
- at least one descriptor associated with each respective text source,
- wherein the entries in the descriptor indicate the relevance of a corresponding descriptor with respect to a text source;
- (c) processing the descriptor to produce at least one result;
- (d) using the at least one result to determine the similarity between at least one of the at least one text source Vi and at least one textual descriptor dj and
- e) outputting at least a subset of the at least one text source with respect to similarity to at least one of the chemical and textual descriptors.
A method of calculating similarity or substantial similarity between a first chemical descriptor and at least one other chemical descriptor from a plurality of chemical and textual descriptors, comprising the steps of:
- (a) creating at least one chemical descriptor and at least one textual descriptor for at least one compound in a collection of compounds;
- (b) preparing a similarity descriptor wherein the similarity descriptor characterizes at least two text sources containing textual and chemical descriptions and indicates the frequency of the descriptions;
- (c) processing the similarity descriptor to produce at least one result;
- (d) using at least a portion of the at least one result to determine the similarity between the first chemical descriptor di and at least one other chemical descriptor dj, ;
- and
- e) outputting at least a subset of the at least one other chemical descriptor with respect to similarity to the first chemical descriptor.
A method of calculating similarity or substantial similarity between a first chemical descriptor and at least one other chemical descriptor from a plurality of chemical and textual descriptors, comprising the steps of:
- (a) creating at least one chemical descriptor and at least one textual descriptor for at least a plurality compounds in a collection of compounds;
- (b) preparing a frequency descriptor, from two or more independent source documents, indicating the frequency of the textual and chemical descriptors within each document for the plurality of compounds
- (c) processing the frequency descriptor to produce at least one result;
- (d) using at least a portion of the at least one result to determine the similarity between the first chemical descriptor and at least one of the other chemical descriptor;
- and
- e) outputting at least a subset of the at least one other chemical descriptor with respect to the similarity to the first chemical descriptor.
A method of calculating similarity or substantial similarity between a first chemical descriptor and at least one other chemical descriptor from a plurality of chemical and textual descriptors, comprising the steps of:
- (a) determining the similarity between the first chemical descriptor and at least one of the other chemical descriptors using a at least two independent text sources each containing a known number of one or more occurrences of at least one textual descriptor and at least one of the other chemical descriptors;
- and
- (b) outputting at least a subset of the similarity or substantial similarity of the first chemical descriptor with respect to the other chemical descriptors.
A method of calculating similarity or substantial similarity between a first chemical descriptor, di, and a second chemical descriptor, dj, associated with one or more compounds, comprising the steps of:
- (a) preparing a descriptor matrix X, wherein the descriptor matrix comprises a frequency of each of one or more chemical descriptors and one or more text descriptors in each of two or more source documents;
- (b) performing a decomposition operation on the descriptor matrix to produce at least a left singular matrix;
- (c) computing a rank of matrix X that is the least squares closest;
- and
- (d) computing the dot product between the ith and jth rows, corresponding to the first and second chemical descriptors, respectively, of the left singular matrix, over the least squares rank.
A method of calculating similarity or substantial similarity between a first chemical descriptor and at least one other chemical descriptor from a plurality of chemical and textual descriptors, comprising the steps of:
- (a) creating at least one chemical descriptor and at least one textual descriptor for at least one compound in a collection of compounds;
- (b) preparing a similarity descriptor, wherein the similarity descriptor comprises:
- a text source containing textual and chemical descriptions, and;
- a descriptor associated with each respective text source,
- wherein the similarity descriptor indicates the relevancy of a descriptor with respect to a text source;
- (c) processing the similarity descriptor to produce at least one result;
- (d) using the at least one result to determine the similarity between at least one of the first chemical descriptor di and the text source with the at least one other chemical descriptor dj, ;
- and
- e) outputting at least a subset of the at least one other chemical descriptor with respect to similarity to at least one of the first chemical descriptor and the text source.
A method of calculating similarity or substantial similarity between a first chemical descriptor and at least one other chemical descriptor from a plurality of chemical and textual descriptors, comprising the steps of:
- (a) creating at least one chemical descriptor and at least one textual descriptor for at least one of the compounds in a collection of compounds;
- (b) preparing a similarity descriptor, wherein the similarity descriptor comprises:
- a text source containing the at least one textual and chemical descriptors, and
- a corresponding descriptor associated with and having a predetermined relevancy with respect to the text source;
- (c) processing the similarity descriptor to produce at least one result;
- (d) using the at least one result to determine the similarity between the first chemical descriptor and at least one of the other chemical descriptor and the text source;
- and
- e) outputting at least a subset of the at least one other chemical descriptor with respect to the similarity to at least one of the first chemical descriptor and the text source.
A method of calculating similarity or substantial similarity between a first chemical descriptor and at least one other chemical descriptor from a plurality of chemical and textual descriptors for at least one compound in a collection of compounds, comprising the steps of:
- (a) determining the similarity between the first chemical descriptor and at least one of the other chemical descriptor and the text source using a similarity descriptor comprising a text source containing the at least one textual and chemical descriptors, and a corresponding descriptor associated with and having a predetermined relevancy with respect to the text source;
- and
- (b) outputting at least a subset of the at least one other chemical descriptor with respect to the similarity to at least one of the first chemical descriptor and the text source.
Dependencies are inferred from the claim text.