Abstract: A system, method, and computer program product for fast and efficient searching of large virtual combinatorial libraries based on a fitness function. According to the method of the present invention, a first set of N reagent combinations are selected, for example, at random, from a virtual combinatorial library. Each reagent combination in the first set is then enumerated to produce a first set of enumerated compounds. M number of compounds of the first set of enumerated compounds are selected based on the fitness function. The M compounds are then deconvoluted into reagents to generate a focused library. Substantially every reagent combination associated with the focused library is enumerated to produce a second set of enumerated compounds. K number of compounds of the second set of enumerated compounds are then selected based on the fitness function. These K compounds represent a near optimal selection of compounds based on the fitness function.

Abstract: A system, method, and computer program product for visualizing and interactively analyzing data relating to chemical compounds. A user selects a plurality of compounds to map, and also selects a method for evaluating similarity/dissimilarity between the selected compounds. A non-linear map is generated in accordance with the selected compounds and the selected method. The non-linear map has a point for each of the selected compounds, wherein a distance between any two points is representative of similarity/dissimilarity between the corresponding compounds. A portion of the non-linear map is then displayed. Users are enabled to interactively analyze compounds represented in the non-linear map.

Abstract: A method, system, and computer product is presented for mapping a set of patterns into an m-dimensional space so as to preserve relationships that may exist between these patterns. A subset of the input patterns is chosen and mapped into the m-dimensional space using an iterative nonlinear mapping process based on subset refinements. A set of n attributes are determined for each pattern, and one or more neural networks or other supervised machine learning techniques are then trained in accordance with the mapping produced by the iterative process. Additional input patterns not in the subset are mapped into the m-dimensional space by determining their n input attributes and using the neural networks in a feed-forward (prediction) mode.

Abstract: A method, system and computer program product are provided for scaling, or dimensionally reducing, multi-dimensional data sets that scale well for large data sets. The invention scales multi-dimensional data sets by determining one or more non-linear functions between a sample of points from the multi-dimensional data set and a corresponding set of dimensionally reduced points. Thereafter, these one or more non-linear functions are used to non-linearly map additional points. The additional points may be members of the original multi-dimensional data set or may be new, previously unseen points. In an embodiment, the determination of the non-linear relationship between the sample of points from the multi-dimensional data set and the corresponding set of dimensionally reduced points is performed by a self-learning system such as a neural network. The additional points are mapped using the self-learning system in a feed-forward/predictive manner.

Abstract: The invention provides for in silico analysis of a virtual combinatorial library. Mapping coordinates for a training subset of products in the combinatorial library, and features of their building blocks, are obtained. A supervised machine learning approach is used to infer a mapping function ƒ that transforms the building block features for each product in the training subset of products to the corresponding mapping coordinates for each product in the training subset of products. The mapping function ƒ is then encoded in a computer readable medium. The mapping function ƒ can be retrieved and used to generate mapping coordinates for any product in the combinatorial library from the building block features associated with the product.

Abstract: A method and computer product is presented for mapping n-dimensional input patterns into an m-dimensional space so as to preserve relationships that may exist in the n-dimensional space. A subset of the input patterns is chosen and mapped into the m-dimensional space using an iterative nonlinear mapping process. A set of locally defined neural networks is created, then trained in accordance with the mapping produced by the iterative process. Additional input patterns not in the subset are mapped into the m-dimensional space by using one of the local neural networks. In an alternative embodiment, the local neural networks are only used after training and use of a global neural network. The global neural network is trained in accordance with the mapping produced by the iterative process. Input patterns are initially projected into the m-dimensional space using the global neural network. Local neural networks are then used to refine the results of the global network.

Abstract: The present invention determines properties of combinatorial library products from features of library building blocks. At least one feature is determined for each building block of a combinatorial library having a plurality of products. A training subset of products is selected from the products, and at least one property is determined for each product of the training subset. A building block set is identified for each product of the training subset, and an input features vector is formed from the features of the identified building blocks for each product of the training subset. A supervised machine learning approach is used to infer a mapping function that transforms the input features vector for each product of the training subset to the corresponding at least one property for each product of the training subset. After the mapping function is inferred, it is used for determining properties of other products of the library from their corresponding input features vectors.

Abstract: The present invention provides a method, system, and computer program product for encoding and building products of a virtual combinatorial library. A chemical reaction and reagent data for forming products of the virtual combinatorial library are encoded in a computer readable form. A compiler then operates on the encoded information and generates reagent mapping data. The compiler compiles the encoded chemical reaction to genenerate computer instructions that control the operation of a processor. A compact data structure containing data is generated and stored in a memory. This data structure is then used to gain immediate access to any of the products of the virtual combinatorial library.

Abstract: A greedy method for designing combinatorial arrays. An array of reagents are initially selected from a list of candidate reagents in a combinatorial library. The reagents in the array that maximize a design objective are determined in an iterative manner, by examining each variation site in the combinatorial library in a strictly alternating sequence. During each step, each candidate reagent at a given variation site is evaluated by constructing and evaluating the sub-array resulting from the systematic combination of that reagent with the selected reagents at all the other variation sites in the library. The candidate reagents at that variation site are ranked according to the fitness of their respective sub-arrays, and the reagents with the highest fitness are selected. The process is repeated for each variation site in the combinatorial library, until the fitness of the full combinatorial array can no longer be improved.

Abstract: A method, system and computer program product are provided for scaling, or dimensionally reducing, multi-dimensional data sets that scale well for large data sets. The invention scales multi-dimensional data sets by determining one or more non-linear functions between a sample of points from the multi-dimensional data set and a corresponding set of dimensionally reduced points. Thereafter, these one or more non-linear functions are used to non-linearly map additional points. The additional points may be members of the original multi-dimensional data set or may be new, previously unseen points. In an embodiment, the determination of the non-linear relationship between the sample of points from the multi-dimensional data set and the corresponding set of dimensionally reduced points is performed by a self-learning system such as a neural network. The additional points are mapped using the self-learning system in a feed-forward/predictive manner.

Abstract: A method, system and computer program product for scaling, or dimensionally reducing, multi-dimensional data sets, that scales well for large data sets. The invention scales multi-dimensional data sets by determining one or more non-linear functions between a sample of points from the multi-dimensional data set and a corresponding set of dimensionally reduced points, and thereafter using the non-linear function to non-linearly map additional points. The additional points may be members of the original multi-dimensional data set or may be new, previously unseen points. In an embodiment, the invention begins with a sample of points from an n-dimensional data set and a corresponding set of m-dimensional points. Alternatively, the invention selects a sample of points from an n-dimensional data set and non-linearly maps the sample of points to obtain the corresponding set of m-dimensional points.

Abstract: A computer based, iterative process for generating chemical entities with defined physical, chemical and/or bioactive properties. During each iteration of the process, (1) a directed diversity chemical library is robotically generated in accordance with robotic synthesis instructions; (2) the compounds in the directed diversity chemical library are analyzed to identify compounds with the desired properties; (3) structure-property data are used to select compounds to be synthesized in the next iteration; and (4) new robotic synthesis instructions are automatically generated to control the synthesis of the directed diversity chemical library for the next iteration.

Abstract: An automatic, partially automatic, and/or manual iterative system, method and/or computer program product for generating chemical entities having desired or specified physical, chemical, functional, and/or bioactive properties. The present invention identifies a set of compounds for analysis; collects, acquires or synthesizes the identified compounds; analyzes the compounds to determine one or more physical, chemical and/or bioactive properties (structure-property data); and uses the structure-property data to identify another set of compounds for analysis in the next iteration. An Experiment Planner generates Selection Criteria and/or one or more Objective Functions for use by a Selector. The Selector searches the Compound Library to identify a subset of compounds (a Directed Diversity Library) that maximizes or minimizes the Objective Functions. The compounds listed in the Directed Diversity Library are then collected, acquired or synthesized, and are analyzed to evaluate their properties of interest.

Abstract: The invention provides for in silico analysis of a virtual combinatorial library. Mapping coordinates for a training subset of products in the combinatorial library, and features of their building blocks, are obtained. A supervised machine learning approach is used to infer a mapping function ƒ that transforms the building block features for each product in the training subset of products to the corresponding mapping coordinates for each product in the training subset of products. The mapping function ƒ is then encoded in a computer readable medium. The mapping function ƒ can be retrieved and used to generate mapping coordinates for any product in the combinatorial library from the building block features associated with the product.

Abstract: A system, method and computer program product for representing precise or imprecise measurements of similarity/dissimilarity (relationships) between objects as distances between points in a multi-dimensional space that represents the objects. Self-organizing principles are used to iteratively refine an initial (random or partially ordered) configuration of points using stochastic relationship/distance errors. The data can be complete or incomplete (i.e. some relationships between objects may not be known), exact or inexact (i.e. some or all of the relationships may be given in terms of allowed ranges or limits), symmetric or asymmetric (i.e. the relationship of object A to object B may not be the same as the relationship of B to A) and may contain systematic or stochastic errors. The relationships between objects may be derived directly from observation, measurement, a priori knowledge, or intuition, or may be determined indirectly using any suitable technique for deriving proximity (relationship) data.

Abstract: A computer based, iterative process for generating chemical entities with defined physical, chemical and/or bioactive properties. During each iteration of the process, (1) a directed diversity chemical library is robotically generated in accordance with robotic synthesis instructions; (2) the compounds in the directed diversity chemical library are analyzed to identify compounds with the desired properties; (3) structure-property data are used to select compounds to be synthesized in the next iteration; and (4) new robotic synthesis instructions are automatically generated to control the synthesis of the directed diversity chemical library for the next iteration.

Abstract: A method, system, and computer product is presented for mapping a set of patterns into an m-dimensional space so as to preserve relationships that may exist between these patterns. A subset of the input patterns is chosen and mapped into the m-dimensional space using an iterative nonlinear mapping process based on subset refinements. A set of n attributes are determined for each pattern, and one or more neural networks or other supervised machine learning techniques are then trained in accordance with the mapping produced by the iterative process. Additional input patterns not in the subset are mapped into the m-dimensional space by determining their n input attributes and using the neural networks in a feed-forward (prediction) mode.

Abstract: An automatic, partially automatic, and/or manual iterative system, method and/or computer program product for generating chemical entities having desired or specified physical, chemical, functional, and/or bioactive properties. The present invention identifies a set of compounds for analysis; collects, acquires or synthesizes the identified compounds; analyzes the compounds to determine one or more physical, chemical and/or bioactive properties (structure-property data); and uses the structure-property data to identify another set of compounds for analysis in the next iteration. An Experiment Planner generates Selection Criteria and/or one or more Objective Functions for use by a Selector. The Selector searches the Compound Library to identify a subset of compounds (a Directed Diversity Library) that maximizes or minimizes the Objective Functions. The compounds listed in the Directed Diversity Library are then collected, acquired or synthesized, and are analyzed to evaluate their properties of interest.

Abstract: A method and computer product is presented for mapping n-dimensional input patterns into an m-dimensional space so as to preserve relationships that may exist in the n-dimensional space. A subset of the input patterns is chosen and mapped into the m-dimensional space using an iterative nonlinear mapping process. A set of locally defined neural networks is created, then trained in accordance with the mapping produced by the iterative process. Additional input patterns not in the subset are mapped into the m-dimensional space by using one of the local neural networks. In an alternative embodiment, the local neural networks are only used after training and use of a global neural network. The global neural network is trained in accordance with the mapping produced by the iterative process. Input patterns are initially projected into the m-dimensional space using the global neural network. Local neural networks are then used to refine the results of the global network.

Abstract: A system, method, and computer program product for visualizing and interactively analyzing data relating to chemical compounds. A user selects a plurality of compounds to map, and also selects a method for evaluating similarity/dissimilarity between the selected compounds. A non-linear map is generated in accordance with the selected compounds and the selected method. The non-linear map has a point for each of the selected compounds, wherein a distance between any two points is representative of similarity/dissimilarity between the corresponding compounds. A portion of the non-linear map is then displayed. Users are enabled to interactively analyze compounds represented in the non-linear map.