US Patent 7,054,757. Method, system, and computer program product for analyzing combinatorial libraries. Inventors: Dimitris Agrafiotis, Victor Lobanov, Francis Salemme. Assignee: Johnson & Johnson Pharmaceutical Research & Development.
Protects a method for in silico analysis of a virtual combinatorial library. A supervised machine learning approach is used to infer a mapping function that transforms the building block features for each product in a training subset of products to the corresponding mapping coordinates for each product in the training subset of products. The mapping function is then encoded in a computer readable medium and can used to generate mapping coordinates for any product in the combinatorial library from the building block features associated with the product.
US Patent 7,054,755. Interactive correlation of compound information and genomic information. Inventors: David O'Reilly, Alan Roter, Keith Bostian, David Morgans. Assignee: Iconix Pharmaceuticals.
Protects an interactive system for facilitating hypothesis construction by correlating and presenting gene expression data, bioassay data, and compound activity data, and associating gene and compound function information with product information, and facilitating product purchase. Claims describe a method for analyzing chemical genomic data that includes a database comprising at least four data types: profile information representing the expression level of genes in a cell exposed to a standard compound; bioassay information; gene information; and compound information for each of the standard compounds.
US Patent 7,054,754. Method, system, and software for deriving chemical structural information. Inventor: Jonathan Brecher. Assignee: Cambridgesoft.
Protects a method for deriving chemical structures from chemical names, including inverted names with missing commas or with extraneous spaces.
US Patent 7,047,137. Computer method and apparatus for uniform representation of genome sequences. Inventors: Simon Kasif, Beth Logan, Pedro Moreno, Baris Suzek. Assignee: Hewlett-Packard.
Protects a method for transforming differing length sequences of biological fragments into uniform length representations. A comparison database stores a predefined number of known biological sequences and a comparison routine compares and scores a subject sequence against each known sequence in the database. An individual score is determined for each known sequence in the database, which serves as a vector element in order to form a fixed-length vector representation of the subject sequence. The vector length equals the predefined number of known biological sequences in the database, and scoring is based on a probability or an occurrence count of the known biological sequence in the subject sequence.
US Patent 7,043,476. Method and apparatus for data mining to discover associations and covariances associated with data. Inventor: Barry Robson. Assignee: International Business Machines.
Covers data-mining techniques for discovering "interesting associations between entries of qualitative text, and covariances between data of quantitative numerical types, in records," according to the patent abstract. According to the inventors, the approach is particularly useful for clinical, pharmacogenomic, forensic, police, financial, and other high-dimensional records.
US Patent 7,043,415. Interactive graphical environment for drug model generation. Inventors: Michael Dunlavey, David Hermann, Jeffrey Wald, Daniel Weiner. Assignee: Pharsight.
Covers a method for interactively constructing pharmacological computational models. The invention includes a graphical user interface through which a user may place and connect objects representing pharmacokinetic and pharmacodynamic elements. "As the objects are placed and connected, they are converted into an internal format representing the statements of the corresponding computational model," according to the patent abstract. These statements are presented to the user in a summarized form, as the computational model is constructed, to enable immediate verification of the model.
US Patent 7,043,371. Method for search based character optimization. Inventor: Ward Wheeler. Assignee: American Museum of Natural History.
Protects a method for selecting a sequence in which a set of sequences is defined, and a subset of the sequences is evaluated with a dynamic programming method to obtain a "numeric indicia of merit," according to the patent abstract. Additional sequences are added into the subset of sequences, and the steps of defining, evaluating, and adding are repeated until the numeric indicia is substantially constant or there are no more additional sequences. One or more of the sequences are selected based on the numeric indicia.
US Patent 7,041,455. Method and apparatus for pattern identification in diploid DNA sequence data. Inventors: Charles Magness, Dake Sun. Assignee: Illumigen Biosciences.
Covers methods for identifying mutations and methylation patterns in diploid DNA sequence signal data. The invention describes methods for obtaining two parental allele sequences from diploid DNA sequence signal data, identifying the mutation and haplotype patterns in the two parental allele sequences, assigning likelihood scores for the mutations, and identifying patterns of methylation.