US Patent 7,117,188. Methods of identifying patterns in biological systems and uses thereof. Inventors: Isabelle Guyon, Jason Weston. Assignee: Health Discovery Corporation.
Protects the use of support vector machines and RFE (recursive feature elimination) for the identification of patterns useful for medical diagnosis, prognosis, and treatment.
US Patent 7,117,098. Machine-readable storage medium for analyzing distribution of macromolecules between the cell membrane and the cell cytoplasm. Inventors: Terry Dunlay, Lansing Taylor, Albert Gough, Kenneth Giuliano. Assignee: Cellomics.
Covers systems, methods, and screens for optical system analysis “to rapidly determine the distribution, environment, or activity of fluorescently labeled reporter molecules in cells for the purpose of screening large numbers of compounds for those that specifically affect particular biological functions,” according to the patent abstract. The invention includes a computerized method for processing, displaying, and storing the data.
US Patent 7,117,097. Methods, computer software products and systems for correlating gene lists. Inventors: David Finkelstein. Assignee: Affymetrix.
Covers methods to compare lists of genes. In one embodiment, the genes are ranked using the Spearman correlation.
US Patent 7,117,095. Methods for selecting nucleic acid probes. Inventor: Earl Hubbell. Assignee: Affymetrix.
Protects methods and computer software products for selecting nucleic acid probes. In one embodiment, dynamic programming is employed to select a set of probes so that the selected probes have a maximum aggregate adjusted quality score.
US Patent 7,113,898. Method for modeling signal transduction in cells. Inventor: Emmanuele DiBenedetto, Daniele Andreucci, Paolo Bisegna. Assignee: Vanderbilt University.
Covers a model that can be use to describe biological processes having a complex spatial geometry, such as cell signaling processes. The method can be used to model the pathology underlying diseases characterized by disrupted signaling pathways, and to identify target proteins at an appropriate site in a signal transduction network. Methods of modeling diffusion processes, as well as machine-readable products embodying the methods, are also disclosed in the patent.
US Patent 7,113,896. System and methods for processing biological expression data. Inventors: Zhen Zhang, Hong Zhang.
Protects a method for processing information in a data set that contains samples of at least two classes using an empirical risk minimization model, in which each sample in the data set has an importance score.
US Patent 7,113,870. Prediction of molecular polar surface area and bioabsorption. Inventors: William Egan, Giorgio Lauri. Assignee: Acclerys.
Covers a method for computing the polar surface area of molecules without reference to three dimensional molecular structures. The method uses a linear equation incorporating counts of nitrogens, oxygens, and related atom clusters.
US Patent 7,110,890. Auto-analysis framework for sequence evaluation. Inventors: Thomas Birkett, Puneet Suri, Kai Yung, Sylvia Fang, Salil Kumar, Benjamin Jones, Jon Sorenson, Yuandan Lou, Catherine Frantz. Assignee: Applera.
Protects an automated system for evaluating biological samples. The system includes a centralized registry that contains protocols and configuration information for instruments and analysis applications. The registry “provides for improved automation of biological process runs using an autoanalysis applications manager component or daemon, which accesses and transmits the appropriate protocol and configuration information to selected instruments and/or applications,” according to the patent abstract. This information is used to “instruct data capture by the biological instruments and direct the analysis of the data by the analysis applications.”
US Patent 7,110,888. Method for determining a shape space for a set of molecules using minimal metric distances. Inventor: Anthony Nicholls. Assignee: OpenEye Scientific Software.
Covers techniques for characterizing molecules based on the shapes of their fields. The minimal distance between two molecular fields is used as a shape-based metric, independent of the underlying chemical structure, and a high-dimensional shape space description of the molecules is generated. These attributes can then be used in creating, characterizing, and searching databases of molecules based on field similarity. “The most immediate application of these techniques is to pharmaceutical drug discovery and design,” according to the patent abstract.
US Patent 7,110,885. Efficient methods and apparatus for high-throughput processing of gene sequence data. Inventor: Tony Frudakis. Assignee: DNAPrint Genomics.
Protects a method of processing gene sequence data that includes the steps of reading gene sequence data; identifying and storing primer pair data within the gene sequence data for one of the coding sequences; repeating the acts of identifying and storing such that primer pair data are obtained for each coding sequence; and simultaneously amplifying the coding sequences from three or more of individuals using the identified pairs of primer sequences.