Recent Patents in Bioinformatics, July — August 2011
US Patent 8,000,900. Association-based predictions of pathogen characteristics. Inventors: David E. Heckerman, Simon Mallal, Carl M. Kadie, Corey Benjamin Moore, Nebojsa Jojic. Assignee: Microsoft.
Describes a system comprised of a machine-learning classifier that’s trained to predict disease states based on associations between a host and pathogen. The method includes creating computerized instructions for using population data to analyze one or more parts of a target molecule that could have an impact on the host's health.
US Patent 7,996,157. Genetic diagnosis using multiple sequence variant analysis. Inventors: Marc Zabeau, Patrick Stanssens, Yannick Gansemans. Assignee: Methexis Genomics.
This patent describes an approach that aims to define a subset of genetic variants that can be used as genetic markers to search for correlations with certain phenotypic traits. These markers can then be used to develop algorithms and computer programs that convert genotype data into constituent haplotypes. The invention has applications in genome-wide association studies, clinical in vitro diagnosis, plant and animal breeding, and identifying micro-organisms, according to the inventors.
US Patent 7,996,155. ANOVA method for data analysis. Inventors: Lee Weng. Assignee: Microsoft.
This patent describes improved analysis of variation, or ANOVA, methods for analyzing measured and transformed data that is useful for analyzing gene or protein expression data. The approach accepts two data types as input — data measurements and a predetermined error associated with the measurements. This added input information increases the statistical power of the method, the patent abstract states.
US Patent 7,991,563. Sequence pattern descriptors for transmembrane structural details. Inventors: Isidore Rigoutsos, Peter Riek, Robert M. Graham, Jiri Novotny. Assignee: IBM.
Describes a method for detecting sequences and identifying structural details in transmembrane proteins by comparing a query sequence to composite pattern descriptors.
US Patent 7,991,560. System, method, and computer software for the presentation and storage of analysis results. Inventors: Shantanu V. Kaushikkar, Teresa Webster, Rui Mei, Linda McAllister. Assignee: Affymetrix.
This patent describes a computer program that processes emission intensity data corresponding to the probes of a biological probe array. According to the patent abstract, the program "includes a genotype and statistical analysis manager that determines absolute or relative expression values based, at least in part, on a statistical measure of the emission intensity data and at least one user-selectable statistical parameter."
Additionally, the analysis manager can determine genotype calls for one or more probes based, in part, on the emission intensity data and "display the absolute or relative expression values based ... on at least one user-selectable display parameter and/or a measure of normalized change between genotype calls."
US Patent 7,991,557. Computer system and methods for constructing biological classifiers and uses thereof. Inventors: Choong-Chin Liew, Thomas Yager, Adam Dempsey, Samuel Chao. Assignee: GeneNews.
Describes a method of constructing classifiers that distinguish between trait subgroups using molecular marker data from blood samples. The system can be used for disease prediction, diagnosis, and prognosis as well as to monitor disease progression and/or regression, among other applications.