Bioinformatics, Statistical Proteomics and Genomics and Systems Biology

Menu

Navigation

Mathematic Descriptions - Multifactorial Gene Expression

Project: Mathematic descriptions - multifactorial gene expression

Agency: NIH / National Institute of General Medical Sciences

PI: Rodriguez-Zas, Sandra.

Grant Number: 1R01GM068946-01

Link: http://crisp.cit.nih.gov/crisp

Abstract:

Social behavior is a complex trait and social interactions are critical determinants of human mental and physical health. Our long-term objective is to explain the interplay among factors influencing complex traits like social behavior by the integration of genomic research, statistical analysis, bioinformatic resources and outreach. Six specific aims support this objective. 1) We will generate expression profiles for more than 7000 genes in the brain of the honey bee as a function of its complex pattern of behavioral maturation; samples will be taken every 3-4 days, as it moves through its in-hive tasks and then shifts to foraging outside. We will characterize these profiles in two distinct races known to differ in the rate of maturation. These biological findings will have a major impact on understanding behavior, neurological and mental processes, health and well-being in bees, humans and other species. Epistatic, pleiotropic, and polygenic background effects will be studied while accounting for systematic noise including array, dye and transcript level effects. 2) We will apply novel linear and nonlinear mixed-effects longitudinal models that seamlessly integrate intensity normalization and analyses. 3) We will use hierarchical models to identify between race variation. These models account for technical sources of variation and for correlation in expression that might arise among genes that are from the same family or code for proteins in the same pathway. This proposal fills an important need for general models suitable for complex sets of gene expression data. 4) Visual approaches will be developed to check for model adequacy. 5) The proposed methodology will be combined with bioinformatic resources in a publicly available suite of routines that facilitate the interpretation of results while assigning gene function. 6) The findings will be used as applied examples of mathematical and science concepts in inquiry-based, open-ended, educational materials; virtual and tangible presentations for high school students, teachers, and the general public will be developed. Our findings constitute stepping-stones for understanding and further studies of human neurological, mental, communication, and aging processes and disorders, giving rise to treatments and cures.

Related publications on gene expression profile:

Rodriguez-Zas, S. L. 2002. Comparison of statistical methods to study cDNA microarray data. Journal Dairy Science, 80 (Suppl. 2):10.

Rodriguez-Zas, S. L., and B. R. Southey. 2002. Linear mixed effects models for microarray gene expression data. In: Proceedings of the World Congress in Genetics Applied to Livestock Production. France. August 2002.

Rodriguez-Zas, S. L., M. R. Band, R. E. Everts, B. R. Southey, Z. L. Liu, and H. A. Lewin. 2003. Analysis of gene expression patterns in the cattle digestive system. Journal Animal Science, 81(Suppl. 1): 628.