Activity Number:
|
154
|
Type:
|
Contributed
|
Date/Time:
|
Monday, July 30, 2007 : 10:30 AM to 12:20 PM
|
Sponsor:
|
Biometrics Section
|
Abstract - #310117 |
Title:
|
Binary Logistic Regression Modeling of Precursor Sequence Cleavage
|
Author(s):
|
Allison Tegge*+ and Sandra Rodriguez-Zas and Bruce Southey
|
Companies:
|
University of Illinois at Urbana-Champaign and University of Illinois at Urbana-Champaign and University of Illinois at Urbana-Champaign
|
Address:
|
1207 Gregory Dr, Urbana, IL, 61801,
|
Keywords:
|
binary logistic model ; bioinformatics ; sequence analysis
|
Abstract:
|
Neuropeptides are signaling molecules critical in neural communication. Approaches to predict neuropeptides resulting from complex posttranslational enzymatic processing of precursors can support efficient experimental confirmation. We used a binary logistic model to predict precursor cleavage using amino acid location information. Logistic regression models were trained and cross-validated using precursor sequences and cleavage information from multiple mammalian species. Complementary variable selection methods were used to identify parsimonious models. The sensitivity of the models to correctly predict cleavages sites ranged from 68 to 75%. The species-specific logistic cleavage predictive equations differed in the explanatory variables and regression coefficients. Logistic modeling helped uncover species-specific sequence features influencing precursor sequence cleavage.
|