Knowledge-based artificial neural network modeling assessment: integrating heterogeneous genomics data to uncover lifespan regulation

Biological analytics and more advanced data analysis techniques have made remarkable advancements as the area of machine learning continues to grow. More specifically, genetic modeling and neural network building are gaining interest as it becomes a fundamental piece of most model building we see today. We propose a Knowledge-Based Artificial Neural Network (KBANN) to predict phenotype while providing insight to effected subsystems. Within KBANN, the input layers are a single or group of Gene Ontology (GO) terms while each layer’s input is a single number between 0 and 1, explaining how expressed the given term is. The expression number provides an average of the number of copies that a gene is producing at its current age compared to that over the average of its entire lifespan. Preliminary results show that KBANN model can potentially be used to predict lifespan phenotype using the Genotype-Tissue Expression data.

Grantor dc:publisher

University of Tennessee at Chattanooga

Author dc:creator

• Day, Taylor

Contributors dc:contributor

• Qin, Hong
• Liang, Yu; Wu, Dalei; Yang, Li
• College of Engineering and Computer Science

• Genetics--Mathematical models
• Neural networks (Computer science)

Licence

• http://rightsstatements.org/vocab/InC/1.0/

Language dc:language

English, eng

Repository record dc:identifier

https://scholar.utc.edu/theses/741

OAI identifier oai:identifier

oai:scholar.utc.edu:theses-1909