Seminar: Inferring Gene Function from Atlantic Cod Gene Expression Data using Machine Learning

Shuyue Qi
M.Sc. Candidate
Supervisor: Dr. Lourdes Peña-Castillo

Inferring Gene Function from Atlantic Cod Gene Expression Data using Machine Learning

Department of Computer Science
Wednesday, July 26, 2017, 11:00 a.m., Room EN 2022


Abstract

Identifying gene function has many useful applications such as plant breeding, reproductive biotechnology in animals and drug discovery. The goal of this project is to identify Atlantic Cod gene function based on gene expression data and gene co-expression module membership. First, I will discuss the application of Support Vector Machines (SVMs) and Neural Networks (NNs) in predicting gene functions. In this project, I used AZURE Machine Learning Studio to apply SMOTE method, to build SVM models and NN models, and to use an Ensemble method to predict gene functions in Atlantic cod. To evaluate the performance of the different models, I used accuracy score, F1-score and recall score to get the best model to infer ten gene functions. Experimental results indicate that ensemble classification methods for gene function prediction in Atlantic cod outperform the second best model by 7.8% in average accuracy.

 

Contact

Department of Computer Science

230 Elizabeth Ave

St. John's, NL A1B 3X9 CANADA

Tel: (709) 864-2530

Fax: (709) 864-2552

becomestudent@mun.ca