Nazia Fatima

 

Date:  Thursday, September 12, 2019

Time:  3:30 pm – 5:30 pm

Location: 3105, Lambton Tower

 

To understand the complex cellular mechanisms involved in a biological system, it is necessary to study the nature and specificity of various parameters and their relations with each other. Only then we can obtain reliable biomarkers to be used for diagnosis, treatment and prognosis of the disease. With the advent of novel high-throughput biotechnological tools such as deep sequencing has lead to a number of very large datasets that require specialized algorithms for analyzing the data. Various machine learning algorithms has been devised to achieve this results. With the advent of Deep learning prediction can be done based on huge data and gives more accurate results. Integration of multi-omics data in applications for the identification of biomarkers that will help understand the transcriptional and genetic mechanisms involved in the development of prostate cancer.

 

In this thesis, we introduce a systematic, and generalized method, called iSOM-GSN, used to transform multi-omic genomic data with higher dimensions to a two-dimensional grid. Afterwards, we apply a convolutional neural network to predict disease states of various types. Based on the idea of Kohonen's self-organizing map (SOM), we generate a two-dimensional grid for each sample for a given set of genes that represent a gene similarity network (GSN). The set of genes that are significantly highly mutated across the whole genome and which are related to each other based on functional interactions. We then test the model to predict breast and prostate cancer using gene expression, DNA methylation and copy number alteration, yielding accuracies in the 94-98% range for tumor stages of breast cancer and calculated Gleason scores of prostate cancer with just 11 input genes for both cases. The scheme not only outputs nearly perfect classification accuracy, but also provides an enhanced scheme for visualization, dimensionality reduction, and interpretation of the results.

Thesis Committee:

 

Thesis Defense Announcement

5113 Lambton Tower, 401 Sunset Ave.,Windsor ON. N9B 3P4 (519) 253-3000 Ext. 3716 csgradinfo@uwindsor.ca