Past Event! Note: this event has already taken place.
Machine learning approaches in cancer genomics
February 22, 2018 at 10:30 AM to 12:00 PM
| Location: | 5345 Herzberg Laboratories |
| Cost: | Free |
Abstract:
The development of high-throughput genomic data acquisition technologies has revolutionized the field of cancer biology. However, the large size of generated data, their diversity and their unique characteristics have necessitated the development of specialized machine learning techniques that are able to analyze and integrate these different types of data in order to obtain biologically meaningful insights.
In the first part of this talk I will describe ProGENI, a novel method for prioritization of genes whose expression determines the sensitivity of cancer cells to different treatments. Although “off-the-shelf” feature selection methods may be used for this task, many important genes would escape identification due to the complexity of drugs’ mechanism of action, noisy data and the fact that these methods overlook known relationships among genes. To overcome these issues, this method utilizes random walk techniques to integrate prior knowledge in the form of a gene interaction network with the gene expression data and significantly improves the prioritization accuracy.
In the second part of the talk, I will describe InPheRNo, a new method for the reconstruction of phenotype-relevant transcriptional regulatory networks (TRNs). A TRN is a directed network reflecting the causal relationships between transcription factors (proteins regulating the expression of genes) and their target genes. This method utilizes a carefully designed probabilistic graphical model to simultaneously capture the effects of multiple transcription factors on their targets and relate target genes’ expressions to the phenotype variation. A pan-cancer analysis using this method reveals known and new regulatory mechanisms controlling the development of each cancer type and identifies cancer-driver transcription factors.
Biography:
Amin Emad is a postdoctoral research associate at the department of computer science at the University of Illinois at Urbana-Champaign (UIUC), working with Saurabh Sinha. Amin received his PhD in Electrical and Computer Engineering from UIUC in 2015 under the supervision of Olgica Milenkovic. He received his MSc degree from University of Alberta and BSc degree from Sharif University of Technology. He is the recipient of various fellowships and awards including the gold medal in physics Olympiad in Iran, the NSERC Alexander Graham Bell Canada Graduate Scholarship, the Sundaram Seshu fellowship, the CompGen fellowship, and the NSERC Postdoctoral fellowship. His research focuses on developing novel machine learning and graph mining techniques for cancer genomics.