Skip to content. | Skip to navigation

Personal tools

• Monday Feb. 25th at 12:00pm: BC Cancer Research Seminar Series presents “Profiling copy number aberrations and loss of heterozygosity mutational landscapes in cancer genomes” by Gavin Ha

— filed under:

Gavin Ha - Lloyd Skarsgard Graduate Student Research Excellence Prize 2012 Recipient, PhD candidate, CIHR Bioinformatics Training Program (UBC), Department of Molecular Oncology, BC Cancer Research Centre

  • Seminar
When February 25, 2013
from 12:00 PM to 01:00 PM
Where Gordon & Leslie Diamond Family Theatre - BC Cancer Agency Research Centre (675 W.10th Ave.)
Contact Name Host: Dr. Sohrab Shah
Add event to calendar vCal


Genome instability resulting from somatic aberrations such as segmental copy number alterations (CNA) and loss of heterozygosity (LOH) is a hallmark of cancer. Gene dosage of tumour suppressors and oncogenes are often impacted by these events; therefore, comprehensive profiling can help to discover novel candidate drivers of tumourigenesis and to characterize individual tumours. High density genotyping arrays and, more recently, whole genome sequencing of cancer genomes have been popular high resolution assay platforms used to interrogate CNAs and LOH. However, accurate assessment of somatic alterations from these technologies requires careful consideration of technical and biological artefacts such as normal admixture and germline variations. To address these limitations, we have developed probabilistic algorithms for more precise profiling of somatic alterations in cancer genomes.

I will introduce the algorithms and discuss the results from two published cancer studies for which we have applied our methodologies. The first study is the analysis of the copy number landscapes in the METABRIC dataset of 2000 breast cancer samples hybridized to SNP6.0 arrays. The consequent effect of the CNAs was determined by comparing to matching gene expression data and subsequently used in combination to discover novel molecular subgroups. The second study involved the development of APOLLOH, a tool which predicts LOH in whole genome sequencing of cancers, and its application to analyze the genomes of 23 triple-negative breast cancers. Analysis of matching transcriptome-sequencing data (RNAseq) was performed to investigate the contribution of genomic events on imbalanced allelic expression.

Detection of CNA and LOH are also complicated by the admixture of normal cells and the coexistence of heterogeneous tumour subpopulations. In the final section, I will present current work in which we have developed a new probabilistic framework designed to predict CNA and LOH while accounting for subclonal heterogeneity in bulk tumour samples. I will present preliminary results for a high-grade serous ovarian cancer sample.     



Gavin Ha is completing his PhD research at the Department of Molecular Oncology at the BC Cancer Research Centre under co-supervisors Dr. Sohrab Shah and Dr. Samuel Aparicio.  He completed his BSc in the Combined Program of Microbiology/Immunology and Computer Science in 2008 at UBC. That same year, he entered the Bioinformatics Training Program with a scholarship from the CIHR. In the program, he completed rotations at the CMMT and the European Bioinformatics Institute in Hinxton, UK. In 2010, he joined the Shah and Aparicio labs while holding an NSERC Graduate Award. Gavin’s research involves the development and application of computational algorithms to profile structural aberrations in high-density microarrays and next-generation sequencing data of breast and ovarian cancer genomes.


Document Actions