Applying high throughput sequencing to cancer and human genetic diseases

Recent advances to DNA sequencing technology allow us to determine the entire human genome sequence of an individual in a single experiment. Such analyses offer exciting possibilities but the data present significant computational challenges. My lab uses a combination of high-throughput sequencing and bioinformatics to study the genetic architecture of cancer and other heritable and sporadic genetic diseases. Though many cancers are not heritable, they arise from the collection of a series of somatic mutations that arise throughout ones life. The "driver" mutations are those that alter the behavior of cells and contribute to malignancy. General aims of my research include the identification of mutations that offer new insights into disease etiology such as the driver mutations in cancer and causal alleles in other diseases. A secondary aim is to identify biomarkers that may be used to predict and follow the clinical course of cancers. Ultimately, I aim to discover mutations that may lead to new treatments for diseases such as cancer and improved methods for detecting tumour dynamics. My lab also studies common sporadic cancers affecting pet dogs (Canis familiaris). We hope that some new insights made in human cancers may facilitate improved treatment and diagnosis of canine cancers such as lymphoma.
To achieve each of these goals we need to produce massive amounts of sequencing data (genomes, exomes or transcriptomes) from patient samples and deep re-sequencing experiments of regions for the quantitative detection of mutations. Sophisticated analyses are required to separate the disease-related signal from normal biological variation and technical noise inherent in any sequencing technology. As data throughput increases and more challenging sample and experiment types are approached, my group will continue to develop algorithms and workflows to improve our ability to accomplish this. Distinct mutation types such as single nucleotide variants (SNVs), insertion/deletions, copy number variants (CNVs) and structural alterations can all act as drivers. My lab will pursue approaches to integrate these disparate mutation types to better understand the complete genetic architecture of individual cancers. As computational infrastructure is a commodity, I ultimately aim to produce cloud-ready tools that operate within Galaxy or some other user-friendly system such that individual researchers can more readily utilize these methods.

Specific research aims are as follows:
1) Determine the genetic events that lead to treatment resistance, relapse and metastasis in common human cancers with a focus on lymphoma and pediatric cancers
2) Develop sensitive assays for detecting the presence of tumour cells and key driver mutations in the bloodstream of patients
3) Identify commonalities and differences between common canine cancers and their human counterparts
4) Develop improved methods for detecting mutations in massively parallel sequencing data and integrating distinct mutation types to aid in identifying driver mutations

Laboratory trainees will have the opportunity to learn how to produce and analyze next generation sequencing (NGS) data using an Illumina MiSeq. I am open to trainees who desire a strictly wet-lab focus and those with purely bioinformatics projects and any blend of the two. Owing to my affiliation with the BC Cancer Agency's Genome Sciences Centre, students will have access to clinical collaborators and additional high throughput NGS instruments such as the HiSeq and IonTorrent Personal Genome Machine.

Students interested interested in a purely bioinformatics project may also apply to the CIHR/MSFHR Bioinformatics Training Program.

Page last modified Sep 04, 2015