Computational statistician/Post Doc/Research Associate

Position

We have an immediate opening for an individual with an interest and background in computational statistics to lead the development of automated approaches for the analysis of high throughput flow cytometry data and apply these to the analysis of clinical data to aid cancer diagnosis and biomarker discovery.

Background

Flow cytometry plays a critical role in basic research and clinical therapy in the areas of cancer, vaccine research, and detecting changes in immune response, to name but a few areas of application. There are many exciting new technological developments that will make flow cytometry data more widely used. However, the increased complexity greatly increases the need for reliable statistical methods, and accompanying software implementations, for the analysis of these data. Indeed many of the leaders in this field have explicitly stated that the largest problem facing them is the analysis of the data. We propose the development of a comprehensive and coordinated set of statistical methods that directly address many important problems.

Many current practices (e.g., gating -- the identification of homogenous populations of cells) require substantial manual intervention, and data analysis has become a rate-limiting step in the application of flow cytometry. Building on previous work that provides substantial software infrastructure, including data representation, transformation, visualization and quality assessment, this application provides a number of specific proposals to overcome these bottlenecks and to provide a sound computational and statistical paradigm that will support current and future needs.

Dependant on the applicant’s skills and interests, and in collaboration with other members of the Brinkman group and our collaborators, they will have the opportunity to be involved in several different components of this effort. The applicant will be involved in the development statistical methods suitable for the preprocessing of large scale flow cytometry data sets. Like all other high throughput genomic data sources, there will be substantial need for normalization and other preprocessing steps in order to separate the noise introduced by varying experimental conditions from the true, important biological variation between the samples. This work could include the development of statistical methods suitable for the automatic discovery and identification of subpopulations of cells. Once appropriate preprocessing has been carried out there will be a need for automatic gating strategies to identify known populations of cells, or to discover new populations. The applicant will also have the opportunity to be involved in the development of supervised and unsupervised analysis methods for the identification of different cell populations for a variety of applications including cancer, allergy, immunity, transplantation and HIV/AIDS. The applicant could also apply their skills towards the development of statistical methods to support the analysis, visualization and quantitative comparison of flow cytometry data from different samples or patients. The basic, and simplest problem is the comparison of data from two different samples. However, different experimental paradigms, such as longitudinal data, clinical trials data, or data in a 384 well plate format, will require specialized statistical methods. Finally, and perhaps most importantly, carrying out the specific analyses on appropriate data is an essential component of the process for developing relevant statistical methods and software implementations. The applicant will play a key role in applying these approaches to leukemia and lymphoma high throughput flow cytometry datasets. In order to ensure that the methods being developed are appropriate and useful the applicant will collaborate with several different cancer groups nationwide to validate and further develop appropriate statistical methods and software tools. The goal of this final component is to translate the research components into practice through close collaboration with our clinical partners, directly impacting patient care.

Qualifications and Experience:

We expect qualified candidates will have a degree in statistics or computer science and at least two years experience in software development. Experience with R is highly preferred.

Position Details:
Position is available immediately and will be for a minimum of two years. Applicant will hold a position in the Terry Fox Laboratory (http://www.terryfoxlab.ca/) at the BC Cancer Agency in Vancouver, BC, Canada.

To Apply:

Please email your cover letter and resume to rbrinkman@bccrc.ca.
Only short listed candidates will be contacted for interviews.