Precision Medicine DTP - Molecular dynamics of the response to breast cancer therapies

Background

Surgery is still the gold standard approach for treating cancer, most therapies are given after tumour excision, making it difficult to examine how cancer cells respond directly in patients. However, pre-surgical treatment is becoming more popular for breast cancer patients to down-stage tumours, so that less invasive operations can be performed and potentially improve predictions of which patients will respond and overall outcomes. Over the last 10 years the Sims group have profiled gene expression responses of breast tumours to a range of different treatments (1–3). Sequential samples collected at diagnosis, during treatment and at excision can be compared to examine changes in the tumour at the molecular level, minimizing patient-patient variation and increasing statistical power. These valuable dynamic datasets can be challenging to analyze due to their high dimensionality – where the large number of features change subtly over time in responding and non-responding patients. Further challenges arise due to the data being incomplete and noisy.

The random-projection ensemble classifier (4) is a general method for dealing with high-dimensional data, which has been shown to be effective in a wide range of applications. The technique can be seen as a way to extend existing low-dimensional methods to the high-dimensional setting.

Aims

Evaluate how modern statistical learning techniques, such as the random-projection ensemble classifier (4) can be used to accurately stratify breast cancer patients to the most appropriate treatment groups. This will include identifying the most important predictive features at different time points before and on-treatment associated with short-term response, long term outcomes and validation across datasets.

Training Outcomes

The student will receive training in a wide range of analysis techniques for making use of high-dimensional genomic data in a pragmatic way towards the benefit of cancer patients, comparing new biomarkers and signatures against established approaches, both pre- and on-treatment.

Data science methodologies will include the state-of-the-art classification and clustering techniques, including data perturbation and random projection approaches. The candidate will have the opportunity to mix with cancer researchers and mathematicians to develop highly desirable skills in applied machine learning techniques.

This MRC programme is joint between the Universities of Edinburgh and Glasgow. You will be registered at the host institution of the primary supervisor detailed in your project selection.

All applications should be made via the University of Edinburgh, irrespective of project location. For those applying to a University of Glasgow project, your application along with any supporting documents will be shared with University of Glasgow.

http://www.ed.ac.uk/studying/postgraduate/degrees/index.php?r=site/view&id=919

Please note, you must apply to one of the projects and you must contact the primary supervisor prior to making your application. Additional information on the application process is available from the link above.

For more information about Precision Medicine visit:
http://www.ed.ac.uk/usher/precision-medicine