April 10, 2011 - Update

We have now submitted a new, much improved version of ISOLATE for publication, called ISOpure. ISOpure is able to perform per-sample purification for tumor gene expression profiles (as well as other array and sequenced based data), and will be available soon as an R package for download. Please check back for updates periodically, or send your email address to me so I can send you an email when it's ready. I can be reached at gerald(dot)quon(at)utoronto(dot)ca .

Overview

We developed ISOLATE, a statistical tool that addresses three problems simultaneously:

  • Identification of the site of origin of carcinomas of unknown primary origin
  • Purify tumor samples of infiltrating healthy cells
  • Identify differentially expressed genes in the carcinoma, with respect to healthy cells

    • The software is available for download below.

    We have had several requests and inquiries as to whether ISOLATE can return purified profiles on a per-patient basis. The current version of ISOLATE cannot do this, and to the best of our knowledge, no method currently exists that has been proven to be able to do so, either. In short, existing purification methods assume all heterogeneous tumour profiles used as input, are assumed to contain the same cancer cells (mixed among infiltrating healthy cells in different proportions). We are working on a version to purify individual profiles, and plan to release it in early 2011. Stay tuned!

    Contact

    gerald.quon @ utoronto.ca ; quaid.morris @ utoronto.ca

    Paper download

    [abstract, pdf]