Almac Sitemap

Signature generation

The identification of gene signatures of response involves the identification of a collection of transcripts whose pattern of expression predicts the response to a therapeutic agent (predictive signature) or indicates disease outcome (prognostic signature).

This approach can lead to the development of a test which can be brought forward to regulatory agencies for approval and marketed as a test or companion diagnostic for a therapeutic agent.

Large datasets are typically required for the generation of such signatures and experimental study design is of critical importance. The experiment must be balanced to ensure that the design reflects the disease in terms of histology, gender, ethnicity of the patients and the subtypes of the disease.

In the case of predictive signatures, sufficient numbers of responders and non-responders are required with each group being balanced for all confounding variables.

Gne signatures of response

Our Approach:

Almac Diagnostics' bioinformaticians have vast experience in both binary and multi-class classification and will adopt the most suitable algorithms and cross-validation methods to develop a signature that best predicts class membership.
We will work with you to identify the relevant variables to aid in the correct design of your study or clinical trial.
We will analyse and work with your pre-clinical data to facilitate the identification of effective and validated signatures.
We will create a test set in order to generate a classifier. This will be validated on the independent test set to determine the specificity, sensitivity, negative and positive predictive values.
Based on this we can move forward with the development of a companion diagnostic, suitable for regulatory approval.