Skip to main content
Figure 1 | BMC Bioinformatics

Figure 1

From: Pathway activity inference for multiclass disease classification through a mathematical programming optimisation framework

Figure 1

Schematic flow chart of the DIGS-based approach for multiclass disease classification problems. Pathway specific gene expression profiles are created by integrating gene expression profile and pathway information. For each pathway, build pathway activity as a weighted (variables) linear summation of expression of member genes, with the objective function maximising the number of samples whose pathway activity are inside the range of their own classes. The maximum number of member genes in a pathway allowed to have non-zero weights is explicitly constrained in the model by specifying the parameter NoG. Create pathway activity profile by assembling all pathway activities and a classifier is trained on the pathway activity profile and predicts the class label of a new sample. It is important to note that training procedure, i.e., inferring pathway activity and training a classifier, is always blind to testing samples to achieve an objective evaluation of classification performance.

Back to article page