The ADVOCATE algorithm uses a Machine Learning based algorithm to combine weak evidence derived from the expression of individual genes into a model that provides an optimal estimate of the compartment-specific composition of a heterogeneous bulk tissue.
Reference:
H. Carlo Maurer*, Sam R. Holmstrom*, Jing He*, Pasquale Laise, Tao Su, Aqeel Ahmed, Hanina Hibshoosh, John A. Chabot,Paul E. Oberstein, Antonia R. Sepulveda, Jeanine M. Genkinger, Jiapeng Zhang, Alina C. Iuga, Mukesh Bansal, Andrea Califano and Kenneth P. Olive "Experimental mi- crodissection enables functional harmonisation of pancreatic cancer subtypes". GUT 2019
∗ These authors contributed equally to this work.