TY - JOUR
TI - Pathway-based subnetworks enable cross-disease biomarker discovery
AU - Haider, Syed
AU - Yao, Cindy Q.
AU - Sabine, Vicky S.
AU - Grzadkowski,
AU - Michal
AU - Stimper, Vincent
AU - Starmans, Maud H. W.
AU - Wang, Jianxin
AU - and Nguyen, Francis
AU - Moon, Nathalie C.
AU - Lin, Xihui
AU - Drake,
AU - Camilla
AU - Crozier, Cheryl A.
AU - Brookes, Cassandra L.
AU - van de
AU - Velde, Cornelis J. H.
AU - Hasenburg, Annette
AU - Kieback, Dirk G. and
AU - Markopoulos, Christos J.
AU - Dirix, Luc Y.
AU - Seynaeve, Caroline and
AU - Rea, Daniel W.
AU - Kasprzyk, Arek
AU - Lambin, Philippe
AU - Lio', Pietro
AU - and Bartlett, John M. S.
AU - Boutros, Paul C.
JO - Nature Communications
PY - 2018
VL - 9
TODO - null
SP - null
PB - Nature Publishing Group
SN - 2041-1723
TODO - 10.1038/s41467-018-07021-3
TODO - null
TODO - Biomarkers lie at the heart of precision medicine. Surprisingly, while
rapid genomic profiling is becoming ubiquitous, the development of
biomarkers usually involves the application of bespoke techniques that
cannot be directly applied to other datasets. There is an urgent need
for a systematic methodology to create biologically-interpretable
molecular models that robustly predict key phenotypes. Here we present
SIMMS (Subnetwork Integration for Multi-Modal Signatures): an algorithm
that fragments pathways into functional modules and uses these to
predict phenotypes. We apply SIMMS to multiple data types across five
diseases, and in each it reproducibly identifies known and novel
subtypes, and makes superior predictions to the best bespoke approaches.
To demonstrate its ability on a new dataset, we profile 33 genes/nodes
of the PI3K pathway in 1734 FFPE breast tumors and create a
four-subnetwork prediction model. This model out-performs a
clinically-validated molecular test in an independent cohort of 1742
patients. SIMMS is generic and enables systematic data integration for
robust biomarker discovery.
ER -