Our work on a novel integration method (WON-PARAFAC) is published in Nature Communication

My last work during the postdoc time @ NKI published in Nature Communication, entitled:

Genomic data integration by WON-PARAFAC identifies interpretable factors for predicting drug-sensitivity in vivo

The method uses a cube-representation of multiple genomics data (gene by sample by data types) and apply a variant of parallel-factor analysis (PARAFAC), named WON-PARAFAC.

WONPARAFAC summarizes multiple genomics data of cell lines using interpretable factors. These factors are invariant over model systems and capable of reconstructing PDX genomics data; moreover, they allow for response prediction and translate predictive drug-response information from cell lines to PDXs.

You can find the source code in MATLAB at the Wessel’s group website (NKI) and also at the Github repository.