CEA, LITEN,
Laboratoire des systèmes solaires (L2S).
50 av. du Lac Léman,
73375 LE BOURGET DU LAC - CEDEX

 


Lespinats, S., Meyer-Bäse, A., He, H., Marshall, A.G., Conrad, C.A. et Emmett, M.R. "Novel insights into the lipidome of glioblastoma cells based on a combined PLSR and DD-HDS computational analysis." Proceedings of Society of Photo-Optical Instrumentation Engineers Conference Series, SPIE 2009, Vol. 7347, Avril 2009, Orlando, USA.


Résumé :
Partial Least Square Regression (PLSR) and Data-Driven High Dimensional Scaling (DD-HDS) are employed for the prediction and the visualization of changes in polar lipid expression induced by different combinations of wild-type (wt) p53 gene therapy and SN38 chemotherapy of U87 MG glioblastoma cells. A very detailed analysis of the gangliosides reveals that certain gangliosides of GM3 or GD1-type have unique properties not shared by the others. In summary, this preliminary work shows that data mining techniques are able to determine the modulation of gangliosides by different treatment combinations.