RT @enniscath: Duh | Accounting for cellular heterogeneity is critical in epigenome-wide association studies http://t.co/pL04EePRos #genome…
@LastBehaviorist @Ersant0s @PBaboins @kareem_carr @EricRWeinstein @ConceptualJames These sorts of issues are constantly lurking in many fields. Frankly, I think it is an illusion that the merely having some experimental control eliminates the need to think
@MahmoodHacan @dan_marinazzo @matteo_brainnet @INM7_ISN Very easy for measurement process to interfere. Sometimes you might not be measuring this at all but capturing something else instead, that proportion of cell-types varies in the data, thus an artifac
@ee_reh_neh @trashystats I've been playing with the approach described in this paper (for methylation): https://t.co/GXlBSLeWXz to adjust for cell type composition. Nutshell: use CpGs that best discriminate cell types as negative control probes in an RUV D