[BioC] heatmap.2 Z-score vs logFC

[Sean Davis]

Hi, Neel.

Some rows have larger logFC than others.  Without scaling, those rows
will be brighter than the rows that have smaller logFC.  Scaling the
rows (columns) makes the rows (columns) have similar variation so that
even rows that have smaller logFC still have visible changes when
plotted.  If most of your rows (columns) have similar variation, then
scaling will not make much difference.

Sean

On Tue, Oct 18, 2011 at 11:33 AM, Neel Aluru <naluru at whoi.edu> wrote:
> Hello Bioc Users,
>
> I am having a hard time understanding why heatmap.2 uses row (or column)
> Z-score instead of directly plotting logFC values?  Based on the formula for
> calculating Z-score, I understand that it is used for scaling or smoothing.
> But, isn't it the same when using logFC. Please correct if I am wrong. I
> really appreciate if anyone can give me a clear explanation.
>
> Sorry for asking a silly question. I
>
> --
> Neel Aluru
> Visiting Investigator
> Biology Department
> Woods Hole Oceanographic Institution
> 45 Woods Hole Road
> Woods Hole 02543
> Massachusetts
> USA
>
> _______________________________________________
> Bioconductor mailing list
> Bioconductor at r-project.org
> https://stat.ethz.ch/mailman/listinfo/bioconductor
> Search the archives:
> http://news.gmane.org/gmane.science.biology.informatics.conductor
>