[BioC] Statistical significance of clinical information

[Daniel Brewer]

Hi,

I have microarray data from a range of tumours.  With each of these
tumours I have some associated clinical information, including test
results/scores, patient characteristics (e.g. age) and other biochemical
tests (FISH etc.).  I would like to do a number of things:
1) Determine whether any of these have a significant effect on the
overall expression of the tumour.
2) Determine which genes expression correlate the best with a particular
characteristics/test result

This seems to the sort of thing you could do with linear models and
limma, but it was unclear to me what is the best way forward.  In
particular limma seems to be set up for the experimental design rather
than other external factors.  Can limma take into account continuous
variables rather than catergories?

Any advice on this would be gratefully received as I find this whole
area a bit confusing.

Dan

-- 
**************************************************************

Daniel Brewer, Ph.D.

Institute of Cancer Research
Email: daniel.brewer at icr.ac.uk

**************************************************************

The Institute of Cancer Research: Royal Cancer Hospital, a charitable Company Limited by Guarantee, Registered in England under Company No. 534147 with its Registered Office at 123 Old Brompton Road, London SW7 3RP.

This e-mail message is confidential and for use by the addre...{{dropped}}