Darlene Goldstein | 1 Dec 13:21 2004

Re: normalisation or analysis with batch effects

Hi, I just wanted to mention that even if you do normalize all the chips
together, you are still likely to see the 'batch' (or 'block') effects.  To try
to assess the extent of the problem, you might cluster the samples and see if
you get samples from the same batch clustering together.

Best regards, Darlene



the 11 tumour sampel are considered as biological replicates, or are these split
into different tumour classes?

We've had a similar problem. Our data was generated in three different
laboratories, each having slightly different protocols, but within each lab we
had the same factors (the same doses of a drug).

I guess, if the tumours are considered as replicates one could include the batch
as a factor (as you suggest below), but if they contain different tumour classes
one could not separate the dmso effect from the "tomour" class effect.

The tissues samples (normal and tumour) are probably from different subjects and
will show strong differences per se. Maybe one get some estimates for the impact
of the batch by using a mixed effects model with each sample as random effect
and the batch as fixed effect.

something like lme(response ~ batch, data=d, rand = ~ 1|sample)

(Continue reading)

James W. MacDonald | 1 Dec 17:42 2004

Re: Re: normalisation or analysis with batch effects

>-----Original Message-----
>From: bioconductor-bounces at stat.math.ethz.ch
>[mailto:bioconductor-bounces at stat.math.ethz.ch]On Behalf Of
>Sent: 30 November 2004 23:51
>To: BioConductor mailing list
>Cc: Andrea Pellagatti
>Subject: [BioC] normalisation or analysis with batch effects
>Dear list,
>If the following question has been asked before, I do apologise in
>advance and hope someone can point to the relevant thread. Otherwise I
>would appreciate some thoughts and pointers to this problem.
>Thank you.
>Problem : My collaborator (cc-ed here) has performed hybridisation for
>11 tumour and 40 normal samples on Affymetrix HGU-133Av2
>(contains ~55k
>probesets) chips. He had hybridised about half of the samples when he
>realised he needed more Affymetrix chips.
>The second batch of chips arrived with the instruction to add DMSO in
>the hybridisation cocktail, which he followed. The first batch did not
>have such instruction. Therefore we believe that the two
>batches are not

(Continue reading)