Bonferroni?

NCurse · 11-06-2009, 09:04 AM

I would seriously need some help regarding a statistical problem.

We measured gene expression levels of several genes in a set of control samples (cc. 15) and other sets of diseased groups as well. We analyzed the gene expression levels gene by gene in Prism by performing Mann-Whitney U test in order to identify significant differences in the levels of gene expression between control samples and diseased samples (disease by disease).

Now the reviewer told us we should perform Bonferroni adjustement (quite common requirement in microarray studies where there are >20,000 probes, but we used Taqman Low Density Array with only 96 genes on it).

Could someone please help me with this?

Thank you in advance!

terzi · 11-07-2009, 05:10 PM

Since its use would increase the probability of a Type II error, The Bonferroni adjustment is only recommended under certain situations, since is only useful to test a general hypothesis. I hope this link helps:

Code:

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1112991/

NCurse · 11-08-2009, 08:33 AM

Thank you, Terzi, for your answer. I still cannot prove that I don't have to use Bonferroni, but this link helps a lot.

Berci

11-06-2009, 09:04 AM	#1
NCurse Junior Member Join Date: Nov 2009 Posts: 2 Thanks: 0 Thanked 0 Times in 0 Posts	Bonferroni? I would seriously need some help regarding a statistical problem. We measured gene expression levels of several genes in a set of control samples (cc. 15) and other sets of diseased groups as well. We analyzed the gene expression levels gene by gene in Prism by performing Mann-Whitney U test in order to identify significant differences in the levels of gene expression between control samples and diseased samples (disease by disease). Now the reviewer told us we should perform Bonferroni adjustement (quite common requirement in microarray studies where there are >20,000 probes, but we used Taqman Low Density Array with only 96 genes on it). Could someone please help me with this? Thank you in advance!

11-07-2009, 05:10 PM	#2
terzi Senior Member Join Date: Sep 2009 Location: Xalapa, Mexico Posts: 102 Thanks: 0 Thanked 22 Times in 22 Posts	Since its use would increase the probability of a Type II error, The Bonferroni adjustment is only recommended under certain situations, since is only useful to test a general hypothesis. I hope this link helps: Code: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1112991/ __________________ Have a nice day!

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search
Display Modes
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode

11-08-2009, 08:33 AM	#3
NCurse Junior Member Join Date: Nov 2009 Posts: 2 Thanks: 0 Thanked 0 Times in 0 Posts	Thank you, Terzi, for your answer. I still cannot prove that I don't have to use Bonferroni, but this link helps a lot. Berci