Clustering in R

I have a file from a microarray experiment. The data is normalised and we are looking to find out which genes are differentially expressed between the groups. We have two groups and the data is nearly normally distributed, so we have performed a t-test for each gene. The p-values are in the last column; they are not corrected for multiple testing.

   V1               V2               V3               V4               V5               V6
1  Names    GSM123264.cel    GSM123265.cel    GSM123266.cel    GSM123267.cel    GSM123268.cel
2   DDR1  9.9278082964302 10.2686357358346 9.75539843223479 10.4895541740877 10.1249575001857
3   RFC2 5.36386267697312 5.08142307160792 5.22893132183885 5.23837679546089  5.1280831260327

How can I calculate how many genes have an FDR <=0.05? Also, how can I cluster the data to see where the genes discriminate between the groups? Thank you!

genes cluster microarray r fdr

• 420 views • link

updated 3.9 years ago by Ram

• written 11.0 years ago by espop23

Table looks like expression values per CEL files only, no p-values. Check "affy" and "limma" packages in Bioconductor.

• 0 views • link

written 11.0 years ago by JC

For clustering I would recommend "cluster 3" and then "Java Tree View" which are very decent programs. If you would like to do it in R, you should explore heatmap.2 function.

• 0 views • link

written 11.0 years ago by GouthamAtla

Looks like the first row of your data should be the column headers. Import the data with read.table("file.txt",header=TRUE). If you want to identify differentially expressed genes, use the limma package. Then use heatmap.2 or better, heatmap.3 on differential genes

• 0 views • link

written 11.0 years ago by Irsan

0 answers

No answers yet.