Hi Jason,
The 90 additional samples only looked for mutations in a subset of 39 genes identified in the original 24 samples. The n=114 should only apply to the frequency calculations for those 39 genes. I couldn't find in their publication or supplementary material an exact list of the 39 genes, but based on the article, they would be a subset of the ~80 genes reported in Supplementary table 7. Through process of elimination, an example would be CTAG2. You'll have to contact the authors to determine exactly what 39 genes were measured across all 114 samples. (http://dcc.icgc.org/martreport/?report=gene_report&ensembl_gene_id=ENSG00000126890&datasets=hsapiens_gene_ensembl_tcgaBRCA) Thanks, Heather From: Jianxin Wang [mailto:[email protected]] Sent: Monday, March 21, 2011 12:52 PM To: Estrella, Heather; Elena Rivkin; [email protected] Subject: RE: [BioMart Users] ICGC BioMart: Pancreatic Cancer (JHU/US)dataset reference Heather, I'm not sure what you mean by saying "some of the genes were only tested in the 24 samples". Can you clarify? Jason ________________________________ From: Estrella, Heather [mailto:[email protected]] Sent: Monday, March 21, 2011 1:33 PM To: Jianxin Wang; Elena Rivkin; [email protected] Subject: RE: [BioMart Users] ICGC BioMart: Pancreatic Cancer (JHU/US)dataset reference Thanks Jason, It's misleading how Biomart reports some of the mutation frequencies as some of the genes were only tested in the 24 samples and not the full 114 so the frequency reported (x/114) is much less than the actual x/24 for that data set. Thanks, Heather From: Jianxin Wang [mailto:[email protected]] Sent: Monday, March 21, 2011 9:44 AM To: Estrella, Heather; Elena Rivkin; [email protected] Subject: RE: [BioMart Users] ICGC BioMart: Pancreatic Cancer (JHU/US)dataset reference Heather, Yes, the 114 samples are from their "discovery" studies (24 samples) and "prevalence" studies (the rest of them). Jason ________________________________ From: [email protected] [mailto:[email protected]] On Behalf Of Estrella, Heather Sent: Monday, March 21, 2011 11:38 AM To: Elena Rivkin; [email protected] Subject: Re: [BioMart Users] ICGC BioMart: Pancreatic Cancer (JHU/US) dataset reference Thanks Elena, That's the reference I was thinking the data was probably from. The total number of samples used for genetic analysis in this study was 24 not 114. Can you please look into correcting the summary stats so that the correct frequency is reported for mutations? In looking at the sample names for this study in BioMart, it looks like additional studies were added to the BioMart dataset. Maybe that's where the n-114 came from. However, that messes up the frequencies as not all the genes were tested in all 114 samples. Many thanks, Heather From: Elena Rivkin [mailto:[email protected]] Sent: Monday, March 21, 2011 9:14 AM To: Estrella, Heather; [email protected] Subject: Re: [BioMart Users] ICGC BioMart: Pancreatic Cancer (JHU/US) dataset reference Heather, These data were derived from the following publication: Core Signaling Pathways in Human Pancreatic Cancers Revealed by Global Genomic Analyses Siân Jones1,* et al. Science. 2008 September 26; 321(5897): 1801-1806. doi:10.1126/science.1164368. Elena From: "Estrella, Heather" <[email protected]> Date: Mon, 21 Mar 2011 12:08:21 -0400 To: "[email protected]" <[email protected]> Subject: [BioMart Users] ICGC BioMart: Pancreatic Cancer (JHU/US) dataset reference Hi, In looking at the mutation frequencies reported for the Pancreatic Cancer (JHU/US) dataset, it reports the frequencies out of n=114. Can you please send me the reference to the source data behind what's reported for "Pancreatic Cancer (JHU/US)"? Thanks, Heather
_______________________________________________ Users mailing list [email protected] https://lists.biomart.org/mailman/listinfo/users
