User: Mark Robinson

gravatar for Mark Robinson
Mark Robinson870
Reputation:
870
Status:
Trusted
Location:
Last seen:
3 years, 3 months ago
Joined:
6 years, 1 month ago
Email:
m************@imls.uzh.ch

Posts by Mark Robinson

<prev • 84 results • page 1 of 9 • next >
0
votes
2
answers
1.1k
views
2
answers
Answer: A: edgeR: Problem with subsetting a DGEList in latest package version
... Hi Katja, Apologies for the slow response. I can reproduce this, thanks for sharing. Gordon or I will commit a fix shortly. In the meantime, try: rc <- readDGE(files, labels = labels) rc <- DGEList(rc$counts) Then, subletting should work. Cheers, Mark On 13.08.2014, at 23:37, Katja He ...
written 3.3 years ago by Mark Robinson870
0
votes
3
answers
582
views
3
answers
Comment: C: "True" normalization factors in edgeR
... Hi Tianyu, That's much clearer, thanks. In calNorm(), I changed: counts_edgeR<-counts[indNonDE,] de <- DGEList(counts = counts_edgeR, group = group ) de <- calcNormFactors(de,method="TMM") to: de <- DGEList(counts = counts, group = gro ...
written 3.3 years ago by Mark Robinson870
0
votes
3
answers
582
views
3
answers
Answer: A: "True" normalization factors in edgeR
... Hi Tianyu, Good questions get good answers. If you want my help, you should make it easier for me: 1. Answer my original questions. In particular, look at your data to see what normalization factors you've introduced. That may give you some indication of if you've done something wrong. 2. Inste ...
written 3.3 years ago by Mark Robinson870
0
votes
3
answers
582
views
3
answers
Answer: A: "True" normalization factors in edgeR
... Hi Tianyu, Sorry for the slow response (was away). I have read your attachment, but your question and simulation is still a little unclear to me. First, some useful diagnostics for me to see whether this simulation is in any way reasonable to a real RNA-seq dataset are an M-A plot -- plotSmear(), ...
written 3.3 years ago by Mark Robinson870
0
votes
1
answers
829
views
1
answers
Comment: C: total count filter cutoff
... In my lab, we typically follow a "CPM of at least X in at least Y samples" rule, where X=1 (arbitrary but reasonable, can be changed) and Y=size of smallest replicate group, according to one of the case studies in the user's guide, for example: ------ 4.3.6 Filtering We fi lter out very lowly expr ...
written 3.6 years ago by Mark Robinson870
0
votes
2
answers
567
views
2
answers
Comment: C: tagwise parameters for negative binomial distribution in edgeR
... Hi Davide, Just to add another option, there is also estimateGLMRobustDisp(), which is a wrapper for an iteratively re-weighted scheme that down weights outliers -- effectively using a constant prior degrees of freedom. This was developed independently of estimateDisp() from a different perspectiv ...
written 3.7 years ago by Mark Robinson870
0
votes
2
answers
843
views
2
answers
Answer: A: differential expression- edgeR
... Hi Ashutosh, Please keep these exchanges on the BioC mailing list. I've cc'd the list here. > But when I exactly follow the script from edgeR manual on the data and on making DGE list, I get library size as NA. This depends on how the data is stored. readDGE() expects read counts where absen ...
written 3.9 years ago by Mark Robinson870
0
votes
2
answers
843
views
2
answers
Answer: A: differential expression- edgeR
... Dear Ashutosh, I wasn't able to (nicely) get exactly the same files from GEO, but here is an alternative: See: http://www.ebi.ac.uk/arrayexpress/experiments/E-MTAB-71/samples/ f <- "ftp://ftp.ebi.ac.uk/pub/databases/microarray/data/experiment/MTA B/E-MTAB-71/E-MTAB-71.raw.1.zip" bf <- basen ...
written 3.9 years ago by Mark Robinson870
0
votes
3
answers
1.7k
views
3
answers
Answer: A: edgeR glmFit
... Hi Jahn, A few things: 1. Please don't email me directly. USE THE BIOC MAILING LIST. This is the third time I've mentioned this. I've cc'd it again. 2. You are using exactTest() with GLM-based dispersion estimation. This is probably ok, although a bit non-standard. Did you read and follow the ...
written 3.9 years ago by Mark Robinson870
0
votes
3
answers
1.7k
views
3
answers
Answer: A: edgeR glmFit
... Hi Jahn, I've cc'd the BioC mailing list; it's best to keep a public record of these exchanges, so others can benefit. So, I believe your problem is a matter of the class of the object that you are sending to estimateGLMCommonDisp(), etc.. Here is an example: > y <- matrix(rnbinom(1000,mu= ...
written 3.9 years ago by Mark Robinson870

Latest awards to Mark Robinson

No awards yet. Soon to come :-)

Help
Access

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 2.2.0
Traffic: 257 users visited in the last hour