User: Federico Lasa

gravatar for Federico Lasa
Reputation:
80
Status:
New User
Location:
Last seen:
3 years, 3 months ago
Joined:
4 years, 2 months ago
Email:
f*****@gmail.com

Posts by Federico Lasa

<prev • 8 results • page 1 of 1 • next >
0
votes
2
answers
489
views
2
answers
Comment: C: Any problem with the design matrix or the contrasts? Thanks
... The difference is that chip is treated as a numeric variable in one case and as a factor variable in the other. I believe the second case is more aproppiate. On Aug 8, 2014 4:03 PM, "Rao,Xiayu" wrote: > Thank you for your kind suggestion, Federico! > > I tried your suggested design, and f ...
written 3.3 years ago by Federico Lasa80
0
votes
2
answers
489
views
2
answers
Answer: A: Any problem with the design matrix or the contrasts? Thanks
... Both a question and a quick suggestion.. For the first case, should/does design <- model.matrix(~0+AR +gender +chip, data=targets) cm<-makeContrasts(ARpos-ARneg, levels=design) produce the same results you have? On Fri, Aug 1, 2014 at 10:07 AM, Rao,Xiayu wrote: > Hello, > > I lea ...
written 3.3 years ago by Federico Lasa80
0
votes
1
answers
445
views
1
answers
Comment: C: finding differential expression pattern between paired columns of data frame.
... You could create a statistic that captures what you want. i.e: s = LineAD / (abs(LineBC) ) this number would be big if lineAD foldchange is big and/or line BC is small. It's not perfect but may somewhat capture what you want. You can alter it as well to prioritize something particular, like: s = ...
written 3.3 years ago by Federico Lasa80
0
votes
1
answers
573
views
1
answers
Answer: A: RMA Normalization Nimblegen with adf.txt file
... Hr Piet, I'm not familiar with adf file format but one thing you can do if you have the raw .xys files is search for the paricular platform in GEO and download the corresponding design (ndf) file from there, for instance: http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GPL17894 has downloads fo ...
written 3.4 years ago by Federico Lasa80
0
votes
2
answers
911
views
2
answers
Comment: C: converting probeset id to gene id's : fRMA
... Another thing to consider is the platform you are using, it might be relevant since some Affy chips (i.e. HuGene 1.0 ST) can be summarized in different ways (probeset or core if i recall correctly). On Sun, Jun 22, 2014 at 11:14 PM, Peter Langfelder wrote: > On Fri, Jun 20, 2014 at 5:05 PM, Abh ...
written 3.4 years ago by Federico Lasa80
0
votes
2
answers
397
views
2
answers
Answer: A: Bioconductor Digest, Vol 128, Issue 7
... Hi, No, the "2" is intentional. It is a table that 'maps' affy probe ids TO (sounds like two) gene symbols. I'm just guessing here, but i think that you need to install and load the hgu133plus2.db package. try running: source("http://bioconductor.org/biocLite.R") biocLite("hgu133plus2.db") libra ...
written 3.9 years ago by Federico Lasa80
0
votes
1
answers
694
views
1
answers
Answer: A: Analysis of Affymetrix Human Gene 2.0 ST arrays
... Hello, I also have doubts regarding this platform (Gene ST v1 1.0 in my case) so I'm gonna hop in with some questions and problems I'm encountering. I also had (a lot of) control/intron probes popping in the topTable of differentially expressed genes and while I realize i could just remove them, i ...
written 4.0 years ago by Federico Lasa80
0
votes
1
answers
630
views
1
answers
Answer: A: how to calculate logFC in limma
... If your data is already log transformed then the logFC (log of fold change not fold change of the logs) is computed as ave_prog - ave_stable. I believe this is what limma does. On Mon, Sep 9, 2013 at 11:13 PM, Wang Peter wrote: > this is output of limma > > ## ...
written 4.2 years ago by Federico Lasa80

Latest awards to Federico Lasa

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: 148 users visited in the last hour