User: cms72

cms720
Reputation:
0
Status:
New User
Last seen:
1 year, 10 months ago
Joined:
1 year, 11 months ago
Email:
c****@student.le.ac.uk

Profile information, website and location are not shown for new users.

This helps us discourage the inappropriate use of our site.

Posts by cms72

<prev • 7 results • page 1 of 1 • next >
1
7.3k
views
1
... And finally, one last question before my analysis is complete! edgeR's glmTreat is equivalent to Limma's treat, and both use Bayes method. However I noticed only Limma uses the adjust method "BH", which calculates an adj P-value. glmTreat does not provide any adj. p-value in its output. Would that ...
written 23 months ago by cms720
1
7.3k
views
1
... Hi Aaron, I'm trying to compare my logFC results to results that are in FC to see if I achieved the same gene expressions as they did, like if gene A is down-regulated in theirs, if its down-regulated in mine too.  I know the computation would be table$FC<- 2^(abs(table$logFC)). But then everyt ...
written 23 months ago by cms720
1
7.3k
views
1
... Hi Aaron,  Just a quick question, would the equivalent of topTreat in Limma be glmTreat in edgeR?  Thank you again in advance. ...
written 23 months ago by cms720
1
7.3k
views
1
... I see. Thank you again. Regarding voom, the reason why I never used it was because I have FPKM values and not read counts (I know its not ideal, but thats what I've got to work with. And yes, I have to use limma specifically and not cummerbund). My understanding is that voom normalises counts, wher ...
written 23 months ago by cms720
1
7.3k
views
1
... In addition to my earlier post. I was fiddling around with my results and found that if I added robust to both lmFit and Treat, it would result to more DEGs, with warnings saying small variances detected (I assume due to its robust regression): fit<-lmFit(Count, design, method = "robust") ..... ...
written 23 months ago by cms720
1
7.3k
views
1
... Hi Aaron, Thank you for your time and clarification. Your explanation makes sense. These are the steps I've undertaken.. I've followed the Limma manual, but I feel like my comprehension could be wrong. fit<-lmFit(Count, design, method = "robust") cont.matrix<-makeContrasts(Treatmentvscontrol ...
written 23 months ago by cms720
1
7.3k
views
5 follow
1
... Hi, I was wondering if I could ask for confirmation regarding my understanding of logFC values. So I know when you get a negative logFC value, the gene is underexpressed, and when you get a positive logFC value, the gene is overexpressed. But is there a cutoff value? I read in an article that they ...
written 23 months ago by cms720 • updated 23 months ago by Aaron Lun24k

Latest awards to cms72

Popular Question 23 months ago, created a question with more than 1,000 views. For LogFC: how do you determine the cutoff for differentially expressed genes?

Content
Help
Access

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.