## User: Koen Van den Berge

Reputation:
170
Status:
Trusted
Location:
Ghent University, Belgium
Last seen:
4 months, 1 week ago
Joined:
5 years, 5 months ago
Email:
k***********@gmail.com

#### Posts by Koen Van den Berge

<prev • 34 results • page 1 of 4 • next >
1
100
views
1
... Hi Sophia, The message about the removal of NA p-values is simply there for your information to check how many features have been removed prior to the analysis. Note that, even though you set allowNA=TRUE, nothing can be said about NA p-values, so they must be removed. Hence you can safely contin ...
written 4 months ago by Koen Van den Berge170
1
156
views
1
... Yes, the different annotations have been a problem for stageR, and I apologize for this inconvenience. This should be fixed in the following major upgrade of the package. Let us know, on the BioC support site, if you have any additional questions. ...
written 5 months ago by Koen Van den Berge170
1
156
views
1
... Yes, the different annotations have been a problem for stageR, and I apologize for this inconvenience. This should be fixed in the following major upgrade of the package. Let us know, on the BioC support site, if you have any additional questions. ...
written 5 months ago by Koen Van den Berge170
1
183
views
1
... Yes, two stageR analyses would be appropriate in that case, and you are correct that according to Shaffer's procedure, you are allowed to not adopt a FWER correction in the confirmation stage, so indeed method="none" will be valid here. ...
written 5 months ago by Koen Van den Berge170
1
183
views
1
... If you would want to analyze both timepoints and all treatments in a single stage-wise analysis, you would have two options: - A three-stage analysis: (i) Global screening test to assess differences between any of the treatments in any of the timepoints. (ii) Conditional on rejecting (i) you may p ...
written 6 months ago by Koen Van den Berge170
1
662
views
1
... Thank you for checking this. In the zinbFit function, could you specify your treatment design as well? This will allow the zero excess probability to systematically vary with the treatment. Also, we recommend setting commondispersion=TRUE. In the meanwhile, I would be happy to take a look at the da ...
written 11 months ago by Koen Van den Berge170
1
662
views
1
... I agree with Mike's suggestions, and would check those first. In addition, could you also let us know how the weights look like for the zeros of those specific genes? Can you share some of your code, e.g. did you include the treatment effect in the ZINB-WaVE model? Note that in our simulations, we ...
written 11 months ago by Koen Van den Berge170
1
277
views
1
... Hi Fiona, I found out that the error stems back from the gene-level p-values that do not carry the names of the genes that they are representing. In your case, the error can be solved by setting names(pScreen)=res[[2]]\$gene_id before inputting pScreen in stageR. I will now make sure to give a c ...
written 12 months ago by Koen Van den Berge170
1
277
views
1
... I am not yet very familiar with DRIMSeq, but it would be great if you could share your d object after running DRIMSeq. From browsing through your code it seems like that should contain most of what I need to run stageR and check where things went wrong. You could for example send it to koen.vanden ...
written 12 months ago by Koen Van den Berge170
1
277
views
1
... Thank you for reporting this. There used to be a problem with getAdjustedPValues in the older versions of stageR. It would be great if you could update your version with ​devtools::install_github("statOmics/stageR") and check if the error still occurs. Would you mind sharing (some of) your dat ...
written 12 months ago by Koen Van den Berge170

#### Latest awards to Koen Van den Berge

Popular Question 8 months ago, created a question with more than 1,000 views. For LIMMA similarity of contrast matrix, F test and topTable
Teacher 8 months ago, created an answer with at least 3 up-votes. For A: Correct use of Nested/Stage-wise testing with limma/stageR for 3-group design.
Scholar 8 months ago, created an answer that has been accepted. For A: Correct use of Nested/Stage-wise testing with limma/stageR for 3-group design.
Scholar 8 months ago, created an answer that has been accepted. For A: stageR NA in getAdjustedPValues() (method=dtu)
Popular Question 8 months ago, created a question with more than 1,000 views. For Hisat2 vs featureCounts summary statistics
Popular Question 2.1 years ago, created a question with more than 1,000 views. For convert S4 DataFrame of Rle objects to sparse matrix
Popular Question 2.1 years ago, created a question with more than 1,000 views. For Hisat2 vs featureCounts summary statistics
Popular Question 3.3 years ago, created a question with more than 1,000 views. For convert S4 DataFrame of Rle objects to sparse matrix
Popular Question 3.3 years ago, created a question with more than 1,000 views. For EdgeR LogCpm and LogFC values calculation

Content
Help
Access

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