User: Aaron Lun

gravatar for Aaron Lun
Aaron Lun14k
Reputation:
13,620
Status:
Trusted
Location:
Cambridge, United Kingdom
Scholar ID:
Google Scholar Page
Last seen:
4 hours ago
Joined:
2 years, 6 months ago
Email:
a***@wehi.edu.au

I am a research associate working in the field of computational biology at the Cancer Research UK Cambridge Institute in the United Kingdom. I am the author and maintainer of the csaw, diffHic, InteractionSet and scran packages, a contributor and co-maintainer for the edgeR package, and an occasional contributor to the limma and scater packages.

Posts by Aaron Lun

<prev • 1,490 results • page 1 of 149 • next >
0
votes
1
answers
27
views
1
answers
Answer: A: Single cells batch effects
... I would say that RUV is not the appropriate tool here. RUV(seq) is designed for detecting unwanted factors of variation. But in this case, you know the factor of variation - the batch/experiment in which each cell was processed. There's not much point running RUVseq to recover something that you alr ...
written 4 hours ago by Aaron Lun14k
0
votes
1
answers
28
views
1
answers
Comment: C: Is there any point to TMM-normalize TPM prior to limma-voom?
... FYI, the precision weights from TMM are calculated from a binomial distribution, and won't make much sense for non-count data. It probably doesn't do much harm, either, but it can be turned off with doWeighting=FALSE. ...
written 5 hours ago by Aaron Lun14k
0
votes
1
answers
104
views
1
answers
Comment: C: limma: how to use pairing and adjust for covariates
... Male.post represents the log-fold change of post over pre in males. Female.post represents the log-fold change of post over pre in females. The contrast Female.post - Male.post tests the null hypothesis that the post/pre log-fold changes are the same between males and females. If this is not clear, ...
written 6 hours ago by Aaron Lun14k
0
votes
1
answers
60
views
1
answers
Comment: C: limma blocking and including covariates
... It's a bad idea to shop around for the analysis strategy that gives you the most DE genes. As I mentioned before, duplicateCorrelation is somewhat liberal and makes a number of assumptions. How do you know that the greater number of DE genes is real, and not just some consequence of loss of type I e ...
written 7 hours ago by Aaron Lun14k
0
votes
1
answers
60
views
1
answers
Comment: C: limma blocking and including covariates
... Ryan is mostly right. However, duplicateCorrelation assumes that the patient effects are homogeneously distributed. This will not be true if there are systematic factors of variation across patients. To illustrate, let's say with have 5 male and 5 female patients, with a pair of treated/untreated sa ...
written 8 hours ago by Aaron Lun14k
0
votes
2
answers
143
views
2
answers
Comment: C: remove batch effect and adjusted model for 4 covariates in Limma
... You are missing the point. It doesn't matter that the patients are in different batches. If both samples for any given patient belong to the same batch, the patient blocking terms will absorb any batch effect. Say all samples in the first batch have increased expression - this will be modelled as la ...
written 9 hours ago by Aaron Lun14k
0
votes
2
answers
143
views
2
answers
Comment: C: remove batch effect and adjusted model for 4 covariates in Limma
... I meant to move your comment as a response to my first answer, via the "Add comment" button. But never mind. Anyway, there are two problems here. The first problem - and the cause of the unestimable coefficients - is that your patients are fully nested in your batch. This is probably because both s ...
written 11 hours ago by Aaron Lun14k
0
votes
2
answers
143
views
2
answers
Comment: C: remove batch effect and adjusted model for 4 covariates in Limma
... Please move this latest comment to a response to my first post, the space is getting too small to answer it effectively here. ...
written 12 hours ago by Aaron Lun14k
0
votes
1
answers
104
views
1
answers
Comment: C: limma: how to use pairing and adjust for covariates
... I would assume so. Check the design matrix to be sure; you should see entries of 1 for post-treatment male samples, and zero for everything else. ...
written 1 day ago by Aaron Lun14k
0
votes
1
answers
104
views
1
answers
Comment: C: limma: how to use pairing and adjust for covariates
... The only difference (in the pruned designed matrices) is that, if you have an intercept, it represents the average log-expression in the first patient, and all other patient terms represent the log-fold change relative to the first patient. Without an intercept, each patient term directly represents ...
written 1 day ago by Aaron Lun14k

Latest awards to Aaron Lun

Scholar 12 hours ago, created an answer that has been accepted. For A: (batch) corrections of RNA-seq data: integrating LIMMA and SVA
Scholar 5 days ago, created an answer that has been accepted. For A: (batch) corrections of RNA-seq data: integrating LIMMA and SVA
Scholar 7 days ago, created an answer that has been accepted. For A: (batch) corrections of RNA-seq data: integrating LIMMA and SVA
Teacher 10 days ago, created an answer with at least 3 up-votes. For A: Limma-voom contrast confusion
Scholar 18 days ago, created an answer that has been accepted. For A: (batch) corrections of RNA-seq data: integrating LIMMA and SVA
Teacher 18 days ago, created an answer with at least 3 up-votes. For A: Limma-voom contrast confusion
Teacher 25 days ago, created an answer with at least 3 up-votes. For A: Limma-voom contrast confusion
Good Answer 26 days ago, created an answer that was upvoted at least 5 times. For A: Using limma when continuous and categorical confounders are present at the same
Appreciated 26 days ago, created a post with more than 5 votes. For A: Combining newer/older RNAseq data, batch correcting
Teacher 4 weeks ago, created an answer with at least 3 up-votes. For A: Limma-voom contrast confusion
Scholar 4 weeks ago, created an answer that has been accepted. For A: (batch) corrections of RNA-seq data: integrating LIMMA and SVA
Scholar 5 weeks ago, created an answer that has been accepted. For A: Run windowCounts only on annotated regions?
Teacher 5 weeks ago, created an answer with at least 3 up-votes. For A: Limma-voom contrast confusion
Scholar 5 weeks ago, created an answer that has been accepted. For A: Can featurecounts count number of mapped reads rapidly in arbitrary regions?
Teacher 6 weeks ago, created an answer with at least 3 up-votes. For A: Limma-voom contrast confusion
Teacher 7 weeks ago, created an answer with at least 3 up-votes. For A: Discrepancy in the output of decideTests (total not equal to sum of up/down gene
Teacher 7 weeks ago, created an answer with at least 3 up-votes. For A: Discrepancy in the output of decideTests (total not equal to sum of up/down gene
Scholar 7 weeks ago, created an answer that has been accepted. For A: Can featurecounts count number of mapped reads rapidly in arbitrary regions?
Teacher 8 weeks ago, created an answer with at least 3 up-votes. For A: Discrepancy in the output of decideTests (total not equal to sum of up/down gene
Scholar 8 weeks ago, created an answer that has been accepted. For A: Can featurecounts count number of mapped reads rapidly in arbitrary regions?
Appreciated 8 weeks ago, created a post with more than 5 votes. For A: Combining newer/older RNAseq data, batch correcting
Good Answer 8 weeks ago, created an answer that was upvoted at least 5 times. For A: limma Time Course Experiment with many time points
Scholar 8 weeks ago, created an answer that has been accepted. For A: Can featurecounts count number of mapped reads rapidly in arbitrary regions?
Teacher 8 weeks ago, created an answer with at least 3 up-votes. For A: Discrepancy in the output of decideTests (total not equal to sum of up/down gene
Teacher 11 months ago, created an answer with at least 3 up-votes. For A: deseq2, edger different number of replicates and extraction kit

Help
Access

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 2.2.0
Traffic: 136 users visited in the last hour