Moderator: Michael Love

gravatar for Michael Love
Michael Love20k
Reputation:
19,890
Status:
Trusted
Location:
United States
Website:
http://mikelove.github...
Twitter:
mikelove
Scholar ID:
Google Scholar Page
Last seen:
11 hours ago
Joined:
5 years, 8 months ago
Email:
m****************@gmail.com

Michael I. Love, Dr. rer. nat.
Assistant Professor
Department of Biostatistics
Department of Genetics
University of North Carolina-Chapel Hill

The Love lab maintains the following R/Bioconductor packages:

the RNA-seq gene-level workflow, and the RNA-seq DTU workflow.

Posts by Michael Love

<prev • 4,075 results • page 2 of 408 • next >
0
votes
1
answers
32
views
1
answers
Answer: A: Using Wald test after lfcShrink (dds, contrast ())
... The fact that the base mean over all samples is low is not necessarily a problem. If you have four groups, and 4 out of 16 samples have a count of ~10 and the rest have 0, that gives you a base mean of 2.5, although there is some evidence here of DE. It depends on the distribution of counts, but I'd ...
written 2 days ago by Michael Love20k
0
votes
1
answers
62
views
1
answers
Comment: C: Question: Pair-wise DESeq Analysis with 3 Factor Levels - RNA-Seq
... To generalize, let me call NB, CW, and CWW just A, B and C, levels of "condition". You can compare C to A and B to A in one design, you don't need to rerun DESeq(), you just use the "contrast" argument of results(). That doesn't deal with the correlations among the cell type though. You can't accou ...
written 2 days ago by Michael Love20k
0
votes
1
answers
59
views
1
answers
Answer: A: what is the correct command line to import my txi data file as a deseq2 dataset
... The basic line is correct, but the design is not according to our time series example: https://bioconductor.org/packages/release/workflows/vignettes/rnaseqGene/inst/doc/rnaseqGene.html#time ...
written 2 days ago by Michael Love20k
0
votes
1
answers
24
views
1
answers
Answer: A: DESeq2/apeglm s-values: Why average FSR (or FSOS) instead of maximum?
... If you think about the expected number of binary events, that motivates the cumulative sum of the posterior probabilities. But also we were following precedent of some of the others working with a set of the smallest posterior probabilities. In the paper we have: "Other methods have suggested using ...
written 2 days ago by Michael Love20k
0
votes
2
answers
85
views
2
answers
Comment: C: Are spurious correlations possible in correlating two DESeq2 foldchanges compute
... Cool, didn’t know about that. ...
written 2 days ago by Michael Love20k
0
votes
2
answers
85
views
2
answers
Answer: A: Are spurious correlations possible in correlating two DESeq2 foldchanges compute
... Yes, under the null of no differences among A, B, and C, the standard MLE for the log2 of C vs A and B vs A will be positively correlated. I  think an LFC shrinkage method will reduce this correlation some but not entirely, because LFCs consistent with 0 for both comparisons will move closer to the ...
written 2 days ago by Michael Love20k
0
votes
1
answers
33
views
1
answers
Answer: A: Differential expression where only certain samples are paired
... The short answer is that we can’t do this with fixed effects but you can do it with limma and duplicateCorrelation(). There are some longer posts on here with this question but it’s difficult to dig them up based on limited search functionality. ...
written 3 days ago by Michael Love20k
0
votes
1
answers
62
views
1
answers
Comment: C: Why does DEseq2 take a long time to run on large dataset ?
... One more comment: no, it's not normal that every sample would have a 0 for every gene in bulk RNA-seq. There are housekeeping genes that you'd expect to have positive counts for all samples. There may be failed samples that are being included and should be excluded based on QC. I recommend using Mul ...
written 3 days ago by Michael Love20k
0
votes
1
answers
62
views
1
answers
Comment: C: Why does DEseq2 take a long time to run on large dataset ?
... On my machine, 400 samples takes 1.5 minutes per 1000 genes with one core. So that extrapolates for 15 minutes for 10,000 genes (probably you should be around this number if you filter out the unexpressed genes). > dds <- makeExampleDESeqDataSet(n=1000, m=400) > system.time({ dds <- DE ...
written 3 days ago by Michael Love20k
0
votes
1
answers
62
views
1
answers
Comment: C: Why does DEseq2 take a long time to run on large dataset ?
... Also while you’re speeding things up, you should remove the genes that don’t have sufficient counts. You can use for such a large sample size: keep <- rowSums(counts(dds) >= 10) >= n Where you might choose n of 10 for such a large dataset. Then subset dds before running DESeq() dds <- ...
written 3 days ago by Michael Love20k

Latest awards to Michael Love

Scholar 6 weeks ago, created an answer that has been accepted. For A: One-sided p-value from DESeq2
Scholar 6 weeks ago, created an answer that has been accepted. For A: Get release date for version of DESeq2
Scholar 6 weeks ago, created an answer that has been accepted. For A: ntop parameter in plotPCA(). DESeq2
Scholar 6 weeks ago, created an answer that has been accepted. For A: DESeq2 - Three Sample Analysis
Scholar 6 weeks ago, created an answer that has been accepted. For A: DESeq2 confounding cartridge
Teacher 6 weeks ago, created an answer with at least 3 up-votes. For A: How does DESeq2 handle zero counts in one condition?
Scholar 6 weeks ago, created an answer that has been accepted. For A: Analysis guidance for data inclusion in deseq2 analysis: data subsets vs contras
Scholar 6 weeks ago, created an answer that has been accepted. For A: Post-processing RNA-Seq after tximport
Scholar 6 weeks ago, created an answer that has been accepted. For A: How to set up appropriate analysis using sample type and condition for DESeq2
Teacher 6 weeks ago, created an answer with at least 3 up-votes. For A: DESeq2 baseMean values for each sample
Scholar 6 weeks ago, created an answer that has been accepted. For A: Should I use alpine and can I convert the output back to counts?
Scholar 6 weeks ago, created an answer that has been accepted. For A: DESeq2 on two groups
Teacher 6 weeks ago, created an answer with at least 3 up-votes. For A: What doest this p.value distribution mean?
Teacher 6 weeks ago, created an answer with at least 3 up-votes. For A: DESeq2 why estimate dispersion on log scale
Scholar 6 weeks ago, created an answer that has been accepted. For A: DESeq2 why estimate dispersion on log scale
Teacher 6 weeks ago, created an answer with at least 3 up-votes. For A: How to set up appropriate analysis using sample type and condition for DESeq2
Scholar 10 weeks ago, created an answer that has been accepted. For A: ntop parameter in plotPCA(). DESeq2
Scholar 10 weeks ago, created an answer that has been accepted. For A: DESeq2 - Three Sample Analysis
Scholar 10 weeks ago, created an answer that has been accepted. For A: One-sided p-value from DESeq2
Good Answer 10 weeks ago, created an answer that was upvoted at least 5 times. For A: difference among tximport scaledTPM, lengthScaledTPM and the original TPM output
Teacher 10 weeks ago, created an answer with at least 3 up-votes. For A: How does DESeq2 handle zero counts in one condition?
Teacher 10 weeks ago, created an answer with at least 3 up-votes. For A: DESeq2 baseMean values for each sample
Scholar 10 weeks ago, created an answer that has been accepted. For A: DESeq2 on two groups
Appreciated 3 months ago, created a post with more than 5 votes. For DESeq2 testing ratio of ratios (RIP-Seq, CLIP-Seq, ribosomal profiling)
Appreciated 3 months ago, created a post with more than 5 votes. For Updated DESeq2 performance on highly replicated yeast RNA-seq data

Help
Access

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 2.2.0
Traffic: 176 users visited in the last hour