User: Giovanni Bacci

Reputation:
0
Status:
New User
Scholar ID:
Google Scholar Page
Last seen:
4 months, 3 weeks ago
Joined:
1 year, 6 months ago
Email:
d******@gmail.com

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

This helps us discourage the inappropriate use of our site.

Posts by Giovanni Bacci

<prev • 9 results • page 1 of 1 • next >
0
votes
1
answer
350
views
1
answers
Comment: C: Deviance decomposition with DESeq2
... Dear Michael, thanks for your help! I changed my code as follows: # Full model dds.full <- DESeq(dds.full, test = "Wald", fitType = "parametric", betaPrior = T) # Deviance estimation of reduced models deviances <- sapply(models[-1], function(m){ dds <- dds.full # Copying the full model ...
written 15 months ago by Giovanni Bacci0
0
votes
1
answer
350
views
1
answers
Comment: C: Deviance decomposition with DESeq2
... Sorry if I come back to this after quite a long time, but I'm still struggling to produce reliable results. It seems that the above calculation produced some negative values. I would say that the factor considered is somehow worsening the model by increasing the deviance of the full model in respect ...
written 15 months ago by Giovanni Bacci0
0
votes
1
answer
350
views
1
answer
Deviance decomposition with DESeq2
... Hello, I'm using DESeq2 to analyse a large dataset with basically three predictors: 1 - Subjects = the ID of the subject involved in the experiment (categorical) 2 - Culture medium = the culture medium (categorical) 3 - Time = the number of days (continuous) I would like to inspect the proporti ...
deseq2 analysis of deviance written 18 months ago by Giovanni Bacci0 • updated 18 months ago by Michael Love25k
0
votes
1
answer
477
views
1
answers
Comment: C: DESeq2 multifactor design with interaction term
... Hi Mark and thank again for your help! If I understood correctly the interaction between, let's say, food B and subject 5002 should be: (5002.B - 5002.A) - (5001.B - 5001.A) if that holds, the interaction between food B and subject 5003 should be: (5003.B - 5003.A) - (5001.B - 5001.A) so if I ...
written 18 months ago by Giovanni Bacci0
0
votes
1
answer
477
views
1
answers
Comment: C: DESeq2 multifactor design with interaction term
... That was my fault! I was giving the unfiltered table to my plotting function, sorry. The function now works well and the zero-counts OTUs have been removed. My last question regarding the interaction part of the original post: If I understood correctly I can extract contrasts between food B and C ...
written 18 months ago by Giovanni Bacci0
0
votes
1
answer
477
views
1
answers
Comment: C: DESeq2 multifactor design with interaction term
... Let me clarify that: the function lfcShrink worked fine with both methods but the output still contains zero-count OTUs. The worst thing is that they have a huge log2FoldChange and a very small p-value so it is really difficult to remove them in a "programatic" way without removing valid difference ...
written 18 months ago by Giovanni Bacci0
0
votes
1
answer
477
views
1
answers
Comment: C: DESeq2 multifactor design with interaction term
... I tried with lfcShrink (both with apeglm and ashr) and it doesn't seem to work. As you said I actually have 16S rRNA data they are not RNA-seq and they could not follow negative binomial distribution (I tried to model them with glmmADMB and in some cases the Poisson family seems to work better than ...
written 18 months ago by Giovanni Bacci0
0
votes
1
answer
477
views
1
answers
Comment: C: DESeq2 multifactor design with interaction term
... Thanks for the reply! Sorry to bother you but I'm trying to fully understand how DESeq2 builds result tables. The steps I reported for subject2 can be considered correct? I was planning to use DESeq2 for other studies where I cannot make the simplification that you suggested so I need to know if (i ...
written 18 months ago by Giovanni Bacci0
1
vote
1
answer
477
views
1
answer
DESeq2 multifactor design with interaction term
... Hello, I'm running DESeq2 in combination with phyloseq to analyze 16S rRNA data coming from a large survey. I have gut microbiome coming from 6 subjects (called 5001, 5002, 5003, 5004, 5005, and 5006) that experimented 3 types of diets (called A, B, and C). I would like to test differences between: ...
deseq2 multiple factor design interaction term written 18 months ago by Giovanni Bacci0 • updated 18 months ago by Michael Love25k

Latest awards to Giovanni Bacci

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 16.09
Traffic: 303 users visited in the last hour