User: chris86

gravatar for chris86
chris86 • 390
Reputation:
390
Status:
Trusted
Location:
UCL, United Kingdom
Website:
https://intobioinforma...
Last seen:
1 day ago
Joined:
4 years, 5 months ago
Email:
c*************@gmail.com

I am a post-doc data analyst working in Rhuematoid Arthritis research in London.

Posts by chris86

<prev • 181 results • page 1 of 19 • next >
0
votes
1
answer
87
views
1
answers
Answer: A: The order used in common practce? batch effect removel and normalization
... I'd normalise first to remove your technical variation within batches, if this was RNA-seq then some samples might have way more reads in than others and that would need correcting for. Then batch is like differences between certain labs and operators etc, so that can be done afterwards. ...
written 12 weeks ago by chris86 • 390
0
votes
0
answers
92
views
0
answers
Comment: C: Adapter trimming before mapping with Salmon
... Depends on what percentage adaptors contamination you have, trim them if the fastqc plot says 5% + in my opinion. Salmon doesn't remove adaptors... no. ...
written 3 months ago by chris86 • 390
1
vote
1
answer
155
views
1
answers
Answer: A: Which one is more trustable for Cox Proportional Hazard Model with Lasso penalty
... Actually (unless this is just for training, then excuse me) none of this is trustworthy because you are not splitting into separate training and testing datasets, e.g. https://stackoverflow.com/questions/18130338/plotting-an-roc-curve-in-glmnet If you want to do cross validation + glmnet just use ...
written 3 months ago by chris86 • 390
0
votes
2
answers
101
views
2
answers
Comment: C: gene Isoforms quantification using RNA-seq data
... Salmon is very good, rsem is also good but very slow. The above comment could be misleading. ...
written 3 months ago by chris86 • 390
0
votes
0
answers
117
views
0
answers
Comment: C: How to apply a linear mixed model onto an RNA-Seq data?
... Correct me if I’m wrong, but I think the duplicatecorrelation function in limma is equivalent to adding a random effect. ...
written 3 months ago by chris86 • 390
0
votes
1
answer
186
views
1
answers
Comment: C: Variance stabilising transformation for machine learning
... Indeed I use the vst function all the time before running caret or clustering etc, it is very useful, you misunderstood my point. There isn’t a lot of good sense to using deseq2 to select features to make a classification model to me, seeing as many algorithms will select features for you according ...
written 3 months ago by chris86 • 390
0
votes
1
answer
186
views
1
answers
Comment: C: Variance stabilising transformation for machine learning
... It’s an unusual idea to pre select genes ranked according to a prior statistical test. If you read the caret manual they suggest allowing the underlying algorithm to select features automatically eg glmnet. There is also the overfitting issue if you haven’t cross validated the feature selection with ...
written 4 months ago by chris86 • 390
0
votes
1
answer
124
views
1
answers
Comment: C: Limma classical interaction model question
... OK Gordon, I actually have already done as you suggest in the manual and used the single factor method by pasting the groups together. So your saying the above when you test response*drug is not the same interaction as I would be testing by using (responderDrug1-non-responderDrug1)-(responderDrug2-n ...
written 5 months ago by chris86 • 390
0
votes
1
answer
124
views
1
answers
Comment: C: Limma classical interaction model question
... OK thanks Gordon, so how would I modify the second block of code to test the same hypothesis as the first? ...
written 5 months ago by chris86 • 390
0
votes
1
answer
124
views
1
answer
Limma classical interaction model question
... Hi So I am trying to work out why these two bits of code give different results. The overall aim is to look for genes that are changing differently between responders and non-responders (to therapy) between two different treatment groups (baseline only data i.e. a single timepoint). However, I am a ...
limma written 5 months ago by chris86 • 390 • updated 5 months ago by Gordon Smyth ♦ 39k

Latest awards to chris86

Popular Question 12 months ago, created a question with more than 1,000 views. For Prior plots from COMBAT
Popular Question 18 months ago, created a question with more than 1,000 views. For Limma camera function never gives any differentially expressed gene sets
Popular Question 18 months ago, created a question with more than 1,000 views. For Limma un moderated t-test
Scholar 2.5 years ago, created an answer that has been accepted. For A: Heatmap of a subset of genes
Popular Question 2.5 years ago, created a question with more than 1,000 views. For Variance filtering prior to RNA-seq differential expression
Popular Question 2.5 years ago, created a question with more than 1,000 views. For Limma camera function never gives any differentially expressed gene sets
Popular Question 2.5 years ago, created a question with more than 1,000 views. For Prior plots from COMBAT
Popular Question 2.5 years ago, created a question with more than 1,000 views. For Limma camera function never gives any differentially expressed gene sets
Autobiographer 3.6 years ago, has more than 80 characters in the information field of the user's profile.
Centurion 3.6 years ago, created 100 posts.
Scholar 3.6 years ago, created an answer that has been accepted. For A: Heatmap of a subset of genes
Student 3.6 years ago, asked a question with at least 3 up-votes. For Limma camera function never gives any differentially expressed gene sets
Scholar 3.6 years ago, created an answer that has been accepted. For A: Apply the plotPCA function in DESeq2 package to quantitative MS-based data
Supporter 3.6 years ago, voted at least 25 times.
Scholar 3.6 years ago, created an answer that has been accepted. For A: Heatmap of a subset of genes
Content
Help
Access
Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 16.09
Traffic: 176 users visited in the last hour