User: felix.klein

gravatar for felix.klein
felix.klein140
Reputation:
140
Status:
Trusted
Location:
Germany
Last seen:
7 months, 1 week ago
Joined:
3 years, 7 months ago
Email:
d******@gmx.de

Posts by felix.klein

<prev • 50 results • page 1 of 5 • next >
0
votes
2
answers
166
views
2
answers
Answer: A: FourCSeq - errors in plotZScores
... Hi Rustavo, right now I cannot tell what the problem is and from which function it is caused. Could you please run traceback() to track this down. Best regards, Felix   ...
written 10 months ago by felix.klein140
0
votes
1
answers
235
views
1
answers
Answer: A: FourCseq, rawRanges Description
... Hello Theodoros, these are a mixture from FourCSeq and DESeq2. For FourCSeq I can tell what they are, for DESeq2 you have to look it up in the documentation. leftSize/rightSize: Size of the left fragment end (1st to 2nd cutter site) leftValid/rightValid: Are the fragment ends valid (fragment of f ...
written 22 months ago by felix.klein140
0
votes
3
answers
307
views
3
answers
Comment: C: FourCseq: getZScores errors, first case 6 digenerated cutterm second case 4 base
... Hello Theodoros, have a look at the dispersion fit whether this makes sense to you (plotDispEsts(fcf)). And have a look at the DESeq2 package which is behind all these calculations. Best regards, Felix ...
written 22 months ago by felix.klein140
0
votes
3
answers
307
views
3
answers
Comment: C: FourCseq: getZScores errors, first case 6 digenerated cutterm second case 4 base
... Hello Theodoros, there should be not much more false "significat" contacts if you change the threshold within a reasonable range. What I meant with the switch in interaction differences is the following: If you use the plotDifferences function and see a clear trend that the log fold change is a ...
written 22 months ago by felix.klein140
0
votes
3
answers
307
views
3
answers
Comment: C: FourCseq: getZScores errors, first case 6 digenerated cutterm second case 4 base
... Hello Theodore, this seems like a quite high number. How many fragments are actually left after you used this threshold? You should definetly double check the results in this case because you probably remove quite a lot of fragments. Especially watch out if you have a switch in the interaction dif ...
written 22 months ago by felix.klein140
0
votes
2
answers
215
views
2
answers
Answer: A: FourCseq: mixed amount of replication/samples
... Hello Theodore, you can create the colDate in the way your experiment has been caried out. And if you have the required information in a csv file then you should use it as import. For simplicity I made use of the each function, but I could have used c(1,2, 1,2,3, ...) instead. Cheers, Felix ...
written 22 months ago by felix.klein140
0
votes
3
answers
307
views
3
answers
Answer: A: FourCseq: getZScores errors, first case 6 digenerated cutterm second case 4 base
... Hi Theodore, how do the scatter plots between your replicates look like? Do the libraries show sufficient agreement or are they quite different? As a straightforward approach you can try to increase the number in minCount. This will set aside low count fragments, which tend to have a higher noise ...
written 22 months ago by felix.klein140 • updated 22 months ago by Wolfgang Huber13k
0
votes
1
answers
227
views
1
answers
Answer: A: FourCseq- adding information manually
... Hi Tal, check that you have the same factor levels. I would assume, that fragemented_Rn6 has all chromosomes as level while colData only has chr5. If you add the missing levels there it should work. Best regards, Felix ...
written 22 months ago by felix.klein140
0
votes
1
answers
232
views
1
answers
Comment: C: local fit and FourCseq
... Hi Tal, this might indeed be to few fragments for the analysis. Did you have a look at the raw counts and how they are distributed? Also check the correlation plots with the different thresholds (after getZScores). How do they look? Best regards, Felix On 07/10/2016 03:03 PM, tlgolan [bioc] wr ...
written 22 months ago by felix.klein140
0
votes
1
answers
232
views
1
answers
Comment: C: local fit and FourCseq
... Hi Tal, this procedure is done automatically by DESeq2, so you cannot force the parametric fit. You can try to change the minCount = 40 threshold in the getZScores function so that the noisier part of the data is removed. Best regards, Felix ...
written 22 months ago by felix.klein140

Latest awards to felix.klein

Scholar 2.6 years ago, created an answer that has been accepted. For A: FourCSeq - how to write tables with the results
Teacher 2.6 years ago, created an answer with at least 3 up-votes. For A: FourCSeq - how to write tables with the results

Help
Access

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 2.2.0
Traffic: 293 users visited in the last hour