Question: cn.mops: interpreting output when returnPosterior=TRUE
0
gravatar for laurabuggiotti
17 months ago by
laurabuggiotti0 wrote:

Hi,

Im using cn.MOPS for detecting CNVs in multiple WGS samples, however i got some questions concerning the interpretation of the output:

  1. When using returnPosterior=TRUE should i have an additional column with a posteroir probability value or is the output filtered according to the latter? I havent a posterior probability column, and here is the command i have used (with no error):
resCNMOPS <- cn.mops(bamDataRanges, returnPosterior = TRUE)

resCNMOPS <- calcIntegerCopyNumbers(resCNMOPS)

       2. I have a coverage of about 10X and a total of 40 samples, which threshold would you suggest to use to select final cnvr (group together CN0-1 as loss and CN3-8 as gain)? I read that i can play around with WL although the software it does estimate it from the data, which is best given that i dont have a training set? Moreover, in another post you talked about sI/NI value which i dont have in my final tables, where can i retrieve it?

       3. These 40 samples represent two populations, would you recommend to run cn.mops per population? 

I really like cn.mops and hope to get some more advice from you.

Thanks a lot in advance for your time and support,

Laura

cn.mops • 281 views
ADD COMMENTlink modified 17 months ago by Günter Klambauer540 • written 17 months ago by laurabuggiotti0
Answer: cn.mops: interpreting output when returnPosterior=TRUE
0
gravatar for Günter Klambauer
17 months ago by
Austria
Günter Klambauer540 wrote:

1. There is no additional column, but an extra slot (resCNMOPS@something) that returns the three dimensional array of posterior probs (segments x samples x copynumberclasses).

2. The default thresholds of cn.MOPS should work fine. If you have a validation set with known CNVs you should adjust the parameters of cn.MOPS to that data set. The sI/NI values are also present in the result object.

3. If you also want to find differences between the two populations then just run cn.MOPS on all samples (both populations). If you are only interested in the intra-population variability, then run cn.MOPS separately for each population.

Thanks for questions and good luck with the analysis!

 

Regards,

Günter

ADD COMMENTlink written 17 months ago by Günter Klambauer540
Please log in to add an answer.

Help
Access

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 16.09
Traffic: 328 users visited in the last hour