Doubt regarding CN analysis

[Vaishali Chakraborty]

Hi Team,

I had 2 doubts regarding the copy number data I have processed from GISTIC2 and bought in cBioPortal:

1. I received a set of CNV results from one of our clinical teams and they had processed the CNV data using the ascat module, it essentially had the chr, location and copy number values from which I was able to create the .seg file to use as an input in the gistic2.0 pipeline. Hence, I was able to create the data_cna and data_log2_cna files and import them in cBioPortal. However, I was not able to create the cna.seg file as I was unable to understand how the seg.mean value was calculated. In the igv/SEG link it just mentions that: "It reads the last column as the numeric value for that locus". Any help on this regard on how to calculate and get this seg.mean value will be appreciated.

2. Is just a view query, since these are ovarian cancer samples, the most important genes for the study are BRCA1 and BRCA2. In the lolliplot view I can see the Copy number alterations for both the genes, example BRCA1 here:

However, in the study view it shows only BRCA2 gene having 1 mutation, BRCA1 is completely missing and the numbers also look odd. In the above image itself I can see at least 4 samples with Copy# alterations. Please find the reference image below showing only the BRCA2 gene when BRCA is being searched. The profiled samples for mutations and copy number and all samples are same as all are tumor samples:

Looking forward to hearing from you.

Best regards,

Vaishali

[Tali Mazor]

Hi Vaishali,

1) I don't know anything about this, but I've cc'ed our lead data curator Ritika, who might be more familiar with this process than I am.

2) The chart in study view only shows amplifications & deep deletions, it will not show shallow deletions or gains. However, all those alteration types will appear in the mutations tab, which is why you see more alterations there than you do in the study view.

I hope that helps,

-Tali

--
You received this message because you are subscribed to the Google Groups "cBioPortal for Cancer Genomics Discussion Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cbioportal+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/cbioportal/CAB5ADusOMn1i_Eboc520xmuZtKU1zepsM3tJ%3Dg3eqPr%3Dzu1TGQ%40mail.gmail.com.

[Vaishali Chakraborty]

Hi Tali,

Now it makes sense, thank you so much. I'll wait for Ritika's response on my first question.

Best regards,

Vaishali 

[Vaishali Chakraborty]

Hi Ritika and Team,

Please let me know if you have any update on my previous question:

I received a set of CNV results from one of our clinical teams and they had processed the CNV data using the ascat module, it essentially had the chr, location and copy number values from which I was able to create the .seg file to use as an input in the gistic2.0 pipeline. Hence, I was able to create the data_cna and data_log2_cna files and import them in cBioPortal. However, I was not able to create the cna.seg file as I was unable to understand how the seg.mean value was calculated. In the igv/SEG link it just mentions that: "It reads the last column as the numeric value for that locus". Any help on this regard on how to calculate and get this seg.mean value will be appreciated.

Also, the seg.cn value used for input in gistic2.0 just to double check I did: seg.cn=log2(CN)-1 where CN is the called copy number value. Only for negative values i.e., when copy number = 0 which means a homozygous deletion, log2 of zero would have been infinity so I changed the seg.CN value to -2 wherever there was a deletion on both the alleles. Hope my understanding was correct on this, please let me know your suggestions.

Looking forward to hearing from you.

Best regards,

Vaishali

[Vaishali Chakraborty]

Hi Team,

Any help on the below query will be greatly appreciated, quite stuck on this point at the moment.

Best regards,

Vaishali

[Vaishali Chakraborty]

++ kun...@mskcc.org 

[Ritika Kundra]

Hi Vaishali,

Sorry, you said "it essentially had the chr, location and copy number values from which I was able to create the .seg file to use as an input in the gistic2.0 pipeline." so why can't cna.seg be created?

Thanks,

Ritika

[Vaishali Chakraborty]

Hi Ritika,

Thank you for your response.

The seg.mean column in the cna.seg file, I was not able to understand the calculation that needs to be used. In the igv/SEG link it just mentions that: "It reads the last column as the numeric value for that locus" and the example.seg present in this link has negative and positive values. I also checked in IGV google groups link and they had mentioned that for a diploid organism IGV auto-calibrates and does a log2 of CN, this did not happen when I imported the data in cBioportal as well as IGV.

Also, even if I do a log2 of CN what will be the log2 of 0 where 0 is a Homozygous deletion. Plus, I also tried reverifying with the example.seg file provided by calculating the antilog 2 of the entries and the entries do not come equal to the called copy number values.

For example, in the first row of example file the seg.mean value = -0.7116 and Antilog2 of (-0.7116) = 0.610642 which is not a whole number like for called ploidy:

0 = Homozygous deletion

1 = Hetro Deletion

2 = Normal

3 = Amplification in one allele

4 = Amplification in 2 alleles or 2 copy Amplification in on allele

So, I figured I might be doing something wrong here. Basically, I wasn't able to figure out the seg.mean calculation which is one of the essential columns in the cna.seg file and any help on that front will be greatly helpful.

The .seg file has the seg.cn value which is used for input in gistic2.0 just to double check I did: seg.cn=log2(CN)-1 where CN is the called copy number value. Only for negative values i.e., when copy number = 0 which means a homozygous deletion, log2 of zero would have been infinity so I changed the seg.CN value to -2 wherever there was a deletion on both the alleles. Hope my understanding was correct on this, please let me know your suggestions.

Best regards,

Vaishali

[Vaishali Chakraborty]

Hi Ritika and Team,

Any update on the below questions? Do let me know if you need any additional information like files from my end to check further.

Best regards,

Vaishali

[kun...@mskcc.org]

Hi Vaishali,

 

Hmm I think you might need to consult with someone from CNA bioinformatics pipeline. But we did find a similar question on Biostar that might be helpful for you:
https://www.biostars.org/p/306180/

It might be actually helpful if you ask on Biostars.

 

Thanks,

Ritika

++ kun...@mskcc.org 

 

However, in the study view it shows only BRCA2 gene having 1 mutation, BRCA1 is completely missing and the numbers also look odd. In the above image itself I can see at least 4 samples with Copy# alterations. Please find the reference image below showing only the BRCA2 gene when BRCA is being searched. The profiled samples for mutations and copy number and all samples are same as all are tumor samples:

 

 

Looking forward to hearing from you.

 

Best regards,

Vaishali

--
You received this message because you are subscribed to the Google Groups "cBioPortal for Cancer Genomics Discussion Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cbioportal+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/cbioportal/CAB5ADusOMn1i_Eboc520xmuZtKU1zepsM3tJ%3Dg3eqPr%3Dzu1TGQ%40mail.gmail.com.

*** Only open attachments or links from trusted senders. Report phishing to inf...@mskcc.org ***

 

=====================================================================

Please note that this e-mail and any files transmitted from
Memorial Sloan Kettering Cancer Center may be privileged, confidential,
and protected from disclosure under applicable law. If the reader of
this message is not the intended recipient, or an employee or agent
responsible for delivering this message to the intended recipient,
you are hereby notified that any reading, dissemination, distribution,
copying, or other use of this communication or any of its attachments
is strictly prohibited. If you have received this communication in
error, please notify the sender immediately by replying to this message
and deleting this message, any attachments, and all copies and backups
from your computer.

[Vaishali Chakraborty]

Hi Ritika,

Thank you for your help and sending the link? What is the pipeline that you use at MSKCC for Copy Number detection which will provide results in coherence with data loading in cBioPortal, I can try that as well.

Best regards,

Vaishali

[Vaishali Chakraborty]

Hi,

Any update on the below question?

Best regards,

Vaishali 

[kun...@mskcc.org]

Hi Vaishali,

 

There are many different methods used throughout MSK for CNA detection. One example is the MSK-IMPACT clinical series where they define their methods in these two publications:
https://pubmed.ncbi.nlm.nih.gov/25801821/

https://pubmed.ncbi.nlm.nih.gov/28481359/

++ kun...@mskcc.org 

 

Error! Filename not specified.

However, in the study view it shows only BRCA2 gene having 1 mutation, BRCA1 is completely missing and the numbers also look odd. In the above image itself I can see at least 4 samples with Copy# alterations. Please find the reference image below showing only the BRCA2 gene when BRCA is being searched. The profiled samples for mutations and copy number and all samples are same as all are tumor samples:

 

Error! Filename not specified.

[Vaishali Chakraborty]

Hi Ritika,

Thank you for your help. Have a great week!

Best regards,

Vaishali