OncoPrinter and Mutation Mapper Questions

199 views
Skip to first unread message

Shannon Glynn

unread,
Jul 20, 2018, 9:51:26 AM7/20/18
to cbiop...@googlegroups.com

To Whom It May Concern,


I am using both the Oncoprinter and Mutation Mapper tools to visualize my own data and have a few questions.


For the Oncoprinter, the data input requires that all mutations are categorized as missense, truncation or inframe. My data includes the following additional mutation types: frameshift, nonstart, nonstop (extension), and splice site. I was wondering if you could provide advice on how to categorize these into the 3 types provided.


For the Mutation Mapper, I am having some issues visualizing my data for mutations in the gene H3F3A. Please see attached for the output as I am getting a nearly 4,000 AA length protein. However, when I use cBioportal to query the IMPACT study for H3F3A mutations, I get a normal lolliplot (also attached). I am similarly getting different lolliplot outputs from when I input my own data for mutations in HIST1H3B versus querying the IMPACT data for mutations in this gene (also attached). Please advise.


Thank you so much in advance.


Sincerely,

Shannon Glynn


Shannon Glynn

MD Candidate | Class of 2021

Weill Cornell Medical College

smg...@med.cornell.edu845-709-0233

IMPACT H3F3A Lolliplot.pdf
H3F3A Lolliplot.pdf
HIS1H3B Lolliplot.pdf
IMPACT HIST1H3B Lolliplot.pdf

Kelsey Zhu

unread,
Jul 20, 2018, 7:30:41 PM7/20/18
to Shannon Glynn, cbiop...@googlegroups.com
Hi Shannon,
cBioPortal categorizes all frameshift, nonstart, nonstop, and splice site into truncating mutations.
I've forwarded your question to one of our senior members for advice.
Best!
Kelsey


--
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+unsubscribe@googlegroups.com.
To post to this group, send email to cbiop...@googlegroups.com.
Visit this group at https://groups.google.com/group/cbioportal.
To view this discussion on the web visit https://groups.google.com/d/msgid/cbioportal/SN6PR06MB3968DCB0C9FDBCF3145392DBE9510%40SN6PR06MB3968.namprd06.prod.outlook.com.
For more options, visit https://groups.google.com/d/optout.

Shannon Glynn

unread,
Jul 23, 2018, 12:44:16 PM7/23/18
to Kelsey Zhu, cbiop...@googlegroups.com

Hi Kelsey,


Thanks so much. As a follow up on the lolliplot issues, I have been experiencing similar problems with the HIST1H2 genes. Please see the attached screenshots. HIST1H2AG, AM and AL are all supposed to produce the same protein, but I keep getting different lolliplot outputs for each of them.


I appreciate your help!


Sincerely,

Shannon Glynn


From: Kelsey Zhu <kelse...@gmail.com>
Sent: Friday, July 20, 2018 7:30:40 PM
To: Shannon Glynn
Cc: cbiop...@googlegroups.com
Subject: [EXTERNAL] Re: [cbioportal] OncoPrinter and Mutation Mapper Questions
 
HIST1H2 Lolliplots.PNG

JJ Gao

unread,
Jul 23, 2018, 6:00:14 PM7/23/18
to smg...@med.cornell.edu, Kelsey Zhu, cBioPortal for Cancer Genomics Discussion Group
Hi Shannon,

Sorry for the issue. We are looking into them.

Thanks,
-JJ

To unsubscribe from this group and stop receiving emails from it, send an email to cbioportal+...@googlegroups.com.

--
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 post to this group, send email to cbiop...@googlegroups.com.
Visit this group at https://groups.google.com/group/cbioportal.

Ino de Bruijn

unread,
Jul 24, 2018, 5:35:26 PM7/24/18
to JJ Gao, smg...@med.cornell.edu, Kelsey Zhu, cBioPortal for Cancer Genomics Discussion Group
Hi Shannon,

Thanks for reaching out! This actually seems to be a bug in the current mutation mapper. There is no mapping for some of those gene names, therefore it picks the longest available protein.

We are working on a fix that we're currently beta testing. All the gene names you mentioned now map to the same protein length. It will be released in the next few weeks. It would be great if you'd like to give the beta version a try and let us know if you run into any issues:


Best wishes,
Ino

Shannon Glynn

unread,
Jul 25, 2018, 1:55:54 PM7/25/18
to Ino de Bruijn, JJ Gao, Kelsey Zhu, cBioPortal for Cancer Genomics Discussion Group

Hi Ino,


Thanks so much! Just used the beta version to re-make my lolliplots and they are all the correct length now. I was wondering, does the green "histone" box that appears on the plot indicate the core of the histone, with either side being the N and C-terminal tails? I am annotating the plots and want to make sure I'm interpreting them correctly.


Thanks so much for your help!


Sincerely,

Shannon


From: Ino de Bruijn <i...@ino.pm>
Sent: Tuesday, July 24, 2018 5:35:09 PM
To: JJ Gao
Cc: Shannon Glynn; Kelsey Zhu; cBioPortal for Cancer Genomics Discussion Group

Ino de Bruijn

unread,
Jul 26, 2018, 2:20:18 PM7/26/18
to Shannon Glynn, JJ Gao, Kelsey Zhu, cBioPortal for Cancer Genomics Discussion Group
Hi Shannon,

The plot only shows the amino acid sequence. Unfortunately I'm not sure what the difference is between just the core of the histone or the entire histone. Do you know how many amino acids that spans? The domain information comes from PFAM, so that would be a good place to find more information. If you mouse over the domains here's a link to their site.

Please do let us know if you gain some new insights, we're curious to know as well

Best wishes,
Ino
Reply all
Reply to author
Forward
0 new messages