Option to use cancer genes to filter genes on study view #5716
Comments
|
@zhx828 could you prioritize this one? Maybe @dippindots can help? |
|
@zhx828 let's not worry about oncogenicity for now -- it'll be harder b/c we will have to annotate all mutations. In this one we just need to filter based on the oncokb cancer gene list. |
|
@zhx828 maybe just add a filter icon before the download icon would do the trick? |
|
After discussing with JJ, we are going to use the realtime oncokb cancer gene list to filter the genes. |
|
@schultzn JJ and I had some discussion about this ticket. We would love to add a oncokb column in the table to indicate whether the gene is from oncokb cancer gene list, and sort the table by oncokb column. Then you would also have some links out to oncokb gene page. |
|
Should it be a column, and do we have room for that? Or a simple check box at the top of the table?
… On Mar 7, 2019, at 3:16 PM, Hongxin ***@***.***> wrote:
@schultzn <https://github.com/schultzn> JJ and I had some discussion about this ticket. We would love to add a oncokb column in the table to indicate whether the gene is from oncokb cancer gene list, and sort the table by oncokb column. Then you would also have some links out to oncokb gene page.
Filtering the table by oncokb cancer gene list would miss some genes that highly mutated in subset of the study.
What do you think? I will post a mock up later.
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub <#5716 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ALV34C3r2OqG1x9_rZayg-S_Zj-uqsVcks5vUXONgaJpZM4awQzi>.
|
|
@schultzn I would vote for a column and it will only be a column with symbols, it should take very minimum of the space. Will send a mock up shortly. |
|
Ok… a column and a checkbox?
And this will only be to indicate whether a gene is a TSG or OG? Or in general in OncoKB - maybe that is better.
And what about step 2: Filter table for oncogenic mutations only, to get driver alteration frequencies. :-)
… On Mar 7, 2019, at 3:51 PM, Hongxin ***@***.***> wrote:
@schultzn <https://github.com/schultzn> I would vote for a column and it will only be a column with symbols, it should take very minimum of the space. Will send a mock up shortly.
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub <#5716 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ALV34PTzM_mr0gKfRfhzr1mER-6ex55tks5vUXvrgaJpZM4awQzi>.
|
|
@schultzn I'm going to include all genes in oncokb cancer gene list |
|
We may not need a checkbox as long as it is sorted by OncoKB by default. This may reduce some confusion as people may search for their genes (non-cancer genes) and they would have to figure out they need to do one more step to bring them in. (we had a number of users asking about missing genes in the mutated genes table previously when we had the filter) |
|
It's a good start - but maybe a little too large / in your face? I like the size of the MutSig icon.
|
|
It's close... so maybe just a little smaller? |
|
@schultzn what about this one? |
|
@zhx828 maybe for OncoKB curated genes, link to the gene page. For other genes, link to the cancer genes page? |
|
Looks good - not 100% sure about the header using the same (but larger) symbol, but I don’t have a better idea.
… On Mar 11, 2019, at 10:41 AM, Hongxin ***@***.***> wrote:
@jjgao <https://github.com/jjgao> sounds good. @schultzn <https://github.com/schultzn> @jjgao <https://github.com/jjgao> what do you think about the design though?
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub <#5716 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ALV34Cf_WUkIrTNsVW4qOMVJbnpf-ATBks5vVmsBgaJpZM4awQzi>.
|
|
Nice!
… On Mar 11, 2019, at 3:02 PM, Hongxin ***@***.***> wrote:
@schultzn <https://github.com/schultzn> : @jjgao <https://github.com/jjgao> suggested to use black in the header. What do you think?
<https://user-images.githubusercontent.com/5400599/54150359-a3368180-440e-11e9-931b-293e75d26f2b.png>
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub <#5716 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ALV34LGK4cxiSWOaF-OvjLpr_d8o37n9ks5vVqhPgaJpZM4awQzi>.
|
|
@schultzn @jjgao I have a running instance here including the feature. Could you take a look and give me some suggestions? Thanks. https://cbioportal-frontend-pr-2109.herokuapp.com/study?id=cellline_nci60 The frequency is sorted by oncokb gene then by frequency. |
|
It looks good - but I still think it is too many blue circles.
Also, what if others want to use a different list of cancer genes? Can we revisit the option of having a simple checkbox in the header of the table? Maybe we can get together and brainstorm briefly?
It looks so much better in this view:
… On Mar 11, 2019, at 10:01 PM, Hongxin ***@***.***> wrote:
@schultzn <https://github.com/schultzn> @jjgao <https://github.com/jjgao> I have a running instance here including the feature. Could you take a look and give me some suggestions? Thanks. https://cbioportal-frontend-pr-2109.herokuapp.com/study?id=cellline_nci60 <https://cbioportal-frontend-pr-2109.herokuapp.com/study?id=cellline_nci60>
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub <#5716 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ALV34OVVligfIYKmSxzO0YRuCd-v774Mks5vVwpzgaJpZM4awQzi>.
|
|
Attachments don't make it through when responding via email. Sorry. This looks good: |
|
Not this (the default):
|
|
@schultzn @jjgao @Sjoerd-van-Hagen do we have a conclusion yet? |
|
@zhx828 based on the discussion above, we can go ahead with the oncokb icon, I think. |
|
I am fine with it but have a personal preference for a filter icon. Mainly because it is a filter. A hover over could show 'filter by oncokb cancer genes. That would still be one click but it is clear that a filter is set, and can be extended in the future. @schultzn what do you think? |
|
How about a cancer ribbon instead of the OncoKB icon? |
|
I would like to make explicit that the list is filtered and the user is not seeing all the genes. How would you do that with the ribbon? |
|
The same as with the OncoKB symbol - there will be two versions, one to indicate “selected”, the other to indicate “not selected”.
… On May 15, 2019, at 3:41 AM, Sjoerd-van-Hagen ***@***.***> wrote:
I would like to make explicit that the list is filtered and the user is not seeing all the genes. How would you do that with the ribbon?
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub <#5716?email_source=notifications&email_token=AC2XPYAN46AAEW7CF5JG6MTPVO5BDA5CNFSM4GWBBTRKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODVNZOQY#issuecomment-492541763>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AC2XPYHENF35GJN7XHDG6MLPVO5BDANCNFSM4GWBBTRA>.
|
|
I think I would still prefer the funnel symbol that is typically used to signify filters, but have no strong objections to using the ribbon. |
|
@schultzn and I had some discussion and agreed that we can use a filter icon with a tooltip as @Sjoerd-van-Hagen suggested. tooltips:
|
|
Thank you everyone.
…On Wed, May 15, 2019 at 5:58 PM JianJiong Gao ***@***.***> wrote:
@schultzn <https://github.com/schultzn> and I had some discussion and
agreed that we can use a filter icon with a tooltip as @Sjoerd-van-Hagen
<https://github.com/Sjoerd-van-Hagen> suggested.
tooltip:
- "Filtered by OncoKB Cancer Gene List. Click to show all genes."
- "Showing all genes. Click to filter by OncoKB Cancer Gene List"
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#5716?email_source=notifications&email_token=ABJGQFYCVGSLL4QP7MVNMMDPVSBPVA5CNFSM4GWBBTRKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODVQCBJQ#issuecomment-492839078>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/ABJGQF3327AB3PHO3PN76B3PVSBPVANCNFSM4GWBBTRA>
.
=====================================================================
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.
|
|
@schultzn @jjgao @Sjoerd-van-Hagen the interface will be the following. I will send out a testing interface soon. After filtering
Before filtering
|
|
@schultzn @jjgao @Sjoerd-van-Hagen here is the testing instance https://cbioportal-frontend-pr-2109.herokuapp.com/study/summary?id=acc_tcga |
|
The filter should be disabled by default imo. For instance, in the example, MUC16 has a higher number of mutated samples than MUC4 but is not shown by default, even though it is associated with ovarian cancer (enough to spend money on trying to create a drug for it that targets this gene (OREGOVOMAB)) where CA-125 is also used as a biomarker, and has a very high occurrence in melanoma. I think people would like to see MUC-16 in this list. In any case, I think by enabling this filter by default we are forcing our opinion, on what the cancer genes are, a bit too much. |
|
Wait, disregard the last comment. There seems to be a bug. If I follow the link I see the icon is enabled and the text says it is filtered, but it is actually not. The behaviour of the switch mirrored. |
|
We can debate whether enabling or disabling is the best route - I am ok with disabling, but would personally always enable it right away. :-)
While CA125 is an ovarian cancer biomarker, MUC16, which encodes it, is one of the largest genes in the genome and notorious for its high background mutation rate. So the high frequency of mutations is very misleading for those who don’t know…
… On May 16, 2019, at 5:51 AM, Sjoerd-van-Hagen ***@***.***> wrote:
Wait, disregard the last comment. There seems to be a bug. If I follow the link I see the icon is enabled and the text says it is filtered, but it is actually not. The behaviour of the switch mirrored.
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub <#5716?email_source=notifications&email_token=AC2XPYC47IJNQAJFDPWWBHDPVUVCBA5CNFSM4GWBBTRKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODVRJNFI#issuecomment-493000341>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AC2XPYDTD7KEVJTVKILHZIDPVUVCBANCNFSM4GWBBTRA>.
|
|
@zhx828 maybe this setting can be saved into local storage so that people can have their own preference? |
|
I think we really need to have user preferences at some point. Or perhaps advanced and beginner modes or something. So many bells and whistles... Problem is then that the solution space for the inputs will become even larger, which will make testing more difficult. |
Before it is hidden by the more recent posts. @zhx828 |
|
@Sjoerd-van-Hagen I don's see the issue though. Could you give me more info which gene is showing up which should not be? |
|
Ah, my mistake. I somehow assumed that the genes with the small icon behind it were in the OncoKB list but it is just a coincidence that if I click the filter the top genes all have the icon... @schultzn I now see that MUC16 is actually in the list too... |
|
@zhx828 anything left for this? |
|
@jjgao nope. Just need to verify everything looks good. Then I will schedule the release. |
|
@zhx828 thanks, please prioritize. |
|
@jjgao could you take a final look at this feature? https://cbioportal-frontend-pr-2109.herokuapp.com/study/summary?id=lgg_tcga |
|
@zhx828 It looks great! Thanks. |
|
@jjgao @schultzn @Sjoerd-van-Hagen this feature is now released. |
|
@Sjoerd-van-Hagen @schultzn @zhx828 should we try to remember user's preference in local storage / cookie? |
|
If it is not too much effort and you are willing to do that, I would say yes. |
|
@jjgao @Sjoerd-van-Hagen created an issue #6274 |
For both Mutated Genes and CNA Genes table, add a toggle option to show only Cancer Genes (the OncoKB list), on by default.
@zhx828 maybe do some mockup first.
The text was updated successfully, but these errors were encountered: