Import patient without sample

[Ji Hen Lau]

Hi cBioPortal team,

I have a question regarding patient inclusion.

In our dataset, we have a few patients who were registered and have relevant clinical data (e.g. age, treatment status), but do not have any associated samples or genomic data because they were not yet sequenced.

From what I observed, these patients who do not have samples present in data_clinical_sample.txt or do not have any associated genomic data seems to be excluded during the import, as I can't find them while exploring the study after import.

My questions:
1. Is it expected behavior that patients without samples or genomic data are not shown in the portal?
2. Is there a supported or recommended way to include such patients — for example, by assigning them dummy sample IDs and setting all sample-level fields to NA?
3. Would adding such "empty" samples affect visualizations or lead to misleading summaries?

We’d like to have all enrolled patients (including those without sequencing data) to be visible in the portal for reference purposes, if possible.

Any advice or best practices would be greatly appreciated. Thank you for your time and for the amazing work on cBioPortal!

Regards, 

Ji Hen

[Benjamin Gross]

Hi Ji Hen,

It is possible to import a study that contains patient with a mixture of clinical and genomic files.  However, as the system was not originally designed for this purpose-as you suspected-the way to do this is to introduce dummy sample ids in the data-clinical_sample.txt file.

Below are links to the cBioPortal and our datahub repository, which contains an example study for your reference.  There are 7 samples that do not have genomic data:

P-0010577-T01-IM5
s_WJ_chol_087_T
P-0000147-T01-IM3
P-0016909-T01-IM6
s_WJ_chol_042_T
P-0023681-T01-IM6
P-0019824-T01-IM6

Please be be cautious about adding the ids to the case list files—particularly the sequenced case list-as they will be included in the denominator and could affect the statistics/results shown on the web pages.

I’ve added Ramya, a member of the cBioPortal curation team, to this email in case you have any follow-up questions.

Best,

-Benjamin

--
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 visit https://groups.google.com/d/msgid/cbioportal/9593125f-171f-4929-914d-69ec5d86004cn%40googlegroups.com.

[Ji Hen Lau]

Hi Benjamin,

Thanks for the clarification !

Ji Hen