Acute Myeloid Leukemia (OHSU, Nature 2018) dataset integrity question

125 views
Skip to first unread message

Richard Darley

unread,
Dec 6, 2021, 2:29:11 PM12/6/21
to cbiop...@googlegroups.com, Alex Tonks

Hi

We have been carrying out analysis on the dataset below (downloaded from your site), but are getting unexpected outcomes from survival analyses

Acute Myeloid Leukemia (OHSU, Nature 2018)

Whole-exome sequencing of 672 acute myeloid leukemia samples (with 454 matched normals) from the Beat AML program.PubMed

 

To check this further we compared the outcome of favorable v adverse patient risk groups using these parameters:

Overall Survival (Months)

ELN 2017 Risk Classification

 

Disturbingly this comparison found no significant difference in outcome (13 v 15 months survival) confirming our suspicions from our other analyses

Is it possible that this information has been corrupted?

 

Many thanks

Richard

______________________________

Richard Darley PhD

Professor of Haematology

Department of Haematology

Division of Cancer and Genetics

School of Medicine

Cardiff University

Heath Park

Cardiff  UK

CF14 4XN

 

Yr Athro R L Darley

Adran Hematoleg

Is-adran o Ganser a Geneteg

Ysgol Meddygaeth

Prifysgol Caerdydd

Mynydd Bychan

Caerdydd CF14 4XN

 

Benjamin Gross

unread,
Dec 6, 2021, 7:43:49 PM12/6/21
to Richard Darley, Kundra, Ritika/Sloan Kettering Institute, cbiop...@googlegroups.com, Alex Tonks
Hi Richard,

Thank you for your email.  With a question like this, we usually have our curator team take a look at the data and ensure its integrity.  If the data looks good based on the publication or data provider, we can pass the issue along to our frontend team for further evaluation.  I’ve add our lead curator, Ritika Kundra to this thread to manage the first step.

Sorry for any inconvenience this may have caused.

Regards,
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 on the web visit https://groups.google.com/d/msgid/cbioportal/LO6P265MB6013C22D361DB25CF8E0954A9C6D9%40LO6P265MB6013.GBRP265.PROD.OUTLOOK.COM.

Reply all
Reply to author
Forward
0 new messages