Question Regarding inferCNV option setting

244 views
Skip to first unread message

KaYoung Seo

unread,
Jan 11, 2021, 12:27:29 AM1/11/21
to Trinity_CTAT_users

Hello everyone,
I have trouble setting options to run inferCNV tool.

I want to set the "analysis mode" as "cells" and also "HMM report by" as "cell" .
However, the tool showed error messages when I set "analysis mode" as "cells" and "HMM report by" as "cell" or "consensus".

How can I set the option if I want to get HMM CNV prediction as cell level and also be reported the results as cell level?

Or if anyone have run the tool as option "analysis mode = cell", please tell me. 

Thank you for reading my issues, and please give me a feedback

Kayoung

cgeo...@broadinstitute.org

unread,
Jan 13, 2021, 3:42:12 PM1/13/21
to Trinity_CTAT_users
Hi Kayoung,

Which version of inferCNV are you currently using? And what error message do you get?

I am able to run the current latest version of inferCNV with the following options for the HMM:
                              analysis_mode="cells",
                              HMM_report_by="cell",
                              HMM=TRUE,

As a note though, running inferCNV in cells analysis mode is not the recommended way, as predictions from a single cell will not look very different different from the residual expression. The usefulness of the HMM is to combine the information within a region from multiple cells to better predict CNVs.

Regards,
Christophe.

KaYoung Seo

unread,
Jan 15, 2021, 1:41:01 AM1/15/21
to Trinity_CTAT_users
Hi, Christophe.
Thank you so much for replying me.
I'm using the latest 1.6.0 version of inferCNV tool. 

I set the option as same as you, but I got the following consistent error :

Error in match.arg(by) :
  'arg' should be one of “consensus”, “subcluster”, “cell”
Calls: <Anonymous> ... generate_cnv_region_reports -> get_predicted_CNV_regions -> match.arg

I double-checked my option setting, but it is exactly same as you. 

Thank you for reading my response...
Sincerely, 
Kayoung

2021년 1월 14일 목요일 오전 5시 42분 12초 UTC+9에 cgeo...@broadinstitute.org님이 작성:
Reply all
Reply to author
Forward
0 new messages