customized annotation through EPACTS?

105 views
Skip to first unread message

Yiyi Ma

unread,
Sep 11, 2015, 3:47:51 PM9/11/15
to EPACTS
Hi,
Is there a way to have the customized annotation into the INFO field of vcf file through EPACTS?

Thanks!
Yiyi

Hyun Min Kang

unread,
Sep 14, 2015, 4:24:20 PM9/14/15
to EPACTS
What do you meant by customized annotation? You can run any annotation software you want, but supporting custom programs are not well supported for now (since many annotation software has been changed a little bit). You can still create a .grp file based on your annotation, and run gene-based annotation test.

Thanks,
Hyun.

--
You received this message because you are subscribed to the Google Groups "EPACTS" group.
To unsubscribe from this group and stop receiving emails from it, send an email to epacts+un...@googlegroups.com.
To post to this group, send email to epa...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/epacts/eff733e0-d06c-4542-ac90-47573ed870b1%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Yiyi Ma

unread,
Sep 15, 2015, 2:07:44 PM9/15/15
to epa...@googlegroups.com
Hi Hyun Min,
Thank you for your reply. Basically, I want to know how can I make a personalized grouping strategy of the variants and create a vcf file in the format that EPACTS can use to analyze the customized group analysis? For example, I have a customized file listing how to group SNPs.


Yiyi

--
You received this message because you are subscribed to a topic in the Google Groups "EPACTS" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/epacts/BXctatwGQoM/unsubscribe.
To unsubscribe from this group and all its topics, send an email to epacts+un...@googlegroups.com.

To post to this group, send email to epa...@googlegroups.com.

Tanya M. Teslovich

unread,
Sep 16, 2015, 3:30:21 PM9/16/15
to epa...@googlegroups.com
You don’t need to modify your VCF at all.  You can provide a customized marker group file in this format:

A1BG 19:58861808_A/G 
A1CF 10:52573772_C/T 10:52595978_A/G 
A2M 12:9229390_T/C 12:9229986_C/T 12:9232325_C/A 12:9264798_C/T 12:9268393_A/C 
A2ML1 12:8975873_C/G 12:8990121_C/T 12:8990937_C/A 12:8990963_T/C 12:9002833_T/C 12:9009835_T/C 12:9013481_G/A 12:9013783_C/T 12:9013786_C/T 12:9020954_G/A 
A4GNT 3:137843333_G/A 3:137843487_G/T 
AAAS 12:53701866_C/T 12:53709162_C/T 
AADAC 3:151545726_G/T 3:151545830_G/A 3:151545958_T/C 

Tanya


Anupriya Dalmia

unread,
Nov 4, 2020, 10:10:26 AM11/4/20
to EPACTS
Hi Tanya, 

Could you please elaborate on how one may create such a marker group file? I am unable to do so using the make-group EPACTS function, I have a list of 250 genes and I want to generate a .grp file with protein coding variants to run SKAT-O. Any advice would be greatly appreciated.

Best,
Anupriya

Tanya Teslovich

unread,
Nov 4, 2020, 10:43:33 AM11/4/20
to epa...@googlegroups.com
Hi, Anupriya.  This is going back several years, so it's all a bit hazy, but I created that marker group file independent of EPACTS and provided it as an input file.  I wrote a separate script that read through the VCF and annotation file(s).  I had a few different sets of criteria based on MAF and variant annotation (missense, nonsense, CADD score, etc), and I built that logic into my script.  Inputs to my script were VCF, annotation file, and selection criteria, and the output was a marker file in the format shown below.  I fed that output file into EPACTS when I was ready to run analysis.  I hope that helps.

Tanya


Anupriya Dalmia

unread,
Nov 4, 2020, 11:24:23 AM11/4/20
to epa...@googlegroups.com
Hi Tanya,

Thank you for your extremely prompt response. Since the gene group file has some very specific criterion on how it is formatted, I was having some trouble generating this independent of EPACTS. And am also unsure why the in-built "make-group" function is returning an empty file to me. 

Of course, since it's been several years for you now, I presume pulling out the code for this would not be possible?

Anupriya

Reply all
Reply to author
Forward
0 new messages