Why the LogFC and delta-beta value are different in ChAMP
1
0
Entering edit mode
Hongbii • 0
@hongbii-24366
Last seen 16 months ago

Good day, I'm studying the DNA methylation using Champ package in R-studio. I have some questions about LogFC and delta-beta value in the result generated by the champ package. The authors have mentioned that logFC and delta value are having the same value in the documentation.

Blockquote (delta value is same as logFC, we kept is here because very old version ChAMP used).

When I tested with the test data and I got the same logFC and delta value. But when I tested with another set of data, the results are different for the logFC and delta value (different in positive and negative). May I know Is that normal? Another question is, May I know for is that the input data set need to follow the format (maybe certain header pattern) like the one in the test data?

Thank you very much,

Hongbii

DNAMethylation ChAMP • 667 views
ADD COMMENT
2
Entering edit mode
Yuan Tian ▴ 250
@yuan-tian-13904
Last seen 8 months ago
United Kingdom

I think it's normal if it's just a positive negative difference.

ChAMP function sometimes will "sort" your phenotypes, so if you want to compare "B" to "A", but after the calculation, the result might be from "A" to "B", then if so, the values are negatively calculated. I may try improve this in future version.

But for now, if you get result, you may check the graph or beta value to confirm which one is higher. I normally only use abs(LogFC) or abs(Delta) for filtering. I think there is an average value for each group in the result you may refer to.

ADD COMMENT
0
Entering edit mode

Hi,

Thank you for the reply, I appreciate it.

I still have some questions on this. The below image shows the simulation result I get. Other than the above method, May I ask Is that any parameters I can set to make the "LogFC" and the "deltaBeta" are compared by the same contrast respectively?

Thank you very much

ADD REPLY
1
Entering edit mode

Sorry, currently there is no parameter for that. Thanks for raising this issue, I can see if I can design this feature in future version.

ADD REPLY
0
Entering edit mode

Thank you very much!

ADD REPLY

Login before adding your answer.

Traffic: 368 users visited in the last hour
Help About
FAQ
Access RSS
API
Stats

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.6