DiffBind issue with edgeR$bTagwise and DESeq2$fitType
1
0
Entering edit mode
jason.kost • 0
@74d92117
Last seen 16 months ago
United States

Noticed a discrepancy in config file parameter names.

In all documentation, the DBA$config$edgeR$bTagwise and DBA$config$DESeq2$fitType parameters are structured as written here. However, in the actual DBA object they are stored as DBA$config$edgeR.bTagwise and DBA$config$DESeq2.fitType

This causes problems when trying to run an analysis on a DBA object resulting in the following error:

"Analyze error: Error in pv$config$DESeq2$fitType: $ operator is invalid for atomic vectors"

I think this may only occur if the values are explicitly set in an external config, and a workaround seems to be to just manually set the correct parameter using the "." version's value.

Config DiffBind Error • 919 views
ADD COMMENT
0
Entering edit mode
Rory Stark ★ 5.2k
@rory-stark-5741
Last seen 4 weeks ago
Cambridge, UK

There appears to be some ambiguity regarding using a config file to specify the configuration parameters. Currently, you are unable to use an external file to set nested configuration parameters, you can't use this method to set the $edgeR$bTagwise or $DESeq2$fitType parameters. You will need to set them manually, either in the call to dba() or after that call by explicitly setting $config$edgeR$bTagwise etc.

The substitution of '.' for '$' is an artefact of this limitation, and setting the config using a '.' will not work (it will not change the parameter). It must be specified as a nested list using '$'.

ADD COMMENT
0
Entering edit mode

Just to clarify, when I mentioned manually setting the parameter, I meant an assignment within R such as:

sample_data$config$DESeq2$fitType <- sample_data$config$DESeq2.fitType

Not trying to use the "." form within the config.

ADD REPLY
0
Entering edit mode

That should work, does it? If not, can you a) include the package versions being used (sessionInfo()) and more of your script so I can see how you are getting the error? I can't reproduce it as you've described it:

data(tamoxifen_counts)
dba.show(dba.analyze(tamoxifen),
         bContrasts=TRUE)
tamoxifen$config$DESeq2$fitType <- "mean"
dba.show(dba.analyze(tamoxifen),
         bContrasts=TRUE)
ADD REPLY

Login before adding your answer.

Traffic: 896 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