Error in `[.data.frame`(data, , 6) : undefined columns selected - error reading in bismark file

[monke...@hotmail.co.uk]

Hi,

I am trying to input bismark CpG context files to MethylKit

file.list=list("CpG_90.txt", "CpG_96.txt")

myobj=methRead(file.list, sample.id=list("IN190","CON96"), skip=1, header = FALSE, pipeline = "bismark" ,context="CpG", resolution="base", treatment=c(1,0), assembly="onilo11")

head(myobj)

but I keep getting the following error

Error in `[.data.frame`(data, , 6) : undefined columns selected

A line of the input file looks like this:

NB501042:94:H77HNBGX3:1:11101:20371:1063_1:N:0:GTAGAG

-

NT_167495.1

1410829

z

Has anyone had a similar problem, I've tried altering the arguements in the script and have run out of ideas now

Best wishes,

Rebekah

[Altuna Akalin]

please read the function help

"Other possible values for pipeline argument are 'amp','bismark', 'bismarkCoverage' and 'bismarkCytosineReport'. For 'amp' and 'bismark' the function expects a tabular format shown in the webpage (http://github.com/al2na/methylKit). "amp" and "bismark" expect identical input and are kept for historical reasons. 'amp' was a pipeline used in Akalin et al. 2012 Plos Genetics paper, publicly available in googlecode."

--
You received this message because you are subscribed to the Google Groups "methylkit_discussion" group.
To unsubscribe from this group and stop receiving emails from it, send an email to methylkit_discussion+unsub...@googlegroups.com.
To post to this group, send email to methylkit_discussion@googlegroups.com.
Visit this group at https://groups.google.com/group/methylkit_discussion.
For more options, visit https://groups.google.com/d/optout.

[monke...@hotmail.co.uk]

That was the first thing I did and it didn’t really help, I tried all the pipelines and options and still get the same error

[Altuna Akalin]

the input format is wrong, check the example files that comes with the package 

On Wed, May 16, 2018 at 9:34 PM, <monke...@hotmail.co.uk> wrote:

That was the first thing I did and it didn’t really help, I tried all the pipelines and options and still get the same error

--
You received this message because you are subscribed to the Google Groups "methylkit_discussion" group.

To unsubscribe from this group and stop receiving emails from it, send an email to methylkit_discus...@googlegroups.com.
To post to this group, send email to methylkit_...@googlegroups.com.

Visit this group at https://groups.google.com/group/methylkit_discussion.
For more options, visit https://groups.google.com/d/optout.

--

Sent from mobile, excuse the brevity

[monke...@hotmail.co.uk]

Hiya, yes that is what I was meaning. If the software is meant to take Bismark files with the option —bismark, I am confused as to why it won’t take the bismark format?

Is it not set up for this?

Best wishes,

Rebekah

[Altuna Akalin]

you can use the bam files from bismark to call methylation using methylKit, "bismark" option is meant for those cases, where methylKit calls methylation from bismark bam files and saves them as text files. Those text files have the identical format to our old pipeline "amp". 

.Otherwise, you can call methylation fraction/percentage in multiple ways and those can be read in via methylKit. The files you are trying to read do not contain methylation fraction/percentage and coverage per CpG. 

--
You received this message because you are subscribed to the Google Groups "methylkit_discussion" group.

To unsubscribe from this group and stop receiving emails from it, send an email to methylkit_discussion+unsub...@googlegroups.com.
To post to this group, send email to methylkit_discussion@googlegroups.com.

[monke...@hotmail.co.uk]

Ah okay! Cheers! Wasn’t clear from the help file. I did try the bam file, but it kept causing a “fatal error” in Rstudio. I will try another method and import into methylkit. As I would really like to use methylkit for my analysis.

Thank you again,
Best wishes,

Rebekah