Search
Question: Errors with champ.block
0
gravatar for tiroshamit
4 weeks ago by
tiroshamit0
tiroshamit0 wrote:

Hi ChAMP users,

I am running the Champ according to the steps in the documentation. However, when trying to analyze the champ.block, I get errors, and from some reason - a different error in each run.

My data includes 37 samples analyzed by EPIC array. All the analyses up to this level (QC, norm, SVD, DMP) went smoothly.

This is the problematic line:

myBlock <- champ.Block(beta=myNorm,pheno=myLoad$pd$Sample_Group,arraytype="EPIC")

And here are the error lines:

Error in names(revOrder) <- seq_along(ret$idx) :
  attempt to set an attribute on NULL

OR

Error in x[[1]] : subscript out of bounds

Any idea how to solve this? Anyone else had this problem?

Thank you

Amit

 

 

 

 

ADD COMMENTlink modified 4 weeks ago • written 4 weeks ago by tiroshamit0
0
gravatar for Yuan Tian
4 weeks ago by
Yuan Tian0
Shanghai Institute for Biology Science, Shanghai, China
Yuan Tian0 wrote:

Hello Amit:

Could you provide two things: one is sessionInfo, I want to know which version of ChAMP you are using, and secondly, could you paste more information here? I remember all champ function should return many messages, by providing them I could locate problem faster.

Best

Yuan Tian

ADD COMMENTlink written 4 weeks ago by Yuan Tian0

Hello Yuan,

Thank you very much for this great tool and for your prompt response.

I have solved this issue, it seems that I had to use the ChAMP.DMR before the ChAMP.BLOCK. Anyway, it runs ok now.

However, in the meanwhile I had another issue, that I saw it was mentioned previously. I am trying to use the DMR.GUI, and the process freezes after the message "Section 1: Calculate DMP Done".

I saw in another forum that there was an issue with data(probe.features.epic), which I found in the spot in the script relevant (?) to my current problem. When I tried to upload it separately, nothing came up. 

Here are my session details: >> Package version 2.8.9 loaded <<

This probably stems from something I missed on the way, but I will really appreciate any help.

Many thanks

Amit

ADD REPLYlink modified 4 weeks ago • written 4 weeks ago by tiroshamit0

Hello Amit:

DMR.GUI() would firstly calculate significant DMPS then map them back to DMRs, so after calculating DMP, there is a process mapping DMPs back to DMRs, in my memory, it would take some longer time. Have you try my Demo data? If Demo Data works, I think there is no problem with function.

Best

Yuan Tian

 

ADD REPLYlink written 4 weeks ago by Yuan Tian0
0
gravatar for tiroshamit
4 weeks ago by
tiroshamit0
tiroshamit0 wrote:

Hello Yuan

I tried running the EPIC test data and had the same problem.

    Edit: I ran the 450K data and it worked smoothly.

In the global environment variables I see "probe.features", but not "probe.features.epic".

Here is the script I am using:

library(ChAMP)

data(EPICSimData)

champ.QC()

myNorm <- champ.norm(beta=myLoad$beta, arraytype="EPIC")

champ.SVD(beta=myNorm,pd=myLoad$pd)

myDMR <- champ.DMR(beta=myNorm,pheno=myLoad$pd$Sample_Group, method="Bumphunter")

DMR.GUI(DMR=myDMR)

 

And here are the messages I get until the R freezes while running DMR.GUI:

!!! important !!! Since we just upgrated champ.DMP() function, which is now can support multiple phenotypes. Here in DMR.GUI() function, if you want to use "runDMP" parameter, and your pheno contains more than two groups of phenotypes, you MUST specify compare.group parameter as compare.group=c("A","B") to get DMP value between group A and group B.

[ Section 1: Calculate DMP Start  ]

  You pheno is character type.

    Your pheno information contains following groups. >>

    <control>:8 samples.

    <case>:8 samples.

  Your pheno contains EXACTLY two phenotypes, which is good, compare.group is not needed.

Calculating DMP

[===========================]

[<<<<< ChAMP.DMP START >>>>>]


!!! Important !!! New Modification has been made on champ.DMP(): 

    (1): In this version champ.DMP() if your pheno parameter contains more than two groups of phenotypes, champ.DMP() would do pairewise differential methylated analysis between each pair of them. But you can also specify compare.group to only do comparasion between any two of them.

    (2): champ.DMP() now support numeric as pheno, and will do linear regression on them. So covariates like age could be inputted in this function. You need to make sure your inputted "pheno" parameter is "numeric" type.


[ Section 1:  Check Input Pheno Start ]

  You pheno is character type.

    Your pheno information contains following groups. >>

    <control>:8 samples.

    <case>:8 samples.

    [The power of statistics analysis on groups contain very few samples may not strong.]

    pheno contains only 2 phenotypes

    compare.group parameter is NULL, two pheno types will be added into Compare List.

    control_to_case compare group : control, case

[ Section 1:  Check Input Pheno Done ]

[ Section 2:  Find Differential Methylated CpGs Start ]

  -----------------------------

  Start to Compare : control, case

  Contrast Matrix

          Contrasts

Levels     pcontrol-pcase

  pcase                -1

  pcontrol              1

  You have found 780385 significant MVPs with a BH adjusted P-value below 1.

  Calculate DMP for control and case done.

[ Section 2:  Find Numeric Vector Related CpGs Done ]

[ Section 3:  Match Annotation Start ]

[ Section 3:  Match Annotation Done ]

[<<<<<< ChAMP.DMP END >>>>>>]

[===========================]

[You may want to process DMP.GUI() or champ.GSEA() next.]

[ Section 1: Calculate DMP Done  ]

 

Thank you again,

Amit

ADD COMMENTlink modified 4 weeks ago • written 4 weeks ago by tiroshamit0
Please log in to add an answer.

Help
Access

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 2.2.0
Traffic: 258 users visited in the last hour