randel / mixrf Goto Github PK

Few genes in my dataset do not have eQTLs. The eqtl.lis argument (of MixRF.impute) maps these genes to an empty character array. However, I am unable to run the MixRF.impute command with these arguments and this error shows up:

Error in { : task 1 failed - "subscript out of bounds"

Should all the genes in eqtl.lis must have at least one associated eQTL (SNP ID)? Otherwise, is there any other way to represent the absence of eQTLs in these genes?

Thanks in advance.

possible to have a demo showing input format of GTEx data?

How exactly do we obtain the arguments eqtl.lis and snp.dat for the function MixRF.impute from publicly available GTEx data (RNA-seq)? The documentation says that we have to obtain the snp.dat argument, after which we can generate the eqtl.lis argument by making use of the code in eqtl.r. So in order to obtain genotype matrix, do we have to make use of the eqtl data available for each tissue in the GTEx portal?

Dear Jiebiao,

I am receiving on of two errors for 10 of the 12 models I want to run using your function

Error in RET@prediction_weights(newdata = newdata, mincriterion = mincriterion, : cannot compute out-of-bag predictions for observation number 6 |
Error in model@dpp(...). : missing values in response variable not allowed

when running the MixRFb function.

I have a binomial (0 or 1) outcome variable (let's call this y1) and four predictor variables. The weird thing is that it works on another similar outcome variable (let's call this y2). I do exactly the same, the data is the same, only the prevalence of 1's is different. y1 (which doesn't work) has low prevalence of 1's and y2 has okay prevalence of 1's. Could this be the problem?

I have a total of 12 y variables for which only 4 run the full iterations, and 8 run into one of these two errors after the first iterations0.

Could you help me in the right direction? Thanks for the package and the help!!

Hello,
Is there a possibility to add permutation tests in the MixRF function (as rfPermute) or even to include random effects on slopes?
King Regards

For correlation=TRUE, I always confront the following issue:

Calls: MixRF.impute -> get.CVlist -> sample -> sample.int
In addition: Warning messages:
1: In split.default(sample(av.samp), 1:nCV) :
  data length is not a multiple of split variable
2: In split.default(sample(av.samp), 1:nCV) :
  data length is not a multiple of split variable

For example, in case setting number of patients=600, number of tissues = 6 (default nCV or nCV = 3).

Have the ability to use Y as a multiclass variable

Thanks for developing this package. It's neat and intuitive.

Would you consider enabling manipulation of the randomForest parameters via the MixRF arguments? The user may want to adjust things like ntree and mtry.