When conducting analysis, I am getting an error during the feature selection process. I can predict immunogenicity using Immunogenicity_Predict if I use the entire feature set, but prior I would like to use the feature selection function.
I receive the following error using the 'Features_MinimumFeatures' function;
Error in makeTask(type = type, data = data, weights = weights, blocking = blocking, :
Assertion on 'weights' failed: Contains missing values (element 1).
This occurs after the message 'converting preprocessed dataframes into a list of tasks'.
FullDataset is similar to the MHCI_Human dataset (containing the appropriate columns), and featureDF_Combi is produced after feature computation. Using the same data i can produce immunogenicity scores if I omit the feature selection process as above and instead use the entire feature set.
I am wondering what the issue may be here? The rest of the pipeline is working fine.
Hi,
When conducting analysis, I am getting an error during the feature selection process. I can predict immunogenicity using Immunogenicity_Predict if I use the entire feature set, but prior I would like to use the feature selection function.
When using the following code;
Features_MinimumFeatures( featureDFList=list(featureDF_Combi[Peptide%in%FullDataset$Peptide,]), metadataDF=FullDataset[,.(Peptide,Immunogenicity)][,Cluster:=.I], seedSet=1:5, corThreshold=0.75, featureNSet=100, criteria="intersect" )I receive the following error using the 'Features_MinimumFeatures' function;
Error in makeTask(type = type, data = data, weights = weights, blocking = blocking, : Assertion on 'weights' failed: Contains missing values (element 1).
This occurs after the message 'converting preprocessed dataframes into a list of tasks'.
FullDataset is similar to the MHCI_Human dataset (containing the appropriate columns), and featureDF_Combi is produced after feature computation. Using the same data i can produce immunogenicity scores if I omit the feature selection process as above and instead use the entire feature set.
I am wondering what the issue may be here? The rest of the pipeline is working fine.
Look forward to hearing from you
cheers
Paul