Hi, The sub challenge 1 expresse lane has patients that don't have any filename on WES columns. Those are discarded, since on training dataset, we could only use WES columns to predict the label. https://www.synapse.org/#!Synapse:syn10763434 Is because of that that I received the error "Must predict all patients in the goldstandard, and must match the correct study (study + patient)." in the end? Thanks Best regards

Created by Ruben Rodrigues rrodrigues
My team still hasn't submitted anything yet, however when looking at the simulated validation clinical file for sub challenge 1 no patients from the Hose Study have any associated molecular data. Will we only be expected to make predicitions for patients, that have a non "NA" value for the following headers? WES_mutationFileMutect WES_mutationFileStrelkaIndel WES_mutationFileStrelkaSNV RNASeq_mutationFileMutect RNASeq_mutationFileStrelkaIndel RNASeq_mutationFileStrelkaSNV Thanks
I found your submission.   You have to predict all the samples in the gold standard. You are missing all the samples from the DFCI validation cohort.   The entire DFCI validation cohort lacks WES data but has RNA-seq based vcfs *with some major caveats as discussed in the challenge webinar*. These RNA-seq based vcfs have a separate column in the clinical annotations files. Please look at the [simulated validation clinical files](https://www.synapse.org/#!Synapse:syn7222257) to get an idea of the file names.     I recommend reviewing the [webinar and its slides](https://www.synapse.org/#!Synapse:syn10288544) for more information on the peculiarities of the RNA-seq based vcfs.
can you provide you submission id?

Sub challenge 1 - patients without any VCF file on WES columns + Error: "Must predict all patients" page is loading…