Hi everyone, I'm sorry if this has been answered before, but I can't find an answer. Within the test CLINICAL data, there are multiple samples for each subject ID, and when I predict on them and submit, an error bounces back saying there can't be multiple rows with the same ID. Should I be using the Sample ID instead? Thank you!

Created by Benjamin Wooden bwooden25
Thanks Solly, I appreciate the clarification.
Benjamin- We are asking for a single prediction for each sample in each of the time ranges (`Up to hour 0` and `Up to hour 24`). It is up to you how you come to those predictions so long as you limit yourself to gene expression from timepoints in the specified ranges. Solly
Ah, so in the case of multiple samples per patient, we should either use the consensus prediction or in the event of discrepant predictions, come up with a tiebreaker? The problem I'm having is, for instance, in the test_Phase1_CLINICAL file, there are two samples from Subject 4036, 4040, etc., representing different time points (i.e., -24 hrs and 0 hrs). Is the expectation that we either predict on only a single sample per subject, or that we collapse the predictions for the multiple samples into one? Thank you!
Your submission files should not have duplicate SUBJECTIDs. Please use the [submission templates](https://www.synapse.org/#!Synapse:syn5705156) to verify you are submitting in the right format. You should submit separate files for your `Up to hour 0` and `Up to hour 24` predictions.

Duplicated subject ids in test data page is loading…