Dear,
I am submitting a dummy model to the SC3 express lane and it looks like I can not get it right.
The problem happens when the prediction file is validated (ACCEPTED -> INVALID). The system complains that all samples are missing (e.g., TCGA-23-1123) but the output file contains almost the same samples (e.g., TCGA.23.1123 - notice the points instead of the dashes). Is this a problem with the validation script or somehow my script has changed the sample names?
Just in case, my last submission:
- ObjectId: 9640956
- Name: syn11017321
Best regards,
lctranchevent
Created by lctranchevent Thank you Mi for your help.
My problem is solved.
I just realized that I had accidentally removed a parameter of the _read.csv_ function (namely _check.names_, which should be set to FALSE to avoid the dashes being replaced by dots - among the other things it does).
Thanks again,
lctranchevent Ok, thank you for your help. I will investigate my script then.
lctranchevent Hi,
If you are using the same script, it should be TCGA-23-1123
The prediction file is not valid if not exactly the same name
Best,
Mi Thank you Mi for your help.
Actually, I started from there (very useful by the way).
My current script is still very similar (same methods to read and write data) but instead of having a model I just used "rnorm" to generate predictions randomly.
Thus, I only read the data to get the sample identifiers as to make predictions for these only.
Still, I want to double check with the organizers that there is no issue with the sample names on their side (using '.' instead of '-' as a separator).
Best regards,
lctranchevent Hi,
Please check the examples of submission script here: https://github.com/Sage-Bionetworks/NCI-CPTAC-Challenge-Examples/blob/master/sc3/Dry_Run_SC3.R
It might be the way to read the data and/or save the result.
Best,
Mi
Drop files to upload
[Solved] SC3 express lane submission error page is loading…