Q: Submission 1 /2/3 requires not using 24/44/64 genes in BDTNP 84 genes respectively in the training step. How about actual position labels computed by DistMap using 84 genes?
A: You can use them, this is your tool for training your predictions
Created by Thomas Yu thomas.yu Hi yes but actually you can use all RNAseq, no limits on this
thanks
Pablo Hi Pablo,
I want to make sure if I'm wrong.
First, I choose 20/40/60 genes from 84.
Then, I get two data(bdtnp20/40/60 + RNAseq20/40/60).
Next, I can do whatever I want with two data, like DistMap, Machine Learning.
Finally, we submit our predict position.
Am I correct?
Thanks Indeed you can, this is how you can evaluate and train your predictions using the 20/40/60 genes in situs +RNAseq from all 1297 cells OK, but after pre-selection of the 20/40/60 genes subset, I can still use the actual positions calculated for all cells? (these actual positions were calculated using 'binarized_bdtnp.csv' and 'dge_binarized_distMap.csfv', each containing information on all 84 driver genes). Just to be clear, you cannot extract features from the 84 in situs in an exhaustive way, you have to pre-select the 20,40 or 60 genes and consider as if the rest did not exist. Can you please clarify this further? This seems contradictory.
The DistMap positions were calculated using the information from all 84 genes. If I use this information, what is the point of only using 20/40/60 genes directly from BDTNP? The omitted information is still available to me.
Drop files to upload
Webinar Question: Usage of actual position labels computed by DistMap page is loading…