Closed DomenicoSkyWalker89 closed 4 years ago
Hi there, Thank you for being interested in cTP-net. Actually, cTP-net has never been tested on R 3.6. Seems to me that the reticulate had an major updates on R 3.6 and that might be problematic. My coming weeks are a bit crazy. I should be able to address your issue next Thursday or Friday. Just remind me if you haven't heard an update from me.
-ZZ
Thank you I will wait.
Hi there, have you fixed the issue? Thanks a lot for you time.
kind regards, Domenico
Hi Domenico,
Thank you for the reminder. I am going to work on it today. I'll let you know when I finished.
-ZZ
Hi Domenico,
I apologize for the late response. It was no easy fix. Since majority of the R users are still using R3.5, I would strongly suggest you too. R.3.6.1 is a very much new release and may subject to different bugs. Many package users haven't fully test their package there. In R, it allows multiple version existed on a same computer. So it should be very easy to install R3.5 and then install cTP-net in the 3.5 environment.
Hope this helps.
Zilu
Hi Domenico,
I recently updated cTPnet and have tested the installation on R3.6.1 without any error. Please remove the original package and reinstall. Please let me know if you have any further question. I will close the issue for now.
Zilu
Hi Zilu, thaks a lot for your support. i'm doing the analysis using SAVERX to denoise my data (60k CD8 cells) and then use cTPnet as you suggest, but i encounter this error:
please help me, Domenico
also if I run cTPnet without denoise my data using SAVERX i had this different issue:
Hi Domenico, Make sure you have update cTPnet python package via pip. You should uninstall it and reinstall it again.
About the issue with SAVER-X, I think you should better post it on SAVER-X issue here (https://github.com/jingshuw/SAVERX/issues). Another option is to use the online portal for SAVER-X denoising (https://singlecell.wharton.upenn.edu/saver-x/)
Best, Zilu
Ok Zilu, tomorrow i will try what you suggest.
thanks a lot, Domenico
Hi Zilu, cTPnet now works fine on my data. My sample is composed only by CD8 + T cells and i don't know if the neural network work fine on this tipe of data. I will study the obtained data and I will understand if they are reliable with my expectations and the result of scRNAseq.
THANKS A LOT.
Best, Domenico
Hi Domenico, Glad it worked! Just to keep in mind that cTP-net capture relative abundance. If you didn't see a high CD8 level, this might due to all of your cells are CD8+. Let me know if you have any more question. I will close the issue for now. -ZZ
hi, thanks for developed this interesting tools. I encountered an issue during the installation, have you any suggestion to solve it?
kind regards