edroaldo
commented
6 years ago Hi,
You are correct. That matrix comes from a tsne analysis. It is normal to obtain different results during tsne analysis due to the the way tsne works. You can also observe different results with different packages. I often use the package 'tsne' but I also use Rtsne. If you want to reproduce the same tsne analysis as in the tutorial, you can follow the first steps from the StemID tutorial here: https://github.com/dgrun/StemID/blob/master/RaceID2_StemID_sample.R, which was how I generated the object "matrix.R" in the tutorial. Please, let me know if you have further questions. You do not need to apologize!
Thank you for your interest in our work!
2018-03-30 15:05 GMT-04:00 LaptopY notifications@github.com:
Hello, In order to apply CellRouter to my own datasets in the future, I would like to try to complete (and master) your tutorial with the mouse bone marrow single-cell RNA-sequencing dataset (I am a new user in R). You mentionned at the beginning of the tutorial that "CellRouter assumes that you are providing a gene expression matrix with genes as rows and single-cells as columns." I understand that these are the normalized and filtered_expression files provided.
However, I was wondering how to obtain in the future the file named "matrix.R" that you use in "Create Cell Router" part. It is a matrix with two columns and single cells in rownames. You mentioned to rename these columns as tSNE1 and tSNE2, so I assumed that this matrix could come from a "tsne analysis", is it correct or am I wrong ?
I tried to do this with filtered_expression files using fdata1<-t(fdata) # transposition "test<-Rtsne(as.matrix(fdata1), dims=2, check_duplicates=FALSE, pca=T, perplexity=30, theta=0.5, max_iter = 5000) matrix <- as.data.frame(test$Y) # matrix 2 columns 1007 rows Using this matrix I obtained a different graph, "similar" but different.
could you indicate me what are the main steps to obtain this "matrix" ? (I apologize for my English)
Thank you
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/edroaldo/cellrouter/issues/3, or mute the thread https://github.com/notifications/unsubscribe-auth/AJqUR6IFnXHGIlW0LSpxtSOl8ktrrt8Sks5tjoH9gaJpZM4TCB62 .
-- Edroaldo
Hello, In order to apply CellRouter to my own datasets in the future, I would like to try to complete (and master) your tutorial with the mouse bone marrow single-cell RNA-sequencing dataset (I am a new user in R). You mentionned at the beginning of the tutorial that "CellRouter assumes that you are providing a gene expression matrix with genes as rows and single-cells as columns." I understand that these are the normalized and filtered_expression files provided.
However, I was wondering how to obtain in the future the file named "matrix.R" that you use in "Create Cell Router" part. It is a matrix with two columns and single cells in rownames. You mentioned to rename these columns as tSNE1 and tSNE2, so I assumed that this matrix could come from a "tsne analysis", is it correct or am I wrong ?
I tried to do this with filtered_expression files using fdata1<-t(fdata) # transposition "test<-Rtsne(as.matrix(fdata1), dims=2, check_duplicates=FALSE, pca=T, perplexity=30, theta=0.5, max_iter = 5000) matrix <- as.data.frame(test$Y) # matrix 2 columns 1007 rows Using this matrix I obtained a different graph, "similar" but different.
could you indicate me what are the main steps to obtain this "matrix" ? (I apologize for my English)
Thank you