search

Coolgenome / iTALK

55 stars 20 forks source link

Raw input #2

Open RBBurl1227 opened 5 years ago

RBBurl1227 commented 5 years ago

Hello,

I was hoping you could share more about the input data structure. From the paper and github, it looks like you used a 10X Genomics pbmc dataset, and this 10X data was filtered and normalized in Seurat. What was the code for the data processing in Seurat, as well as how you generated the .txt file from the Seurat data. Thank you! I'm really looking forward to using this program.

ywang65 commented 4 years ago

Hi,

The input data is not necessarily processed by Seurat or any specific package/library. A data frame/matrix of expression data with rows representing cell id and columns representing gene names is fine. Our data is pre-processed using Seurat followed their pipeline. You can find their vignettes here.

Key0624 commented 4 years ago

Hi,

The input data is not necessarily processed by Seurat or any specific package/library. A data frame/matrix of expression data with rows representing cell id and columns representing gene names is fine. Our data is pre-processed using Seurat followed their pipeline. You can find their vignettes here.

Hi,

I’m wondering if it is ok not to correct the batch effect before I find the high expression genes by using ‘rawParse’ function？I'm really looking forward to your reply, thanks a lot!

ywang65 commented 4 years ago

Sorry for the late reply. My computer went wrong for a while. iTALK is just a visualization tool. We can't help you decide whether you need to correct the batch effect or not. Sometimes it's fine using the original combined data, but sometimes it's better to use the corrected one.

HWYDH commented 4 years ago

Hi, I have a data frame containing scaled data from a seurat subject. Following the instructions in your papers, I have set each column to be a gene and each row as a cell; in addition, i set a cell_type column containing cell type information. HOWEVER, i kept getting error for the first step:

highly_exprs_genes<-rawParse(data.frame(data),top_genes=50,stats='mean') Error in colnames<-(*tmp*, value = c("gene", "exprs", "cell_type")) : attempt to set 'colnames' on an object with less than two dimensions

Could you please let me know your insights ? Thanks a lot

Abelsufu commented 4 years ago

Hi ywang65,

Could you please extend your answer? I am having the same problems as RBBurl1227 and HWYDH, and I should say that so far, your replies didn't help. I agree with RBBurl1227 that to see you Rmarkdown to check how you did process the data in Seurat (more than Seurat Vignette) would be great.

Thanks in advance.