wyguo / ThreeDRNAseq

A pipeline for differential expression and differential alternative splicing analysis
https://github.com/wyguo/ThreeDRNAseq
GNU General Public License v3.0
61 stars 26 forks source link

[Help] #26

Open realextract opened 3 years ago

realextract commented 3 years ago

Thanks for this RNA-seq app.

As a first time user, I have encountered a few problems while using it. When I was in the "Data Pre-processing page", and tried to plot the PCA, it kept prompting "Please select a factor to group and colour the samples on PCA plots. ", while there was no selection for factors in the "Colour samples based on", or even I tried to type the condition, it still didn't work. Same thing happened to Contrast group selection etc, everything with the dropdown bar I didn't have anything.

PS: does this app fit for single sample RNA-seq analysis.

Thanks very much in advance!

wyguo commented 3 years ago

Hi, Thanks for using our App. In the PCA plot and the "Colour samples based on" box, the App will read the factors from the column names of your input metadata table. There might be something wrong with your metadata table if it didn't read the factors. Could you please show me the table? PS: the App does not work for single sample/condition. This App is designed to compare the expression between different conditions. You need at least two conditions to compare. Best, Wenbin

realextract commented 3 years ago

Hi Wenbin,

sorry for the confusion, I meant only one biological replicate there. And I used pair end mode in Kallisto for TPM, so I guess I don't have the sequence replicates here. Should I run Kallisto for each sequence replicate?

My metatable is attached.

Screenshot 2021-01-14 at 15 02 59

Thanks very much!

wyguo commented 3 years ago

Hi, The App does not work for dataset withouth biological replicates. The pipeline needs biological replicates to calculate the testing statistics for the significance of expression changes.

The sequence replicate comes from the same biological replicate beening sequenced multiple times. If you don't have sequence replicates, you just need to tick the "No" in the Data generation page -->Does the data have sequencing (technical) replicates?

From your metadata table, it seems that you only have one biological replicate for each condition. That's why the App and the PCA plot didn't work properly. Best, Wenbin