Taiji-pipeline / Taiji

All-in-one analysis pipeline
https://taiji-pipeline.github.io/
BSD 3-Clause "New" or "Revised" License
34 stars 9 forks source link

Got different node weight when switched the expression files #5

Closed avialee closed 4 years ago

avialee commented 4 years ago

Hi,

I am running Taiji on our ATAC-seq and RNA-seq data of WT and KO samples. We have two repeats on each type of data. Actually we already got the peak files from ATAC-seq and expression matrix from RNA-seq using our own pipeline. So I fed Taiji with peak files and expression files. However when I switched the 2 repeats of expression files in the input.yml file, I got different values in the nodes.csv. I have no idea how the weight of nodes is calculated. They should be same even the repeat file is switched, right?

Thanks.

kaizhang commented 4 years ago

Could you show me the changes you had made in the input.yml file?

avialee commented 4 years ago

ATAC-seq part is the same, only RNA-seq part is changed:

Before switched: ATAC-seq:

RNA-seq:

After switched:

ATAC-seq:

RNA-seq:

kaizhang commented 4 years ago

There shouldn't be any differences. In the RNASeq directory, there is a file named expression_profile.tsv. Please check whether the file stays the same between the two runs.

avialee commented 4 years ago

Here are the fist a few lines in expression_profile.tsv, it is also switched:

before switched: WT WT KO KO Ubl5 565.0413975 670.3987042 558.0231156 736.6467091 Gm38348 0.0 0.0 2.783157684 0.0 Ttc32 77.0089261 73.81095833 73.75367862 89.25241697

after switched: WT WT KO KO Ubl5 670.3987042 565.0413975 736.6467091 558.0231156 Gm38348 0.0 0.0 0.0 2.783157684 Ttc32 73.81095833 77.0089261 89.25241697 73.75367862

kaizhang commented 4 years ago

I see. This is probably a bug (Taiji-pipeline/Taiji-RNA-seq#1) introduced in v1.1.0. I will fix it this weekend.

avialee commented 4 years ago

Thanks!

kaizhang commented 4 years ago

Fixed in v1.2.1.