Problem extracting differentially expressed genes using diffxpy
1
1
Entering edit mode
2.5 years ago
zenbio ▴ 10

Hi,

I am new at performing RNAseq single cell analysis, so perhaps my question is naive. But here it is: I used the diffxpy package to find differentially expressed genes between two conditions in a given cluster of cells using wald test. To extract the genes with significant log fold change, I apply the following condition on pandas results dataframe:

condition_m = (test_megak.summary()["qval"] <= 0.05) & (abs(test_megak.summary()["log2fc"]) >= 1) 
gene_indices_m = test_megak.summary().index[condition_m]
gene_indices_list_m = gene_indices_m.tolist()

I end up with a list of around 3000 indices out of the 23000 gene tested. But when I run the volcano plot of diffxpy on the test results, it only shows a few point meeting my criterias for differentially expressed (i.e corrected pval <= 0.05 and abs(log2fc) >=1):

test_megak.plot_volcano(corrected_pval=True,log10_p_threshold=-20,log2_fc_threshold=5, min_fc=2, alpha=0.05, size=20)

enter image description here

I actually know that there should be only a handful of genes differentially expressed, certainly not 3000. So the volcano plot corresponds more to what I am expecting.

Can anybody tell me what I am doing wrong? How can I extract the genes names corresponding to the colored points I see on the volcano plot using the diffxpy api?

diffxpy genes differentially expressed • 1.4k views
ADD COMMENT
0
Entering edit mode
2.5 years ago
Marco Pannone ▴ 790

I would suggest you run DE analysis using DESeq2, which is more of a gold standard in RNA-seq analysis. You can find documentation here with proper examples: https://bioconductor.org/packages/release/bioc/vignettes/DESeq2/inst/doc/DESeq2.html

All you need to know about DESeq2 is written there, also showing you how to make plots and extract DEGs accordingly to your desired thresholds.

Good luck!

ADD COMMENT
0
Entering edit mode

DESeq2 is definitely not the gold standard for scRNA-seq.

ADD REPLY
0
Entering edit mode

It's totally fine to perform DE analysis using DESeq2 also with scRNA-seq data, just keeping in mind to change some specific parameters, as also stated in the official documentation.

ADD REPLY

Login before adding your answer.

Traffic: 3519 users visited in the last hour
Help About
FAQ
Access RSS
API
Stats

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.6