I am currently performing DEG analysis using a plant F2population.
Specifically, I am conducting RNA-seq on an F2population created by crossing a dark-colored variety with a light-colored variety. I am then using R's “DESeq2” package for DEG analysis to identify genes involved in color intensity. However, when I split this population into two groups—dark-colored and light-colored—and performed DEG analysis to create a volcano plot, several genes with very high Log2 Fold Change and Adjusted P-values were detected. Yet, these genes showed extremely high expression levels only in some samples from the light-colored group, while other samples showed low expression or no expression at all. In this case, should such genes be filtered out?
For reference, hierarchical clustering was used to divide the group into two clusters, and the data used included SPAD, chlorophyll a, and chlorophyll b measurements.
I would appreciate your ideas.
•
written
by
CaERu _no_ korokoro •
0 answers
No answers yet.
Log in to answer this question.
-
Setting Thresholds and Interpreting Results in GWAS
written by CaERu _no_ korokoro •
-
How to perform GO enrichment in non-model organisms?
written by CaERu _no_ korokoro •
-
WGCNA sample-data preprocessing order
written by Reuben •
-
RNA-seq differentially expressed genes (DEGs) with low fold change values
written by singlecell_bio •
-
Why after normalization of scRNA-Seq data by DESeq2 the p-value of a large number of genes equal ze…
written by Zahra •
-
What is the best method for Differential Gene Expression analysis with log2 transformed data with m…
written by gillesmarodon •
-
High logfold change but padj > 0.1 in DESEQ2
written by thjnant
-
The Cancer Genome Atlas (TCGA) gene expression analysis
written by lovely.molbio •
-
High correlation between two genes or more genes indicative of similar localization?
written by chinsanity •
-
Maximum Number of Gene in a DESeq2 and edgeR
written by rieza.aprianto •
I think there are a lot of facets to consider when answering this question. So its a little difficult to give a good round response. A couple things that come to mind:
Firstly; and maybe most importantly is that you are dealing with samples that are not clones of one another; therefore there should be some genetic diversity, and it is possible that some of your F2 have a different chlorophyll profile due to the these differentially expression genes.
Second, the clustering you have done is on continuous measurements and therefore maybe the binary grouping doesn't capture a more varied spectrum. Maybe try modelling a continuous trait instead.
Thank you for your reply.
Would this mean using regression analysis with the raw numerical data instead of a simple two-group comparison? We are performing GWAS analysis since we have whole-genome data. However, I was hesitant to infer genes based solely on GWAS results, so I'm exploring alternative approaches.