Details
-
Type:
Task
-
Status: Closed (View Workflow)
-
Priority:
Major
-
Resolution: Done
-
Affects Version/s: None
-
Fix Version/s: None
-
Labels:None
-
Story Points:2
-
Sprint:Fall 2 2023 Sep 17, Fall 3 2023 Oct 2
Description
We observed from previous work that VF36 and F3H are very similar with respect to gene expression patterns. To understand how the are mutant differs from wild-type, we will focus on comparing the are mutant to the VF36 cultivar.
Goal: Find and interpret how gene expression in are differs from gene expression in VF36
Rationale: Based on past PCA and MDS plot analysis, we hypothesize there will many genes with expression differences between ARE and VF36. And yet, as far as we know, ARE is nothing but VF36 with a single point mutation in the F3H gene. Why so many differences in expression?
Strategy:
Create Markdown that perform differential analysis comparing are to VF36 within the control 28 degree temperature and under one time.
Attachments
Activity
Testing:
- Markdown: No issues with build or knitting.
- PDF: The pdf is comprehendible and visually has no issues
- Output files: Output files make sense and have no issues
- IGB Images: Images show genes from the SL4 genome and compare A vs. V genotypes
Everything looks great! Moving to done!
Thanks for the above comments [~molly]!
I made a new Markdown that does compares are control samples to VF36 control samples, including time as a variable.
My rationale for including time as a variable was to ensure that the model could take time effects into account when comparing are to VF36 samples. I did not include the treatment samples because we know that temperature has an effect on gene expression, and I did not want to make results harder to interpret by including the treatment as a factor.
Due to our tight time line I committed my changes to the main branch.
For review and/or testing, please examine the following files:
Images (from IGB) confirming the DE observation for two genes:
For this, I used DESeq2. I analyzed SL4 and SL5 genes separately, and saved the results to the ".txt" files above.
I also checked some of the genes using IGB. I visually confirmed that coverage graphs for genes with small log2 fold-changes between are and VF36 looked pretty much the same. Larger fold-changes were more obvious, even when the expression levels were low, e.g., scaled coverage graph peaks of 6 or smaller.