Details
-
Type:
Task
-
Status: Closed (View Workflow)
-
Priority:
Major
-
Resolution: Done
-
Affects Version/s: None
-
Fix Version/s: None
-
Labels:None
-
Story Points:3
-
Epic Link:
-
Sprint:Summer 2 2023 May 29, Summer 3 2023 June 12
Description
Create a new Markdown that uses 72_F3H_PollenTube/results/muday-144-SL5_counts-salmon.txt as input and outputs the same exact file except that the values are scaled using the EdgeR "cpm" function.
The Markdown should answer the question: Did the scaling "work" ? That is, do the observed differences in expression (e.g., scaled numeric values per gene per sample) reflect actual biological variation and not sequencing depth variation.
Identify control genes that you don't expect to exhibit changes in expression in response to the treatments being tested.
Ask the Muday lab (Gloria) for some examples.
Make barcharts that show before and after scaling for a set of genes you do not expect to change in response to time or treatment. If the scaling worked, then all the samples should have the same height, or all samples within the same time duration should have the same height.
Name it: MakeScaledCountsFile.Rmd and save the output to "results"
Name the output the same thing as the input but with "scaled" in the name.
Attachments
Issue Links
Activity
Waiting on email response from muday lab to know which control genes to use.
We needed a list of gene names for genes that never vary from Muday lab, but they are not able to provide such a list due to not enough being known about pollen tube gene expression.
Previously, we found no over-expression of F3H gene in the over-expression lines in the the older, two-hour gene expression data set. See: Markdown report titled "Assessing flavanone 3 beta-hydroxylase (F3H) expression."
For this new Markdown, let's examine RNA-Seq count levels of F3H pre- and post-scaling.
The question to answer in the Markdown could be: "Did scaling via the cpm function reduce variance between replicates?"
Decided to not bother with assessing scaled vs unscaled counts. Instead, focused on the F3H gene.
Added new files to the repository folder 72_F3H_PollenTube:
- InvestigateF3H-and-MakeScaledCounts.Rmd
- InvestigateF3H-and-MakeScaledCounts.pdf
- results/muday-144-SL5_counts-salmon_scaled.txt
Please review using the same review protocol as for linked issue IGBF-3356.
I do not see these file in the bitbucket repo and I do not see that you have a fork made. Could I have some help finding these new files you added?
Attn: [~aloraine]
Sorry for the confusion. I broke protocol and pushed the changes to the main branch directly to the team repository. The files can be found here:
Testing:
Proof-read knitted .PDF: Everything on the knitted pdf looks great!
Read .Rmd file: The date isn't "`r Sys.Date()`" but that's ok.
Check in-line R code: All in-line code is knitted properly.
Attempt to knit .Rmd file: I knitted the markdown from my machine and everything came out correctly and the same.
Open the new output file: Output file is accurate and useful.
Nice job!
Moving ticket to done!
Resource for using cpm() function:
https://bitbucket.org/mdavis4290/molly-flavonoid-rnaseq/src/main/72_F3H_PollenTube/FindDifferentiallyExpressedGenes.pdf
Resource for barplots:
https://bitbucket.org/mdavis4290/molly-flavonoid-rnaseq/src/main/72_F3H_PollenTube/MakeMdsPlots.pdf