Bug Update: The abolsute value function, abs(), worked with the code and counts genes with a -1 and 1 log2foldchange threshold.

Edits in markdown:

v = rs$padj<=Q & abs(rs$log2FoldChange)>=lfcThreshold

Checking the method code: To check if this method was working accurately, I made a separate script that grabbed just one group and performed the same strategy and saved the results as a csv and went through the file and saw that the statement was true and that the adjusted pvalues were all less than or equal to 0.01 and the log2fold change threshold was >= 1 or <=-1.

res= read.csv('MvW-SL5.txt',sep = "\t",stringsAsFactors = F,
                  header=T)

V.34.75 <- res[res$group1 == 'V.34.75',]
V.34.75

V.34.75 <-data.frame(V.34.75)
final1 <- subset(V.34.75, padj<=0.01 & log2FoldChange>=1)
final2 <- subset(V.34.75, padj<=0.01 & log2FoldChange<=-1)
final <- rbind(final1, final2)
final <- data.frame(final)
nrow(final)
write.csv(final, "/Users/mollydavis333/Desktop/finalresults.csv", row.names=FALSE, quote=FALSE) 

Note: There are more result conditions required in the markdown compared to my script so numbers my vary by small amounts but the statements are accurate and perform the required task of fixing the threshold bug in the markdown.

Branch: https://bitbucket.org/mdavis4290/molly4-flavonoid-rnaseq/branch/IGBF-3487

Files to review:

• FindMutantVsWildtypeDEGenes-DESeq2.Rmd
• FindMutantVsWildtypeDEGenes-DESeq2.pdf

Check to see if the data frame 'comps' accurately displays the correct number of genes that have an adjusted pvalue less than or equal to 0.01 and the log2fold change threshold was >= 1 or <=-1.

Reviewed and ready for pull request.

However, before you do that, please also fix the same bug that appears earlier in the code.
Change: :

  vs[[assembly]] = rss[[assembly]]$padj<=Q &
    rss[[assembly]]$log2FoldChange>=lfcThreshold

to:

  vs[[assembly]] = rss[[assembly]]$padj<=Q &
    abs(rss[[assembly]]$log2FoldChange)>=lfcThreshold

attn: [~molly]

PR: https://bitbucket.org/hotpollen/flavonoid-rnaseq/pull-requests/32

Testing:

Markdown = FindMutantVsWildtypeDEGenes-DESeq2.Rmd

• There was a bug with the 'comps' table shown in the knitted markdown. The bug had an issue with the threshold only counting greater than or equal to 1 and not less than or equal to -1 also. This made the sum of DE genes in the table wrong. This ticket fixes this issue and updated the code with the abs() function.
• Please pull this file down to your machine (https://bitbucket.org/hotpollen/flavonoid-rnaseq/src/main/72_F3H_PollenTube/FindMutantVsWildtypeDEGenes-DESeq2.Rmd) and knit the markdown.
• If knitting has no errors, next check the FindMutantVsWildtypeDEGenes-DESeq2.pdf out put file and check the 'comps' table in the pdf as well.

I pulled the code down.

I knitted. All good.

Created a PDF.

But I have not yet had a chance to look at the details in the PDF.

And I have been separated from laptop where I do this! (Main campus woes)
Will recreate on this machine.

I have read through the PDF, and that looks good.

It is readable and straightforward.

Let's check it off as done?

Thanks [~RobertReid]

Moving to Done!