Vignette Examples edits

[gabrielodom]

This is the continuation of Issue #28. Now that we have the examples written, we will need to enhance, modify, and edit these examples. This is the "catch-all" issue for general vignette example edits.

[gabrielodom]

For Example 4 (colorectal prediction): Steven will ponder on the odd shape of the survival ROC curves. For Example 3 (Multi-omics): See Issue #48, add circle plots (@lizhongliu1996). Remove SuperPCA part (@gabrielodom). For Example 2 (KIRP interaction): fix table (aespcaSignifSexInter_df) to show terms, estimate, std.error, statistic and p.value columns.

For all examples: can we access data in the vignettes from the internet? See the last comment of Issue #28.

[gabrielodom]

As mentioned in Issue #5, we now have a vignette. Thus, this thread will discuss edits to the vignette examples. Edits to the vignette itself will be handled through Word Track Changes.

[gabrielodom]

For example 2, I've fixed the tibble display for aespcaSignifSexInter_df and superpcaSignifSexInter_df within the KIRP_Sex_PC_Interaction.html report.

[gabrielodom]

For example 3, I've removed the super PCA section. @lizhongliu1996, if you plan to copy your results into the main report, please make any edits to the updated version: Xena Multi-Omics Ovarian/Reports/summary_ovarian_multiomics.Rmd

[gabrielodom]

@lxw391, in order to include the vignette you are working on, it must have an associated Sweave or Rmarkdown file (see the DESeq2 documentation for an example). Do you want to wait until we have the draft more polished before we build the Rmarkdown file?

[lxw391]

yes please edit the word file with track changes. we can build Rmd file after we finalize it.

[gabrielodom]

Please see the updated code and figures in Vignettes/odomEdit_lily_introduction_vignette.docx in the package Dropbox folder.

[gabrielodom]

Change from WnT to IL-1 pathway for ovarian proteomics and copy number.

[gabrielodom]

Add the pathway names onto the p-value / FDR columns (see page 4 of NAR draft). This should replace the column graph in chapter 5.

[gabrielodom]

@jamesban2015 will attempt to replicate table 1 from in Antonio's TCGAbiolinks paper. We will compare pathwayPCA to GoSeq, GSEA, and EnrichR.

[gabrielodom]

I will attempt to replicate figure 2 from in Antonio's TCGAbiolinks paper. Our sections will be

• Data: all functions from chapters 2 and 3
• Analysis: SuperPCA and AESPCA from chapter 4
• Results: all functions from chapter 5

"Check out MindManager or Xmind" - @lizhongliu1996

[gabrielodom]

From @lxw391, the remaining tasks for pathwayPCA are:

1. Update vignette to match work in NAR paper (change pathway to IL-1, pick NFKB1 gene) DONE
2. Assemble all figures to a PPT document DONE
3. Edit NAR draft

[gabrielodom]

Updated figures are either in the word document (odomEdit_lily_introduction_vignette.docx) or in the Figures/ subdirectory.

[gabrielodom]

@lizhongliu1996 and I worked on the powerpoint last night and this morning. We are still missing the Circo plot (from @lizhongliu1996 and Shirley) and the methods comparison table (from @jamesban2015). This file is Vignettes/Figures/NAR_draft_figures_20181211.pptx

@lxw391, please take a look when you get a chance.

[gabrielodom]

Added the Circo plot from Issue #48. Since the plot will only need minor edits, we can move the discussion here.

[gabrielodom]

I'm editing page 5 of the NAR draft. My comments and changes are in LWtest/NAR-DRAFT_12-10-2018.docx

[gabrielodom]

I've finished my edits to the NAR draft.

[gabrielodom]

Lily's vignette has replaced the "advanced users guide"

[lxw391]

@gabrielodom I got an error when running the vignette. Gabriel -- could you look into this?

[gabrielodom]

Did you pull changes and re-install the package? I’m assuming this is from colonSurv_df?

[lxw391]

Ok let me try rebuilding the package first thing tomorrow morning

Sent from my iPhone

On Dec 13, 2018, at 4:50 PM, Gabriel J. Odom notifications@github.com wrote:

Did you pull changes and re-install the package? I’m assuming this is from colonSurv_df?

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or mute the thread.

[gabrielodom]

@lizhongliu1996 had an issue with R CMD check. Basically, don't leave the Word output as the default vignette option. Use this option to build the vignette, but then change it back to html.

[lizhongliu1996]

That fix the problem!

[gabrielodom]

I think we can finally close this monster issue.