chip-seq large margin data viz

[tdhock]

TODO add to new gallery https://animint.github.io/gallery/
source is here https://github.com/tdhock/PeakSegFPOP-paper/blob/master/figure-large-margin.R
old publication here https://rcdata.nau.edu/genomic-ml/animint-gallery/2016-11-10-Max-margin-supervised-penalty-learning-for-peak-detection-in-ChIP-seq-data/index.html
referenced from https://animint-manual-en.netlify.app/ch16/ch16-change-point#Ch16-exercises

[ANAMASGARD]

Sir @tdhock

I've successfully modified figure-large-margin.R to add the required parameters for the new gallery format:

✅ Changes completed:

• Changed viz <- list(...) to viz <- animint(...)

• Added source="https://github.com/tdhock/PeakSegFPOP-paper/blob/master/figure-large-margin.R"

• Retained existing title parameter

• Removed animint2dir() call

  I am currently facing these problems

❌ Blockers preventing local execution:

1. Missing custom package: quadmod="2017.8.17"

   • Not available on CRAN
   • Appears to be a custom package from 2017
   • No GitHub path specified in packages.R

2. Missing data files:
   data/H3K4me3_TDH_immune/5/regions.RData
   data/H3K4me3_TDH_immune/5/counts.RData
   data/H3K4me3_TDH_immune/5/dp.model.RData

If you prefer I continue attempting to recreate the environment locally, could you please:

• Share what quadmod package is and where to obtain it
• Provide the missing .RData files
• Confirm if it's feasible to run this with R 4.5.2

THANK YOU !

[tdhock]

links to the data files are like this https://rcdata.nau.edu/genomic-ml/chip-seq-chunk-db/H3K4me3_TDH_immune/5/dp.model.RData and you should please add some code that downloads the file if it does not exist: if(!file.exists(local.file))download.file(remote.url, local.file)).
quadmod is on R-Forge but I do not think it is used in that script so I think you could delete that requirement (please simplify the code so that your source link points to a file with only minimal requirements to make the data viz).
yes it should be feasible to run with recent R-4.5.2.

[tdhock]

actually now that I look at the code, I think quadmod is required for some functions like standard.form.constraints() so please install from R-Forge https://quadmod.r-forge.r-project.org/

[ANAMASGARD]

Sir @tdhock,
I've been working on this and made progress, but I'm stuck on one issue.
What I did:

• Downloaded the data files from rcdata.nau.edu
• Installed quadmod from R-Forge and PeakError/penaltyLearning from GitHub
• Modified the code to use animint() instead of list() and added source parameter
• Removed animint2dir() call

The problem:
The original code uses showSelected.variable and showSelected.value syntax:
showSelected.variable=paste0(sample.id, "peaks"),
showSelected.value=peaksBut animint2 throws this error:

Error: Invalid parameter(s): showSelected.variable, showSelected.value.
Please use geom(showSelected = character_vector_of_variable_names)valent in the new animint2 syntax.

Question:
Is there a way to achieve this in animint2, or should I simplify the visualization to use basic showSelected="sample.id" instead? That would lose the per-sample peak selection feature but would still show the main visualization.

Thanks!

[tdhock]

you should use named clickSelects and showSelected instead, as explained in the error message
https://github.com/animint/animint2/wiki/Differences-with-other-packages#old-animint

[ANAMASGARD]

Sir @tdhock I have raised a PR related to this issue can you please review it and give your kind feedback , Thanks

#46