Patience Please…

Mussel Methylation
Published

April 13, 2026

Plan of the Week: April 13 - April 19, 2026

High- level outline for the week. Adjusted daily to reflect progress of the day before

  • This week is all about biomarkers and methylation.

Monday - Methylation, methylKit run

Tuesday - UW-RUA, biomarkers

Wednesday - Biomarkers

Thursday - Biomarkers

Friday - Task management & ferry to FH/ Yellow

Saturday - Yellow Island Surveys

Sunday - Yellow Island Surveys

Plan of the Day

Granular level task list to accomplish the high- level goal outlined above

  • My primary goals for today are as follows:
    • Fix my RStudio/ R updates that crashes my local machine before it gets the printer treatment…
    • Get methylKit loaded in Raven or open an issue for help
    • Biomarker drafting

Projects Touched Today

  • Mussel Methylation

Progress Notes

  • I left off yesterday making a mess in R and RStudio. My local machine caught me slipping and I updated the working R version when switching between projects. Cue the BS.
    • To add insult to injury, I was already making an installation mess in Raven, so I wrapped up, made my notes, and considered leaving my laptop out in the rain…
    • A little sleep and a timer made all of the difference. First, I uninstalled and then re-installed R and then RStudio and fixed my local machine issue.
    • I also disabled my .RData and any .Rprofile docs temporarily until I can figure out what I setup that keeps making a mess in my stuff!
  • I pivoted to some UW-RUA traveler tasks that are time sensitive before returning to fixing my software.
  • Returning to Raven and the methylKit install from hell…
    • During my working block with KPJ, I decided to go back with fresh reasoning and install methylKit since I’d done some clear- headed thinking about the issues and Kristin is a great resource for coding conundrums.
    • First, working through all of the options back to back yesterday clouded my goal. The goal is to get the correct versions of BiocManager and methylKit installed so I can keep moving forward with the analysis.
    • I verified I was in my bio-cli environment to prevent competing versions of R, and maximize capabilities in my micromamba ‘container’ of sorts, aka bio-cli. Once I activated that, I navigated to R in terminal, verified the version (4.3.3), verified where my R libraries are located, and my working directory:

[1]"/home/shared/8TB_HDD_02/cnmntgna/R/x86_64-pc-linux-gnu-library/4.2" [2]"/home/shared/8TB_HDD_02/cnmntgna/micromamba/envs/bio-cli/lib/R/library"

[1] "/home/shared/8TB_HDD_02/cnmntgna"

  • I then went back to the BiocManager documentation, verified the version should be 3.18, not 3.16 (which I was trying yesterday) for R versions 4.3.x. I then crossed my fingers and watched the install…
Warning

Trying to solve an installation or actual coding problem 99.999999999999% requires me to take a step back and identify - out loud - what I am actually trying to do before accessing the internet! I lost a few hours yesterday trying to jump into the middle to fix something I already laid a foundation for. Go back to the basics…

Moving on…

  • BiocManager installed and version 3.18 verified.

    • methylKit failed. Error: object ‘key<-’ is not exported by 'namespace:data.table' Execution halted

    • Remembering to go back to the basics… I checked my paths, my working directories, and the methylKit documentation in the al2na GitHub repo. Even in my updated environment, R version 4.2 is overrunning 4.3 - what a bully. So I will fix that, and then the data.table issue.

    • Working with KPJ, I created new directories in my environment that are exactly the same except they terminal in 4.3 instead of 4.2, and re-pointed my environment to chose the newer libraries…

    • Next hurdle - the data.table package is too new- it is 1.18.2.1. Installed the older 1.16.4 version successfully.

find.package("data.table") packageVersion("data.table")

[1] "/home/shared/8TB_HDD_02/cnmntgna/R/x86_64-pc-linux-gnu-library/4.3/data.table" [1] ‘1.16.4’

  • Now, for methylKit… to fail. Segmentation fault (core dumped)… sounds egregious. More troubleshooting.

  • Halfway through trying to understand the error message, Kristin and I remembered we have this really nice conda/ mamba environment. So, I found the Bioconda GitHub page with instructions to install via my micromamba environment.

    • Helpful within that is the compatible versions list of the gazillion packages needed for this program.

    • rtracklayer was the next culprit, so installing and loading that, version 1.62.0, finally opened the door to methylkit version 1.28.0. I am rating this escape room 0/10.

  • Now for the true test- the code in the markdown.

    • Which failed.

  • After knocking out some UW-RUA tasks, I returned with a fresh set of eyes.

    • Problem 1 - no matter how many times (or ways) I set my working directory or activate my working environment, the markdown is fighting me.

    • Problem 2 - my launch/ .Renviron / .Rhistory files are in a fight to pull me into insanity.

    • Fixing problem 2 first will hopefully solve have the fight.

      • I became a recursive remover with impunity! I suspect the repeat attempts and fails created a bunch of mixed signals.

      • I did that because in my documentation perusal, I found that while conda and mamba are discussed kind of interchangeably, they are companions, not options - this helped me clarify my primary conflict - several ‘environments’ that could be one.

      • Once I renamed my old history and environment files, I changed my global options in R to stop loading from the previous session, restarted and followed the bioconda instructions to get my tools in one environment.

    • I now have a wgbs environment - outside of the methylation project - that I can use for this work and all subsequent projects. The environment includes the QC and alignment tools as well. I am very pleased!

      • It did activate within the markdown once I rendered the markdown as quarto & now my next step is to fix the actual coding errors I have created by modifying the chunks to test against the other environments and workarounds I attempted earlier today.

Outcomes: Products & Word Count

  • WGBS environment!

Today’s total: 0 words

Monthly total to date: 4420 words

Annual total to date: 37,093 words

Annual target total to date: 50,500 words

Next Up: Tomorrow’s Plan

  • UW-RUA and biomarker edits.