DivRef is a resource for including common human variants and haplotypes in CRISPR off-target searches. I rebuilt the generation workflow to make its inputs, assumptions and population choices easier to inspect and change. blog.fulcrumgenomics.com/p/a-crispr-o...

A CRISPR off-target search is only as good as the sequences it searches

Rebuilding DivRef so CRISPR off-target searches can account for human variation

blog.fulcrumgenomics.com

A CRISPR off-target search cannot report a sequence it was never given. @ameynert.bsky.social explains how DivRef brings common human variants and haplotypes into the search space—and how her rebuild complements @timd.one work on Sassy. Read Alison’s post: blog.fulcrumgenomics.com/p/a-crispr-o...

A CRISPR off-target search is only as good as the sequences it searches

Rebuilding DivRef so CRISPR off-target searches can account for human variation

blog.fulcrumgenomics.com

Hand-wrote almost 1300 lines of documentation plus test data to exercise a set of workflows. Still got it! 💪

Reading Jane of Lantern Hill (better than Anne of Green Gables IMO!) to my 10yo. She's totally engaged. It reminds me that reading aloud used to be a perfectly normal family activity of an evening, regardless of age. We listen to podcasts and audiobooks in the car together, but it's not the same.

intheweights.com going around at work today. Not a bad summary - though most models still think I work in Edinburgh. It's handy having an unusual last name sometimes - worth the trade off of having to spell it out every time.

TCL (Today Claude Learned): When a temporary environment is required for testing something, first go for `pixi exec`. If it becomes more complex, I want to run it myself, or needs repeating, write a temporary `pixi.toml`, install the environment, and use `pixi run`.

Old school. My teen and I are currently working our way through Rush Hour, one puzzle a day over breakfast. Unsurprisingly, his young brain is very good at it, but I always get there in the end! 🚗

TIL - nailed down a pytest-workflow tag / GHA matrix set up where the tests were launching workflows on an external platform and copying back the output for the checks. Some tricky concurrency pieces to sort out, but this pattern is going to be highly useful!

My first real job was as a software developer writing file format conversion tools. Almost 25 years in bioinformatics and many LOC later I'm still doing it.

Now that Sassy had been published and presented, it's time for it to break out of my bubble :) curiouscoding.nl/posts/sassy/ www.reddit.com/r/rust/comme...

Open-source bioinformatics works best when the tools are fast, inspectable, and reproducible. Our recent releases cover UMI workflows, sequencing QC, reference identification, ECS analysis, and HGVS parsing. See the one-page overview: shorturl.at/uo8iX