naturepoker
@binomicalabs
Binomica Labs independent researcher by day, warehouse worker by night Small thoughtful science #phylogenetics #archaea #evolution #bioinformatics #phage @naturepoker@genomic.social
Quick film photography snap from my trip to Washington DC National Gallery. It was a welcome relief from everything outside those walls.
Wildfire smokes haven't cleared up yet. Looking forward to the supposedly pretty big thunderstorm later this morning. #nyc
I think @martinsteinegger.bsky.social lab has some of the best lab sticker designs I've seen. Definitely going on my nanopore. Thanks for the chat!
Also spent some quality time at the Smithsonian museum of American Indian. The exhibit on Native American art 1900s onward was extremely compelling. And the exhibit on the history of treaties between Indian nations and the USA... Man that cut deep.
In DC again for ISMB/BOSC. DC metro is certainly something else. Some of the entryways basically look like bunker entrances. Also, the platform's air conditioned. Something NYC will never have for many reasons har har.
Got it. Yeah we just sent some of the genome extracts to an Illumina vendor for comparison data, spent some time on asm to asm synteny and alignment stat script too (fig). Looking forward to analyzing properly soon. Thanks for responding & all the write-ups you do! They're great learning materials
And here's an animation of a demo set - sequencing quality for this run was... Not good
Mad 4AM dash to wrap up a #Julialang tool for #archaea lab collaboration. Pure terminal UnicodePlots pipe for assessing & plotting raw BAM reads quality then generating length x quality plot animation in mp4. The whole thing takes seconds even for pretty large BAMs - great for bulk analysis! 💻🧬🧫🦠
IMHO we need to resist the natural urge to make associations when viewing these paintings. They're not supposed to resemble anything, and pattern recognition is a barrier to taking them in entirely. Interesting in light of our LLM world and cultural push to associate pattern matching with humanity.
Got some much anticipated quality time with the Rothko rotation at the Washington DC NGA. The rotating collection (from 1000+ works!) is located at E615, best way to get there is an elevator to the rooftop and then walking in via the glass doors.
My good friend Gustave, you're there wherever I go, and life is better for it. La Grotte de la Loue
I couldn't pass up the opportunity to stop by the DC National Gallery of Art! Really looking forward to the Rothkos.
Leaving for DC on Amtrak - #ASMicrobe2026 woot! If anyone around here wants to chat about #archaea, #sequencing, or independent research please feel free to drop a note or come say hi!
Trying out one of those newfangled low/non alcoholic beer the young people love so much. Hey it's not too bad? #beer
I've been deep into ONT basecalling model land as part of my recent archaea study, trying to see if it's worth it in the first place. Here's read to phred map plot per hour of sequencing time- ONT SUP model on the left, and archaea tuned custom model on the right. Interesting! #archaeasky 💻🧬
Almost done rewriting BAM score to length plotting script by @jpelbers.bsky.social! Can't believe this was a month's worth of digging :P New version is based off CairoMakie, thinking about playing with 3D WGL output using topographic map. Also, this was a half decent sequencing run! #julialang
Found a dog-eared copy of Flower Wreath Hill. Lost my own a while ago and had to bring it home. Fun fact. When I got here as an 11 year old I didn't know English. A local college lit student took a shine to me and this is one of the passages we went over night after night. Good memories.
Picked up a used copy of The Penguin Book of the Prose Poem at a used bookstore while coming back from work. So happy it didn't get drenched in the sudden rainstorm. I've been really enjoying Rilke's Dinggedicht pieces recently - curious to see if there's a prose poem match for the practice.
I've been wondering about role of reference assembly contiguity when training/tuning an ONT basecaller model. Stumbled into implementing a crude forward pass in #Julia with tuned weights, contig level genome ref vs complete genome ref. Model behaves very differently on same POD5 signal input! 💻🧬
Korean painting of a cat from the 1800's, artist unknown. I don't know why but older Korean works from the era tend to veer on the humorous side.
Put together a #Julialang monitoring script to keep an eye on gradient norm spikes in ONT basecalling model training. Need to figure out if I can afford another epoch and if learning rate needs to decrease. All done on my phone while on a warehouse shift coffee break! What a world.
Found during a morning shift - maybe I'm just uninitiated but this part absolutely blew my weariness away. For those who don't know I've been digging into extra chromosomal rrns in Archaea for about two years now - preprint coming this summer! #archaeasky 🧫🦠 journals.plos.org/plosbiology/...
Spring night in #nyc Queens. It was nice to feel a bit at ease, if just for a moment.
Done disassembling shelves, moving in a new height adjustable desk and unpacking the new #Opentrons and Bamboo A1 at the lab. Six whole hours of work. This is prep to work with local community college kids, surveying and sequencing local superfund waterways late this summer!
AMD-gigabyte MB combo still wonky on linux after nearly 6 years. I'm pretty sure none of my workstation component's running at -55C lol.
This is probably one of the better shots of the city I took on film. TriX developed in caffenol, from upper floors of our warehouse-lab. #nyc