David Bass

David Bass

@davidb256.bsky.social

he/him | PhD student worker doing computational genetics at JHU | TRU-UE Local 197 member | quizbowler, PACE member | UVA alumnus

212 followers257 following63 posts

My first preprint of grad school is up on bioRxiv. It's been an exciting journey, from getting inspired by a conversation about squid polymerases ~1.5 years ago, presenting at ISMB this summer, and now submitting for peer review!

Steven Salzberg@stevensalzberg.bsky.social

Check out our new preprint on thousands of novel recursive exons in the human genome (if you don't know about recursive exons, they are pretty cool), led by PhD student David Bass @davidb256.bsky.social : www.biorxiv.org/content/10.6...

the Mississippi–Louisiana border was drawn at the Mississippi River, but a Mississippi River long past. Extinct oxbows are visible all along the border on Google Maps.

the map on the Wikipedia page "Autonomous communities of Spain" (Habbit 2007) abstracts out Morocco into just a word, making Melilla and Ceuta appear as islands

the Wikipedia page for the United States Central Command specifies that the United States Space Command "is among four [commands] that are headquartered outside their area of operations"

figure from [Sibley et al. 2015](doi.org/10.1038/natu...) that shows how the brain dominates every other tissue's expression of long genes. each line is essentially a histogram, normalized w.r.t. the brain transcriptome's distribution. cool that neural complexity manifests in such a simple stat

interesting use of popgen in distinguishing SNVs from RNA edit sites: "a pair of SNPs within the mRNA ... possess perfect allelic linkage. In contrast ... RNA editing sites do not generally show significant allelic linkage with each other nor with SNPs" (L-GIREMI, Liu et al. 2023 Genome Biol.)

beating the dead horse here, but I just tried my hand at asking GPT-4.1 (JHU gives us access to a bunch of premium LLMs for free) to help me look for a certain type of data, and it told me about four papers, precisely one of which actually exists. impressively wasteful

two of the three fake DOI URLs take me to an error page, while the third takes me to "Elicitation of broadly protective sarbecovirus immunity by receptor-binding domain nanoparticle vaccines." sounds cool, but very much not what I research

fascinated by this plot for listing permutations (namely, the orders in which four introns can be spliced from a given RNA transcript) and denoting their frequencies with line thickness and shade. It avoids needing to enumerate all 4! permutations. from Choquet et al. 2023 (Nat. Struct. Mol. Biol.)

is there a pre-built/easy way to filter a BAM file to the reads that are aligned spliced over a given intron? The best that I can come up with is to do initial filtering with pysam's fetch function, then examine each read's alignment position and CIGAR string

just learned that you can seamlessly concatenate sequences produced by brace expansion with explicit strings in Bash, e.g. `{1..22} X Y` to list all chromosome names. truly a language of many tricks and wiles

I just finished Marilynne Robinson's Gilead. Incredible book about living on the edge of political violence and celebrating uncertainty. Makes me want to read the New Testament

my Jupyter notebooks are vibrant ecosystems in which data analysis for my research and Balatro odds calculations coexist side-by-side (the expected xMult from five Bloodstone procs is 3.05)

my partner and I just watched Koyaanisqatsi on archive.org and it was reversed, but we didn't notice until after finishing it... I was like "oh cool the music is Einstein on the Beach but backwards" and "wow you don't notice that the space shuttle explodes in reverse until a minute in"