David Bass
LightNews LightNews
banner
davidb256.bsky.social
David Bass
@davidb256.bsky.social
170 followers 230 following 44 posts
he/him | PhD student doing computational genomics at JHU Dept of Bio and CCB | TRU-UE Local 197 member | quizbowler, PACE member | UVA alumnus
Posts Replies Media Videos
davidb256.bsky.social
biblically accurate sequence logo (Hsiao et al. 2018, Genome Res., fig. 2G)
November 5, 2025 at 3:17 PM
davidb256.bsky.social
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.)
A figure panel showing a sketch of how, due to genetic linkage, SNPs are distinguished from RNA editing sites by sharing more mutual information.
October 10, 2025 at 2:59 PM
davidb256.bsky.social
tfw you get to "A Squeeze of the Hand"
September 29, 2025 at 6:47 PM
davidb256.bsky.social
accidentally applying `.set_xbound(0, 1)` instead of `.set_ybound(0, 1)` turns violin plots into KDE plots that cling to the walls
August 20, 2025 at 7:39 PM
davidb256.bsky.social
just read The Unbearable Lightness of Being, and the quote "kitsch is the absolute denial of shit" reminded me of this post about genAI refusing to draw accurate baby birds because they're not cute
A tumblr textpost, which I of course screenshotted from Reddit, that shows two images of "baby pigeons" created by generative "AI." The images are cute, depicting the babies as downsized adults instead of the ugly scruffy things that most baby birds actually are. A tumblr user writes "the ai's aversion to producing anything that's not aesthetically pleasing dentist office artwork leaves it unable to comprehend true beauty" and attaches a real image of a baby bird, which is ugly and lacks feathers.
August 12, 2025 at 2:04 PM
davidb256.bsky.social
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.)
Panel of figure 2 of Choquet et al. 2023 "Pre-mRNA splicing order is predetermined and maintains splicing fidelity across multi-intronic transcripts" (Nat. Struct. Mol. Biol.) that uses zig-zagging lines of different thicknesses and (gray) colors to show frequencies of permutations of splicing order of four introns in a representative transcript.
August 5, 2025 at 5:21 PM
davidb256.bsky.social
today is a good day because I saw the Sucellus statue at the Walters
Picture from the Walters Museum of Art of their ~foot-tall bronze statue of the Celtic/Gallo-Roman deity Sucellus, standing in contraposto and holding up a mallet that has been lost to time. Floating behind him is a large mallet with five smaller mallets protruding from its head.
June 23, 2025 at 12:37 AM
davidb256.bsky.social
the MD pride flowing through my veins today has got me like
An image of Mithrun, an anime character from Dungeon Meshi, photoshopped to be decked out in Maryland gear and hanging out in Baltimore's Innher Harbor. She is non-chalant, but conveys great pride in her state.
April 16, 2025 at 9:50 PM
davidb256.bsky.social
had a great time competing in @naqt.com's Intercollegiate Championship Tournament this weekend with my JHU teammates!
Group photo of the Johns Hopkins University quizbowl team, shortly before round 1 of NAQT's 2025 Intercollegiate Championship Tournament. We are (mostly) wearing matching navy team sweaters. We sit in a large hotel convention space.
April 8, 2025 at 2:55 PM
davidb256.bsky.social
how it feels to zoom out in order to see previous lines of output in a Linux screen after having your (foolish) attempt to scroll rewarded with a bunch of "^]]A"s
A promotional image of MacGyver, looking cool and carrying a missile, for the original '80s TV series.
March 26, 2025 at 3:28 AM
davidb256.bsky.social
it's Sextidi the 26th of Pluviôse in the year 233, the day of the topologically associating domain
Figure 1 from Matharu and Ahituv 2015 "Minor Loops in Major Folds" shows (A) a cartoon of chromatin in the nucleus partitioned into differently colored topologically associating domains (TADs), (B) a cartoon of an individual TAD's structure and interactions with the nuclear membrane, and (C) a genetic track displaying regulatory elements, genes, and TAD boundaries within a single (active) TAD compartment, highlighted with a green triangle evocative of HiC heatmap output.
February 14, 2025 at 3:18 PM
davidb256.bsky.social
French Republican calendar but instead of rural day names it's genomics day names and today is recursive splicing
Excerpt from Fig. 2a of Sibley et al. 2016's "Lessons from non-canonical splicing" that illustrates the signature "sawtooth" pattern of RNA-seq read coverage in recursively spliced introns.
February 13, 2025 at 3:41 PM
davidb256.bsky.social
i love these little guys
A complicated, multi-colored, rectangular visualization of eras of geologic time from the Wikipedia page for the Precambrian. The vertical axis shows time in millions of years ago, and the horizontal axis shows increasing precision of terms in the hierarchy of geologic eras, epochs, eons, whatever, etc.
January 8, 2025 at 3:18 PM
davidb256.bsky.social
this is a rhodopsin structure fan account
3D structure of the Drosophila melanogaster protein RH2, a G-protein coupled receptor homologous to human rhodopsin. As a GPCR, it features seven alpha-helical transmembrane domains; one of them is highlighted.
November 11, 2024 at 1:38 AM
davidb256.bsky.social
Interesting shape of a quadratic regression fit to data that is, uh, not particularly quadratic and also log-transformed
A dense blue scatterplot of putative introns' lengths versus the count of aligned reads that were spliced over them. A red quadratic regression curve fits poorly, interestingly plunging down below the (positive-valued) y-axis between ~7500 and ~20000 bp in length due to the curve being fit to non-log-transformed data, but then displayed with a log-transformed y-axis.
September 17, 2024 at 12:52 PM
davidb256.bsky.social
Happy about these Google Slides illustrations that I made for a rotation project presentation on inference of rate constants from (simulated) experimental data describing kinetic trapping
A graph illustrates the formation of a heterotrimer. The reaction network is parameterized by two stage-specific, rate constants, k_dim and k_tri. A presentation slide is titled "Kinetic trapping inhibits molecular self-assembly. Three sketches of a cell attempting to complete a sufficiently long microtubule are labelled, respectively, "Cell seeks to span self with microtubule," "Tubulin subunits bind to existing microtubule: No kinetic trapping," and "Tubulin subunits form new microtubule: Kinetic trapping."
January 29, 2024 at 3:53 AM