iiinova — where your digital self keeps evolving

Rich Abdill

@richabdill.com

853 followers 1.2K following 102 posts

Bioinformatics, open data and meta-research at Sage Bionetworks

Posts Replies Media

Rich Abdill @richabdill.com · 14/07/2026

New Sequence Read Archive standards, you have until next year to start uploading files with (checks notes) data in them ncbiinsights.ncbi.nlm.nih.gov/2026/06/29/s...

5. Sequence file content requirements

To be accepted, each read file must:

Contain at least one valid read—Empty files or files with no valid reads will be rejected
Contain no empty reads—Files containing empty or incomplete reads will be rejected
Include per-base quality scores—FASTQ and BAM/CRAM must contain standard quality information for all bases and reads
Use valid IUPAC nucleotide codes—Allowed characters include:
A, C, G, T, U, R, Y, S, W, K, M, B, D, H, V, and N. Files may be flagged or rejected if >50% of bases are ambiguous
Reference valid INSDC assemblies—If aligned, the reads must reference a publicly available INSDC assembly. If using a private or unreleased reference, submit as unaligned
File should not be previously deposited—Duplicate submissions are not permitted; reference the existing accession instead

041

Rich Abdill @richabdill.com · 20/01/2026

Google's browser is now cutting off its own ads for eating up too many resources

Screenshot of a banner ad at the top of a website. It's been replaced by a dark box labeled "This ad used too many resources for your device, so chrome removed it."

010

Rich Abdill @richabdill.com · 30/08/2025

Growing thread of alternative neutral network illustrations. Reminds me of how much trouble I had with gradient descent until I found enough cartoons that looked like a tennis ball on a Pringle

An illustration of the iterative search for a local minimum along a gradient. It looks like a path (the output of the cost function for a given vector) down a mountain (defined by whatever this function is that they're minimizing)

010

Rich Abdill @richabdill.com · 26/08/2025

Things are getting pretty dark

Stephen Miller on @seanhabbity: "The Democrat Party is not a political party. It is a domestic extremist organization."

000

Rich Abdill @richabdill.com · 14/08/2025

These @septaphilly.bsky.social‬ cuts look worse every time I see them. What does the Main Line look like after this?

021

Rich Abdill @richabdill.com · 06/08/2025

NY Times headline from 1933 reading, "German fugitives tell of atrocities at hands of Nazis / American Bear Out Tales of Outrages and Cruelties in Racial "Purging." / Jews flee persecution / Pour Across Borders, Fearing for Lives—Police Apathetic and Courts Powerless. All news is censored...

010

Rich Abdill @richabdill.com · 31/01/2025

I think poop is as funny as any microbiome researcher would, but it's still a little shocking to see data plotted in brown

A visualization plotting "16S reads per sample" in three categories: "water" points are colored blue, "isolation control" points are colored green, and "rectal swab" points are colored brown.

2253

Rich Abdill @richabdill.com · 14/01/2025

Not sure if this is new, but I just noticed the Cell homepage lists papers by the lead contact ("Di Paolo and colleagues"), rather than the first author ("Guo et al."). Even when the lead contact isn't in the explicit author list www.cell.com/cell/home

Screenshot of the listing for a paper on the Cell homepage, "Trans-ancestry genome-wide study of depression identifies 697 associations implicating cell types and pharmacotherapies." The authorship is listed as "McIntosh and colleagues."

Screenshot of the header for the research paper "Trans-ancestry genome-wide study of depression identifies 697 associations implicating cell types and pharmacotherapies," which only lists a single author, "Major Depressive Disorder Working Group of the Psychiatric Genomics Consortium." A footnote attached to this author states, "Lead contact (Andrew M. McIntosh)"

010

Rich Abdill @richabdill.com · 12/10/2023

Maybe this gets me put on the "College Terror List" that Dr. Rechavi is promoting for Bill Ackman, but I find it unacceptable that a supposed leader thinks an appropriate response to a terrorist attack includes making sure students in the Harvard Kennedy School Muslim Women's Caucus never get a job.

Tweet from Twitter user @BillAckman saying, "I have been asked by a number of CEOs if @harvard would release a list of the members of each of the Harvard organizations that have issued the letter assigning sole responsibility for Hamas’ heinous acts to Israel, so as to insure that none of us inadvertently hire any of their members. If, in fact, their members support the letter they have released, the names of the signatories should be made public so their views are publicly known. One should not be able to hide behind a corporate shield when issuing statements supporting the actions of terrorists, who, we now learn, have beheaded babies, among other inconceivably despicable acts."

Tweet from twitter user @mualphaxi say says, "Here is the list Ackman wants, and not just Harvard." with a link to a Google Sites page. A screenshot of the page has the headline "College Terror List: A Helpful Guide for Employers" with a photo of Harvard University's Dunster House undergraduate housing.

110

Rich Abdill @richabdill.com · 12/10/2023

I really wanted to avoid this, but @odedrechavi.bsky.social is exhorting us to "remember who tweeted what," so I need to say that as a Jewish scientist, I find this rhetoric very distressing. Any professor spreading calls to make student blacklists to his 138,000 followers is being a menace.

Screen shot of a retweet by Oded Rechavi of twitter user @BillAckman saying "I have been asked by a number of CEOs if @harvard would release a list of the members of each of the Harvard organizations that have issued the letter assigning sole responsibility for Hamas' heinous acts to Israel, so as to insure that none of us inadvertently hire any of their..."

Tweet from @OdedRechavi saying "Remember who tweeted what."

151

Rich Abdill @richabdill.com · 22/09/2023

Google Docs with a clutch new way to distract myself

Checkbox labeled "Display word count while typing"

010

Rich Abdill @richabdill.com · 07/09/2023

Back in the days of print journals, was it common to make giant figures and then put basically the entire legend at the *bottom* of the following page? Unreadable. Probably the best example of moves that suggest PDF publications are a formality that journals don't particularly care about.

Figure legend in a PDF in which the first two (of five) panels are described, followed by the words "(legend continued on next page)"

010

Rich Abdill @richabdill.com · 01/09/2023

Free speech means banning groups I don't like

Tweet from X user KeithWoodsYT saying "Elon Musk likes my call to #BanTheADL The people have spoken, we want free speech!"

000

Rich Abdill @richabdill.com · 28/08/2023

Great system, feels good

Screenshot from Nelnet.com, the website of the company that services all federal student aid, with the error message "Our automated phone system and online accounts are currently unavailable due to technical difficulties, and we're working quickly to resolve the issue. We apologize for the inconvenience. Please check back here for updates."

000

Rich Abdill @richabdill.com · 04/08/2023

Eugene Chang's lab was looking at PYY, an appetite-regulating hormone, and happened to notice it was also popping up in unexpected cells in the gut epithelium. Turns out there's another version targeting pathogenic C. albicans, but not the friendly forms: 🔒 doi.org/10.1126/science.abq3178

From caption: "C. albicans hyphae abundance (Hyphae-GFP) on confluent Caco2 cells exposed to vehicle (Control) or PYY1-36 (20 μm) after 6 hours." Left panel ("Control") shows notably higher abundance.

030