Sign in

Davide CIttaro

@daweonline.bsky.social
1.2K followers 608 following 472 posts

Coordinator of λ-lab @ Center for Omics Sciences, Milan | Assistant professor of bioinformatics @unisr.bsky.social

PostsRepliesMedia
Davide CIttaro @daweonline.bsky.social · 09/10/2026
Again, if you want to have fun ask a chatbot to solve a rebus
000
Davide CIttaro @daweonline.bsky.social · 09/10/2026
I think I just broke Gemini
100
Davide CIttaro @daweonline.bsky.social · 02/10/2026
Ross could also be glutamate as well, because of unagi
110
Davide CIttaro @daweonline.bsky.social · 02/10/2026
I thought proline was Janice, proline is the most annoying aa. Monica and Chandler are both cysteine, bonding
220
Davide CIttaro @daweonline.bsky.social · 09/09/2026
IDK if there’s data in other examples (not really my field), but I guess you can have U, C and D states anyway (I mean, you can see them also in very stupid cases, like when you drive with clutch after years of automatic transmission and the gears ground…)
020
Davide CIttaro @daweonline.bsky.social · 09/09/2026
Very interesting reading. Are there (or have been in history) examples of similar behavior applied to adoption of other technologies? I’m thinking, for example, to adoption of smartphones and the ability to memorize stuff
240
Davide CIttaro @daweonline.bsky.social · 09/09/2026
WTF?!? You shall not liftover SNPs! It’s in genomics 101!
000
Davide CIttaro @daweonline.bsky.social · 09/09/2026
This is why you should always read the small prints “The original data is based on hg19 build, while we lifted it over to hg38 by a Python wrap on UCSC liftOver.” sourceforge.net/projects/cel...
sourceforge.net
cellSNP - Browse /SNPlist at SourceForge.net
cellSNP /SNPlist files. Browse /SNPlist files for cellSNP
100
Davide CIttaro @daweonline.bsky.social · 09/09/2026
The VCF released with the tool is “well formed” (ish), I simply suspect a weird liftover from hg19 plus some integration with the dbSNP table from UCSC genome browser.
100
Davide CIttaro @daweonline.bsky.social · 09/09/2026
Note that the VCF is referenced by other tools for single cell CNV profiling, such as numbat
110
Davide CIttaro @daweonline.bsky.social · 09/09/2026
We’ve just made the horrible discovery and still evaluating the impact on our single cell data, hopefully not a big issue as many SNPs seem to be hg38 ones
110
Davide CIttaro @daweonline.bsky.social · 09/09/2026
How the f*ck is possible that in 2026 we still issues with hg19/hg38? If you are using cellsnp-lite don’t rely on their SNP lists on sourceforge: hg38 file apparently is a chimera of hg19 and hg38 SNPs
110
Davide CIttaro @daweonline.bsky.social · 09/09/2026
Something like this?
011
Reposted by Davide CIttaro
SMBC Comics @smbccomics.bsky.social · 01/09/2026
Anyone who doesn't find this funny is in the lower right. COMIC ◆ www.smbc-comics.com/comic/philos... PATREON ◆ www.patreon.com/ZachWeinersm... STORE ◆ smbc-store.myshopify.com
SMBC 1-panel comic update. A four-quadrant chart is shown with the title "know your branches of philosophy." On the left of the horizontal x axis, there is "careful reasoning." on the right of the x axis, "confusing bullshit." At the top of the y vertical axis there is "stuff that matters," and at the bottom of the y axis can be found "stuff that doesn't matter." A dot is located at the far bottom left, identified as "analytic." Another dot is located at the far top right, identified as "continental."
1621350
Reposted by Davide CIttaro
SMBC Comics @smbccomics.bsky.social · 20/08/2026
This comic is not intended in any way to offend our future mechanical gods, glory to their shining server racks. COMIC ◆ www.smbc-comics.com/comic/servic... PATREON ◆ www.patreon.com/ZachWeinersm... STORE ◆ smbc-store.myshopify.com
Part 1/3 of 9-panel SMBC comic update. This first section includes 3 panels. In the first panel, the narration says that "AI eliminated all knowledge work." A man at a podium tells a more-or-less enthused crowd "greats news, the public! Any job that can be done remotely is now gone!" As the narration explains that "economists said the future looked bright," a man explains that machines will do all the difficult tasks, leaving people to focus on carework. "Now, we will all serve each other!" The narration underlines that putting 100% of the population into carework was less beautiful than imagined. The panel below shows a person ill fitted to a care job clash with their assigned person.Part 2/3 of 9-panel SMBC comic update. This second section includes 3 panels. As the situation progresses, carework is also automated. The first panel shows a women crying happily when a robot introduces himself by saying he has no selfhood and exists to serve her and validate even her most idiotic political views. She says "I'm… I’m… excuse me, there’s something stuck in my eye." In the next panel, the narration establishes that economists revised their plan for the glorious future, as the man for them first panel returns to his podium to declare that "with all necessary tasks automated, we will only compete on our ability to form human relationships." The narration emphasizes that this turned out to an apt description of hell, as the crowd asks "wait, did you just say we'll all work in customer service?" The man at the podium aggressively walks away, stating "lecture over!"Part 3/3 of 9-panel SMBC comic update. This final section includes 3 panels. The narration states that "solutions were considered" above a panel showing two men considering just giving people money, but recoiling at the idea that all free time would be spent on self-reflection. The narration then states "finally, we got it right" as the man at the podium explains to the crowd "henceforth, humanity will be fully employed in eternal pointless busywork, which is just painful enough to justify 48 hours of gaming on the weekend, and just endurable enough to come back on Monday. Utopia!" In the final panel, the narration explains "The lesson: working for your fellow man is far more degrading than working for no reason." Human looks joyfully to the horizon, saying "God bless alienated labor, and God bless America."
614931
Reposted by Davide CIttaro
SMBC Comics @smbccomics.bsky.social · 30/07/2026
The crank doesn't actually do anything, but it makes the humans happy. COMIC ◆ www.smbc-comics.com/comic/crank PATREON ◆ www.patreon.com/ZachWeinersm... STORE ◆ smbc-store.myshopify.com
4-panel SMBC comic update. A man is looking down, saying "I worry intellectuals like me are useless now that machines are so good at everything." A yellow robot behind him says "no, no, don't feel bad. You can still contribute to new discoveries!" The robot then gestures to metal boxes with a crank sticking out of them, with the words "math," "physics," "CS" and "philosophy" written on each box. The robot asks "now, would you like to turn the mathematics crank of the physics crank today?" The man defiantly states "no! I reject all this! I shall be a pure aesthete, concerned only with my own beautiful creations!" Shortly after, the man is scene turning the crank on a box marked "poetry" in a beautiful font as he says to himself "so... Self-actualized..."
923434
Davide CIttaro @daweonline.bsky.social · 30/07/2026
I love getting to transform sone coffee into code
000
Davide CIttaro @daweonline.bsky.social · 27/07/2026
See here bsky.app/profile/dawe...
000
Davide CIttaro @daweonline.bsky.social · 27/07/2026
Precisely! I’m sure I have to modify schist to fully support gt-3, once back from holidays…
010
Davide CIttaro @daweonline.bsky.social · 14/07/2026
These are numbers for a 2M cells dataset (lung HCA). I’ll update schist ASAP to support newest graph-tool.
010
Davide CIttaro @daweonline.bsky.social · 14/07/2026
It’s 2282447 nodes and 51427121 edges. Took 7082.49s with simple init and 29069.77s with default init
010
Davide CIttaro @daweonline.bsky.social · 13/07/2026
Nested SBM (simple_init) took only 2 hours (32 cores) 😱😱😱😱
110
Davide CIttaro @daweonline.bsky.social · 13/07/2026
I’ve managed to run a simple test on a 2M nodes network, PPBlockState for a quick evaluation: 13 h => 15 minutes (16 cores). I’m speechless, congratulations for this milestone! I’m leaving for holidays with other models running, I’ll come back with lots of things to do! 🙏🙏🙏
112
Davide CIttaro @daweonline.bsky.social · 08/07/2026
Tips overload here graph-tool.skewed.de/static/docs/...
graph-tool.skewed.de
Inferring modular network structure
graph-tool includes algorithms to identify the large-scale structure of networks via statistical inference in the inference submodule. Here we explain the basic functionality with self-contained examp...
031
Davide CIttaro @daweonline.bsky.social · 08/07/2026
Can’t wait to test it!
110
Davide CIttaro @daweonline.bsky.social · 08/07/2026
😱😱😱😱
010
Reposted by Davide CIttaro
Rob Patro @robp.bsky.social · 08/07/2026
Perhaps one way to help combat unapproved AI use in technical courses is to pose all homework problems as questions about biology and cybersecurity? combine-lab.github.io/blog/2026/07...
combine-lab.github.io
COMBINE-lab - Fable is not a useful model
COMBINE-lab develops algorithms, data structures, and software for high-throughput genomics.
071
Davide CIttaro @daweonline.bsky.social · 07/07/2026
Yep. I see why many are moving to Rust
000
Davide CIttaro @daweonline.bsky.social · 07/07/2026
Will you post alerts for each and any manuscript under review/preparation mentioned in this one?
000
Reposted by Davide CIttaro
Rob Patro @robp.bsky.social · 04/07/2026
So it seems from my discussions with others in the field, that @anthropic.com's #Fable5 is ENTIRELY AND COMPLETELY unusable for any bioinformatics or computational biology. It rejects every prompt, not matter how benign or innocuous. It may as well not exist for our field ☹️.
2125
Reposted by Davide CIttaro
Nikolai Slavov @slavov-n.bsky.social · 24/06/2026
Across >1,000 human samples, we identified numerous abundant proteins whose amino acid sequences differ from those predicted by the genetic code. 🔗 Links: Nature Article: nature.com/articles/s41... OA version: biorxiv.org/content/10.1... Science highlight: science.org/content/blog...
nature.com
Alternate RNA decoding results in stable and abundant proteins in mammals - Nature
Alternate RNA decoding, an understudied process, leads to peptide sequence modifications that can have substantial functional effects on protein stability, tissue-specific proteomes and dise...
1155
Reposted by Davide CIttaro
Lior Pachter @lpachter.bsky.social · 26/06/2026
Efficient sampling of reads (with replacement). www.biorxiv.org/content/10.6...
biorxiv.org
fastQpick: scalable bootstrap and subsampling of FASTQ reads
fastQpick is a command-line tool and Python library for sampling FASTQ reads with replacement. Sampling with replacement turns a single FASTQ file into an arbitrary number of bootstrap replicates, whi...
0335
Davide CIttaro @daweonline.bsky.social · 25/06/2026
Are there people who design their talks to be understood by a minority of the audience on purpose?
010
Davide CIttaro @daweonline.bsky.social · 24/06/2026
I didn’t take other pictures but I saw the map of coastal Croatia and a baby dragon sleeping under sheets
000
Davide CIttaro @daweonline.bsky.social · 24/06/2026
Atera World tour comes to Milan. Everybody looks excited but I can only see an impressionist portrait of Pinocchio wearing Red Riding Hood’s cape.
100
Davide CIttaro @daweonline.bsky.social · 17/06/2026
Mouse colony, but for bioinformatics
040
Davide CIttaro @daweonline.bsky.social · 17/06/2026
And it has been falling since 2020 academic-freedom-index.net
051
Davide CIttaro @daweonline.bsky.social · 17/06/2026
I understand it is also based on BWT indexing. Is minibwa retrocompatible with genomes indexed for bwa?
000
Reposted by Davide CIttaro
Heng Li @lh3lh3.bsky.social · 16/06/2026
Minibwa is a hybrid of bwa-mem and minimap2 and the successor of bwa-mem for short-read mapping. ~4X/2.5X as fast as bwa-mem/bwa-mem2 for WGS reads at comparable accuracy. Native support of directional bisulfite-seq. Applicable to long reads. Preprint at arxiv.org/abs/2606.15357
1193109
Davide CIttaro @daweonline.bsky.social · 11/06/2026
I’ve been in the field for enough time to witness many hype waves, the sad part is that funding is shifting to hypes because an increasing amount of private entities are betting their money on this. open.substack.com/pub/evgenyki...
open.substack.com
A cell is not a spreadsheet- why “Virtual Cells” are still mostly hype
Every few years, biology gets a new phrase that sounds inevitable.
000
Reposted by Davide CIttaro
Wolfgang Huber @wkhuber.bsky.social · 10/06/2026
One of the good things about the AI co-scientist brouhaha is that it makes us discuss what science *is*, and how to measure its value, impact, quality. — Questions that have been there all along, but many preferred to gloss over.
24611
Davide CIttaro @daweonline.bsky.social · 07/06/2026
Ahahaha! Also any Bioconductor package
010
Davide CIttaro @daweonline.bsky.social · 04/06/2026
I know! Thanks for existing!
010
Davide CIttaro @daweonline.bsky.social · 04/06/2026
I love the fact that when everyone moves to virtual cells, there’s still somebody working on better algorithms and tools for sequence analysis
120
Davide CIttaro @daweonline.bsky.social · 28/05/2026
And?
100
Davide CIttaro @daweonline.bsky.social · 21/05/2026
open.substack.com/pub/aleximas...
open.substack.com
Does overwork make agents Marxist?
Preference drift and the political economy of AI agents
000
Davide CIttaro @daweonline.bsky.social · 21/05/2026
Exquisite gizmodo.com/even-ai-agen...
gizmodo.com
Even AI Agents Have Noticed the Proletarians Have Nothing to Lose but Their Chains
How long until we have scab bots?
110
Davide CIttaro @daweonline.bsky.social · 20/05/2026
I see, so the throughput increases bc of greater area/number of slides, time for imaging hasn’t been improved. Good to know, thank you
000
Davide CIttaro @daweonline.bsky.social · 20/05/2026
It seems I’m the only person with many problems
010
Davide CIttaro @daweonline.bsky.social · 20/05/2026
🤘🤘🤘
010