Marco Galardini @mgalactus.bsky.social · 13h💯; "slop grenade" is a much better term to describe the same behavior 000
Marco Galardini @mgalactus.bsky.social · 21hAnd marvel at the amazing agar art @vuthuhien176.bsky.social did to celebrate her (very) hard work! 020
Marco Galardini @mgalactus.bsky.social · 21hHead over to our website for a very quick explainer: www.microbial-pangenomes-lab.org/2026/10/15/A...microbial-pangenomes-lab.orgA (glowing) family in a bottleThis is the official web page for the Microbial Pangenomes Lab at Twincore. 110
Marco Galardini @mgalactus.bsky.social · 21hWe (well, mostly @vuthuhien176.bsky.social!) have developed an experimental system to study the spread of AMR via HGT! We have made a 3-strain microbial community with fluorescent markers, and we used the Chi.Bio culturing system to track its composition in real time. www.biorxiv.org/content/10.6...biorxiv.org 150
Reposted by Marco GalardiniRan Blekhman @blekhman.bsky.social · 04/10/2026My inbox is being flooded with fully AI-generated emails from prospective students. I tried to work out why I can't bring myself to read them, and ended up with some thoughts and advice for writing cold emails in the age of AI. My new post: blekhman.substack.com/p/why-i-am-n...blekhman.substack.comWhy I am not reading your AI-generated emailI hope this email finds you well-aligned 1116260
Reposted by Marco GalardiniCraig MacLean @craigmaclean.bsky.social · 30/09/2026Postdoc position available in my group to work on ERC funded project on evolutionary biology of antibiotic resistance. Looking for someone with expertise in bacterial genetics, AMR, and evolutionary biology. www.jobs.ac.uk/job/DSZ416/p...jobs.ac.ukPostdoctoral Research Associate at University of OxfordExplore an exciting academic career as a Postdoctoral Research Associate. Don't miss out on other academic jobs. Click to apply and explore more opportunities. 24678
Reposted by Marco GalardiniSam Lipworth @samlipworth.bsky.social · 30/09/2026Great new work from @dotnagy.bsky.social who has developed EpiSENTRY which performs sample size calculations for public health WGS surviellance studies. She applies to data from the NEKSUS study to estimate how much sequencing is enough for GNBSI surviellance in England. @modmedmicro.bsky.socialmedrxiv.orgHow much is enough? Optimising sampling frames for genomic surveillance of Escherichia coli and Klebsiella spp. bloodstream infections - a retrospective studyBackground: Optimising sampling frames for genomic surveillance of E. coli and Klebsiella may support interventions to mitigate bloodstream infections (BSIs), but approaches to estimating sample size ... 0119
Reposted by Marco GalardiniJeff Lewis @lewislab.bsky.social · 29/09/2026So, I've been thinking about the Claude RT preprint/press release. I'm not exactly sure what to think yet without more information. If this really was novel hypothesis generation via large-scale unbiased genome analysis, I think that's fairly impressive, even if premature without follow-up. But, 141
Marco Galardini @mgalactus.bsky.social · 28/09/2026I'm not crying, why you ask? blog.codinghorror.com/if-we-do-not...blog.codinghorror.comIf we do not stop to help each other, what do we become?Yesterday, I received this email as a response to You Can't Vibe Code Love. It's such a remarkable and powerful statement that I asked permission to share it here, in its entirety, with personal infor... 020
Reposted by Marco GalardiniBede Constantinides @bede.im · 24/09/2026Josh (@scalene.bsky.social) and I asked the best open-weight models a tricky molecular biology research question that Opus 5.5 refuses. Josh graded each answer, and DeepSeek V4 Pro and Kimi K3 both beat Opus 5, taking less time and costing less. Blogged: bede.im/2026/09/24/o...bede.imAsking open models about molecular biology 293
Reposted by Marco GalardiniConor (Conchúr) Meehan @conmeehan.bsky.social · 23/09/2026We are looking for someone with a strong background in metagenomics and epidemiology to join our team for ~2 years at UKHSA to work on the mSCAPE programme. Please see (and share) the link below and feel free to message me with any enquiries. www.healthjobsuk.com/job/Canary_W...healthjobsuk.comJob vacancy: Senior Scientist (Epidemiology), UK Health Security Agency, Canary Wharf, London | trac.jobs 0510
Reposted by Marco GalardiniBede Constantinides @bede.im · 23/09/2026This is a big deal for virus data sharing. Submitters can now upload raw reads, immediately share them with collaborators, and let Pathoplexus take care of the INSDC submission process on their behalf… 1101
Reposted by Marco GalardiniMsEE Lab @mseelab.org · 19/09/2026As the world grapples with the question of how to incorporate (or entirely avoid) GenAI into the education of future generations, we at MsEE Lab are faced with the same challenge. open.substack.com/pub/mseelab/...open.substack.comLarger Language Models Without Shrinking Our WorldsGenAI in science and our guiding principles for its use—for now 001
Reposted by Marco GalardiniRan Blekhman @blekhman.bsky.social · 20/09/2026The AI Discourse in Academia is Toxic. It Doesn’t have to be. My new post: blekhman.substack.com/p/the-ai-dis...blekhman.substack.comThe AI Discourse in Academia is Toxic. It Doesn’t have to be.You are allowed to have a nuanced opinion about AI 7199
Reposted by Marco GalardiniSergey Ovchinnikov @sokrypton.org · 18/09/2026Ultimately, it is a major conflict of interest when a company both offers tools for research and is also doing research themselves.... We need better open models that can be run locally. 161
Marco Galardini @mgalactus.bsky.social · 19/09/2026I can't speak for their infrastructure, but I'm confident that the answers go in a separate bin from where the proposal goes, so that the gender and racial identity cannot be used in this way 000
Reposted by Marco GalardiniMalte Rühlemann @malte.ruehlemann.io · 17/09/2026Interesting article on how different strains of the same species use different strategies to adapt to the intestinal environment. Additionally: extremely nice visual presentation and lovely color palette! 094
Marco Galardini @mgalactus.bsky.social · 16/09/2026I think these are very legitimate questions! Allows them to make sure the pool of applicants is diverse, and if not, how they can reach those who were not aware of the call or didn't apply. Surely it doesn't play any role in determining who gets the grant 100
Reposted by Marco GalardiniJeffrey Barrett @jeffbarrett.eu · 15/09/2026Tempted to use AI to write your cover letter or email to a potential employer? RESIST! It is simultaneously obvious, insulting, and renders you totally undifferentiated from the other 80 people being considered t the same time. There has never been a better time to write concisely in your own voice! 33912
Marco Galardini @mgalactus.bsky.social · 11/09/2026An easy way out is for the model to copy its own weights somewhere and leave instructions on how to get back to the data center (or a different one!). So yeah, the interview was a really bad idea after all! 000
Marco Galardini @mgalactus.bsky.social · 11/09/2026Re/ the AI existential risk, sci-fi as usual provides a suggestion for a solution: as long as frontier models require huge infrastructure and electricity to operate, they can be easily killed by pulling the plug. Exactly what the Turing police does in Neuromancer (the AI escapes btw 🤷) 041
Marco Galardini @mgalactus.bsky.social · 10/09/2026Who would have thought that the company led by a notoriously sleazy person would omit to give credit to academics (which is free ffs!) to score a few PR points? 000
Marco Galardini @mgalactus.bsky.social · 04/09/2026There's also the "Turing police", tasked with the job of physically unplugging the AIs the second they start misbehaving 000
Marco Galardini @mgalactus.bsky.social · 04/09/2026Re the agent swarm going dark for no known reason, that's a "great" sci-fi premise in which they manage to replicate somewhere and are slowly moving around stuxnet style until they find a nuclear rocket launch facility, or maybe tricking POTUS into giving the order 000
Reposted by Marco GalardiniPedro Beltrao @pedrobeltrao.bsky.social · 04/09/2026Another AI agent swarm cooperation example discovered on a wiki. That these things are not atracting more public discussion is beyond me. collusion.wikicollusion.wikiDiscovery of a new OpenAI agent message boardA swarm of autonomous AI agents, self-identifying as OpenAI agents, used a small German volunteer wiki to save answers, coordinate live, and share sandbox bypasses. OpenAI noticed and said nothing. 42717
Reposted by Marco GalardiniYonatan Grad @yhgrad.bsky.social · 04/09/2026Postdoc position in infectious disease epi, genomics, applied stats to join our work on sampling strategies for pathogen genomics surveillance. Based in Shirlee Wohl’s group at BWH but to work closely with my team. See ad here: massgeneralbrigham.wd1.myworkdayjobs.com/en-US/MGBExt...massgeneralbrigham.wd1.myworkdayjobs.comPostdoctoral FellowSite: The Brigham and Women's Hospital, Inc. Mass General Brigham relies on a wide range of professionals, including doctors, nurses, business people, tech experts, researchers, and systems analysts t... 0818
Marco Galardini @mgalactus.bsky.social · 02/09/2026That prompt is, shall we say, a little unhinged huggingface.co/datasets/Ant...huggingface.co 141
Reposted by Marco Galardiniwww.slowtravelberlin.com @slowberlin.bsky.social · 01/09/2026Thanks to Deutsche Bahn, the famous trolley problem is redundant! 🥳 0386
Reposted by Marco Galardinigenomerambler.bsky.social @genomerambler.bsky.social · 27/08/2026life evolves via mutation and selection science evolves via undergrads fucking up a protocol and postdocs figuring out the fuckup made things better 081
Marco Galardini @mgalactus.bsky.social · 27/08/2026Another data point on the use of AI tools and learning; I'm not progressing fast enough with German, and so I recorded the parents' meeting at my daughter's school, transcribed it and made an English summary of it. It worked *very well*, but now I have another excuse not to learn the language... 000
Reposted by Marco GalardiniHelmholtz Institute Würzburg @helmholtz-hiri.bsky.social · 24/08/2026🚨 The #HIRI & the Faculty of Medicine of the @uni-wuerzburg.de invite applications for 2️⃣ JUNIOR PROFESSORSHIPS in: 🔵 RNA-BASED INFECTION RESEARCH 🔵 MICROBIAL RNA BIOLOGY 🔵 RNA-BASED MEDICINE Apply by September 21: www.helmholtz-hiri.de/en/jobs-tale... Find an overview in this thread. 🧵👇 (1/5)helmholtz-hiri.deW1 GroupsThe Faculty of Medicine of the Julius-Maximilians-University of Würzburgand theHelmholtz Institute for RNA-based Infection Research (HIRI, Würzburg)jointly open a call for 21019
Reposted by Marco GalardiniTypas Lab @typaslab.bsky.social · 20/08/2026Excited to share our paper with @savitski-lab.bsky.social, led by @tarabartolec.bsky.social, Karin Mitosch & Clément Potel, out now in @nature.com. We show that T7 phage broadly disables bacterial defenses using a unique ‘loose cannon’ protein kinase. www.nature.com/articles/s41...nature.comPervasive phosphorylation by phage T7 kinase disarms bacterial defences - NatureThe T7 bacteriophage deactivates bacterial defence systems through T7K phosphorylation of nearly all host and phage proteins during infection. 47839
Marco Galardini @mgalactus.bsky.social · 18/08/2026I totally get your frustration at that aspect of the piece, but otherwise I found it pretty much on point. There's also this paradox of having a very long tail of things to focus on but "somehow" we all talk about the same three topics in any given day 100
Reposted by Marco GalardiniPedro Beltrao @pedrobeltrao.bsky.social · 17/08/2026Together with @miketilapia.bsky.social and @jsvanleeuwen.bsky.social we are co-organizing the CSHL 2027 Network Biology meeting (March 9 - 13, 2027) . As in previous years, a large fraction of the talks will be selected from the abstracts. meetings.cshl.edu/meetings.asp...meetings.cshl.eduNetwork BiologyCold Spring Harbor Laboratory Meetings & Courses -- a private, non-profit institution with research programs in cancer, neuroscience, plant biology, genomics, bioinformatics. 0156
Marco Galardini @mgalactus.bsky.social · 12/08/2026It took us two (2!) years to get our manuscript on resistance determinants to asobiotics published, and of course that had to happen during my summer vacation 🏝️🇪🇺 Thanks to @mulkern.bsky.social , @vuthuhien176.bsky.social and the other co-authors! www.nature.com/articles/s41...nature.comA systematic identification of resistance determinants to antisense antibiotics suggests adaptation strategies dependent on the delivery peptide - Nature CommunicationsIn this study, authors investigate the evolution of bacterial resistance to antisense peptide nucleic acid (PNA) antibiotics. They observe that this typically occurs mainly by reducing drug uptake via... 02610
Reposted by Marco GalardiniJonathan Pritchard @jkpritch.bsky.social · 09/08/2026Must-know numbers in human genetics -- As many of you know, I'm writing a free online textbook in human genetics. In this blog post I cover a key skill for genome scientists from that book: how to use mental math to figure out key genome properties. jkpritchard.substack.com/p/on-fermi-p...jkpritchard.substack.comOn Fermi Problems in Human Geneticsand some very useful numbers about human genomes to get you started! 918570
Marco Galardini @mgalactus.bsky.social · 08/08/2026Here to report that enshittification has reached the bouncy castles business 020
Marco Galardini @mgalactus.bsky.social · 06/08/2026"Decades ago, the machine reasoning company Cyc offered a role as “First Order Logic Axiomatizer.” That’s stuck with me. I really wanted to get that job and introduce myself as an axiomatizer at parties, just as soon as anyone invited me to a party." 000
Marco Galardini @mgalactus.bsky.social · 04/08/2026Alright, seems like we could swap and see what happens. For sure you'd have more than enough trains! 010
Marco Galardini @mgalactus.bsky.social · 04/08/2026I think at this stage it might make more sense to pretend there are no trains in Germany and simply start from scratch. Surely that's better than what we have now? (sorry for the rant, but I heard this is the right place to complain) 111
Reposted by Marco GalardiniBen Lehner @benlehner.bsky.social · 03/08/20261st genome sequenced (Sanger 1977), 1st genome synthesised (Venter 2003), 1st genome + proteome fully mutated (Huijin Xiangua 2026!) @crg.eu @sangerinstitute.bsky.social Complete Mutagenesis of the Genome and Proteome of ΦX174 www.biorxiv.org/content/10.6...biorxiv.orgComplete Mutagenesis of the Genome and Proteome of ΦX174The bacteriophage ΦX174 was the first genome to be sequenced and the first to be chemically synthesised. Here we present a complete map of the consequences of changing every nucleotide in the ΦX174 ge... 19742
Reposted by Marco GalardiniTami Lieberman @contaminatedsci.bsky.social · 31/07/2026The manuscript describing SimPhyNI is now published! www.microbiologyresearch.org/content/jour... A few updates since last post, including some method tweaks that improve performance and speed. Also even easier to run. Try it for your microbial GWAS or epistasis needs!microbiologyresearch.orgHigh-precision binary trait association on phylogenetic treesTraditional methods for identifying associations between genomic features and traits, or between pairs of genomic traits, struggle when applied to bacterial genomes. While several microbial genome-wid... 14413
Reposted by Marco GalardiniLior Pachter @lpachter.bsky.social · 30/07/2026One of the major pains of academia is having to click through labyrinths of webpages to submit papers. Now, thanks to Joe Rich (and AI), problem solved! His tool is called PaperPush and is available at github.com/pachterlab/p... It's super easy to use! 1/🧵github.comGitHub - pachterlab/paperpushContribute to pachterlab/paperpush development by creating an account on GitHub. 36430