Sign in

Rafal Mostowy

@rafalmostowy.bsky.social
608 followers 136 following 75 posts

Group leader at Jagiellonian University in Krakow, cofounder of Polonium Foundation. Working on evolution of sex in bugs.

PostsRepliesMedia
Reposted by Rafal Mostowy
Vivek Mutalik @vivekmutalik.bsky.social · 29/09/2026
Happy to share that @averynoonan.bsky.social paper “Phylogeny-agnostic strain-level prediction of phage–host interactions from genomes using machine learning” is now published in @natmicrobiol.nature.com 🎉🎊 👉🏽 rdcu.be/nR2NizD7H5va
Machine learning workflow for GenoPHI platform
27028
Reposted by Rafal Mostowy
Leah Smith @leahmsmith.bsky.social · 28/09/2026
Adapting the power of transposon sequencing to phages! Our final paper is out now👇 Our plasmids 🧬 are available on @addgene.bsky.social – try phage Tn-seq on your favorite phage!
0248
Reposted by Rafal Mostowy
Wytamma @wytamma.bsky.social · 28/09/2026
One of the reasons I'm interested in local AI models is you're data can't be used for training - people say “I already put my data in onedrive so what's the difference to uploading it to an AI model?” Well one difference is that onedrive won't use your data to scoop you
044
Reposted by Rafal Mostowy
Dorentina Humolli @humollidorentina.bsky.social · 25/09/2026
Very happy to share that HIDEN-SEQ is now out in @natmicrobiol.nature.com! Using this technique, we recapitulated decades of established phage genetics and uncovered new gene functions, including previously unknown anti-defense factors against several defense systems. www.nature.com/articles/s41...
38443
Rafal Mostowy @rafalmostowy.bsky.social · 04/09/2026
🧬 Postdoc ad is now live: evolutionary genomics of Klebsiella pneumoniae phages. How do prophage depolymerases vs SGNH-esterases shape phage host range? Krakow, Poland. Deadline 21 Sept: mostowylab.com/job-postdoc-...
22629
Rafal Mostowy @rafalmostowy.bsky.social · 04/09/2026
Very well deserved, congrats @rafomics.bsky.social!!!
110
Rafal Mostowy @rafalmostowy.bsky.social · 15/07/2026
PhD opening in my lab @jagiellonskiuni.bsky.social: using AI to predict how phage enzymes break down bacterial capsules — joint work with ML group GMUM and Zuzanna Drulis-Kawa's phage lab (Wrocław), spanning genomics, AI & evolution. mostowylab.com/job-phd-ai-d... Deadline: 15 August 2026
11016
Rafal Mostowy @rafalmostowy.bsky.social · 10/07/2026
Klebsiella phages often modify the bacterial capsule instead of degrading it, stripping acetyl groups. Nobody's traced how this evolves. I'm hiring a computational postdoc to find out — Kraków, apply by 21 July. mostowylab.com/job-postdoc-...
21510
Reposted by Rafal Mostowy
Xavier Charpentier @labxc.bsky.social · 07/07/2026
#microsky New preprint from the lab! Widespread immune systems protect bacteria against conjugative plasmids www.biorxiv.org/content/10.6... Kudos to the extremely talented lead author Ludovic Poiré! Key contributions by @francoisrousset.bsky.social and also help from C. Lesterlin/TacC lab. 🧵⬇️
biorxiv.org
Widespread immune systems protect bacteria against conjugative plasmids
Conjugative plasmids are a class of mobile genetic elements capable of efficient transfer between bacterial cells. Although they can introduce beneficial traits such as antibiotic resistance to recipi...
410355
Reposted by Rafal Mostowy
Conlon Lab @bconlon.bsky.social · 02/07/2026
Loss of O-antigen due to wbbL mutations is common and associated with increased mortality in Escherichia coli bloodstream infections | Science Advances www.science.org/doi/10.1126/...
science.org
Loss of O-antigen due to wbbL mutations is common and associated with increased mortality in Escherichia coli bloodstream infections
O-antigen loss in E. coli is not benign—wbbL mutations are associated with increased septic shock and higher mortality.
193
Reposted by Rafal Mostowy
Vivek Mutalik @vivekmutalik.bsky.social · 11/06/2026
Phage therapy via receptor-constrained evolutionary traps: Trends in Microbiology www.cell.com/trends/micro... #phage #phagesky
cell.com
Phage therapy via receptor-constrained evolutionary traps
Phage therapy often fails when bacteria evolve resistance. We argue that phage selection should begin with receptors whose modification imposes predictable costs, turning resistance into reduced virul...
0176
Rafal Mostowy @rafalmostowy.bsky.social · 10/06/2026
Huge congrats to @vyshakhrp.bsky.social for pulling this through, and to our fantastic collaborators from University of Wroclaw, University of Copenhagen, Lund University and Manchester Metropolitan University!
000
Rafal Mostowy @rafalmostowy.bsky.social · 10/06/2026
We propose 5 receptor-recognition strategies targeting the K. pneumoniae surface: capsule, O-antigen, LPS core, outer membrane proteins, resulting in diverse infection strategies. The full structural atlas is freely available at Klebsiella-Phage-RBP-Atlas. 🔗 vyshakhrp.github.io/klebsiella-r...
220
Rafal Mostowy @rafalmostowy.bsky.social · 10/06/2026
Structurally different RBPs often share near-identical N-terminals. The most striking case: within Webervirus, near-identical genomes swap a depolymerase for an L-shaped tail fiber at a single locus — 85% N-terminal identity preserved, host-range changed.
100
Rafal Mostowy @rafalmostowy.bsky.social · 10/06/2026
RBPs showed remarkable levels of structural modularity. The most common type was N-terminal sharing, but other types were also observed (C-terminal sharing/variation, region conservation, recomb. hotspots). Many of these were relatively recent.
110
Rafal Mostowy @rafalmostowy.bsky.social · 10/06/2026
We modelled all 382 RBPs with AlphaFold3 and resolved 39 structurally distinct classes — including 18 never described before — across all morphotypes. This is what the structural diversity of Klebsiella phage RBPs actually looks like.
100
Rafal Mostowy @rafalmostowy.bsky.social · 10/06/2026
The dataset: 192 Klebsiella phages across 3 morphotypes — podoviruses, siphoviruses, myoviruses (incl. jumbophages) — spanning 96 capsule types, with halo phenotypes and empirical and genetic host-range all determined. The broadest K. pneumoniae phage collection studied structurally to date.
101
Rafal Mostowy @rafalmostowy.bsky.social · 10/06/2026
Phage receptor-binding proteins are known for their LEGO-like modularity, but we still do not fully understand their evolutionary potential and how it shapes phage host range. Our new preprint examines this systematically in Klebsiella phages. 🔗 www.biorxiv.org/content/10.6... Thread 🧵
34622
Rafal Mostowy @rafalmostowy.bsky.social · 08/06/2026
And I presume labs will still hire PhD students who use AI 😆
100
Reposted by Rafal Mostowy
Piotr Rozwalak @prozwalak.bsky.social · 11/05/2026
Mushuvirus is the most widespread phage genus in the human gut. 🌍 Together with other family members, these viruses occur in 89% of humans worldwide, including the Iceman Ötzi! How is it possible that they were hidden in previous metagenomic analyses? [1/7] Read: doi.org/10.64898/202...
23114
Rafal Mostowy @rafalmostowy.bsky.social · 01/06/2026
DepoCatalog maps the landscape of depolymerase diversity. What drives it — modularity, recombination, specificity switching — is exactly what we're working to uncover. More on this soon. 👀
000
Rafal Mostowy @rafalmostowy.bsky.social · 01/06/2026
The paper was led by co-first authors Aleksandra Otwinowska & Sebastian Olejniczak (Wrocław) — with important contributions from our own Janusz Koszucki & @vyshakhrp.bsky.social, all supported by our @ncn.gov.pl grant. A genuine team effort between Wrocław and Kraków depocat.uwr.edu.pl
depocat.uwr.edu.pl
DepoCat
100
Rafal Mostowy @rafalmostowy.bsky.social · 01/06/2026
Thrilled to share a major milestone from our long-running collaboration with Zuzanna Drulis-Kawa's lab in Wrocław — DepoCatalog, is now out in @natcomms.nature.com: 129 Klebsiella phage depolymerases, 75 capsule types, 5-class structural framework and public database. www.nature.com/articles/s41...
3237
Reposted by Rafal Mostowy
Andrew Roger @andrewjroger.bsky.social · 06/05/2026
IQ-TREE 3: Phylogenomic Inference Software using Complex Evolutionary Models academic.oup.com/mbe/article/...
academic.oup.com
IQ-TREE 3: Phylogenomic Inference Software using Complex Evolutionary Models
Abstract. IQ-TREE (https://iqtree.github.io/) is a widely used open-source software tool for efficiently inferring phylogenetic trees under maximum likelih
2210105
Reposted by Rafal Mostowy
Martin Polz @polzlab.bsky.social · 06/05/2026
Very excited to share our new paper out in Nature. Congrats to Xiaoqian (Annie) Yu and all the coauthors. www.nature.com/articles/s41...
nature.com
Genome-wide sweeps create ecological units in the human gut microbiome - Nature
Genome-wide selective sweeps commonly occur in the human gut microbiome and can spread across the world within decades to produce epidemic-like population structures.
28243
Reposted by Rafal Mostowy
David Bikard @dbikard.bsky.social · 06/05/2026
Master Internships open at Eligo: www.linkedin.com/posts/duport...
linkedin.com
#internship #syntheticbiology #genetherapy #microbiometherapeutics #biotechcareers | Xavier Duportet
If "engineer bacteria to fight diseases" is your kind of fun, come build it with us. 4 Master's internships open at Eligo for Fall 2026 to work on real drug projects: engineered phage-particles to del...
075
Rafal Mostowy @rafalmostowy.bsky.social · 04/05/2026
Polish phage science meets Polish visual storytelling 🇵🇱🦠 To celebrate our last paper, we collaborated with an artist who previously worked on The Witcher 3 — to create a comic about phages and why understanding them matters in the fight against antibiotic-resistant bacteria. Comic below 👇
23011
Rafal Mostowy @rafalmostowy.bsky.social · 30/04/2026
That is fascinating! Could you point me to a reference please? I’d like to better understand how this works. Thank you!
100
Reposted by Rafal Mostowy
Elisabeth Bik @elisabethbik.bsky.social · 30/04/2026
J. Craig Venter, Scientist Who Decoded the Human Genome, Dies at 79 A risk-taking outsider, he brought speed, competition and controversy to one of science’s biggest races. #GiftLink Nicholas Wade writes @NYtimes.com www.nytimes.com/2026/04/30/s...
nytimes.com
J. Craig Venter, Scientist Who Decoded the Human Genome, Dies at 79
410862
Rafal Mostowy @rafalmostowy.bsky.social · 29/04/2026
That would be my guess but just a hypothesis at this point. Very not obvious how sugar modifications affect the activity of depos. There is K2 work suggesting it might not always do, but it should depend on a lot of stuff.
210
Rafal Mostowy @rafalmostowy.bsky.social · 29/04/2026
Thanks Olaya, I missed this paper and have to read it urgently now! But I must admit we have been engaging quite a lot with your work on the CPS evolution in another MS on phage RBPs that we plan to preprint soon.
120
Rafal Mostowy @rafalmostowy.bsky.social · 29/04/2026
The big picture: depolymerases recognise the sugar backbone; SGNH hydrolases recognise acetylation state. Two different keys for the same lock — and an ongoing molecular arms race over capsule chemistry with fascinating implications for prophage host dependence and evolution.
120
Rafal Mostowy @rafalmostowy.bsky.social · 29/04/2026
Some GWAS hits pointed to completely uncharacterised RBPs with unknown folds — new biology waiting to be understood. Others resembled acetyltransferases, suggesting phages may actively add modifications to capsules, not just remove them. This suggests fascinating underlying evolutionary dynamics!
110
Rafal Mostowy @rafalmostowy.bsky.social · 29/04/2026
The SGNH hydrolases: we produced two GWAS-predicted candidates and tested them. Both confirmed as active esterases — stripping O-acetyl groups from the capsule rather than degrading it. Not destroying the armour. Quietly removing its chemical decorations. A different mechanism entirely.
100
Rafal Mostowy @rafalmostowy.bsky.social · 29/04/2026
Both activity and specificity were hard to predict. Structurally related depos targeted different K-types. Divergent sequences sometimes converged on the same target. Even near-identical sequences could behave completely differently. The genotype-phenotype map of these enzymes is genuinely complex.
120
Rafal Mostowy @rafalmostowy.bsky.social · 29/04/2026
We screened 50 candidates: 14 active, 34 did not express — and nothing in sequence, structure or genomic context predicted which ones would work. Of the 14 active, 5 targeted a different K-type than their host — consistent with capsule switching (Haudiquet et al., PLoS Biol 2021).
100
Rafal Mostowy @rafalmostowy.bsky.social · 29/04/2026
GWAS found predictors for 16/35 K-loci. Structural modelling revealed striking diversity: classical β-helix depolymerases yes — but also a major class of enzymes (SGNH hydrolases), novel RBPs and even acetyltransferases. The K-type associated toolkit is far broader than expected.
110
Rafal Mostowy @rafalmostowy.bsky.social · 29/04/2026
Two parallel approaches. GWAS on >2,500 Klebsiella genomes — no assumptions about protein function, just asking which prophage proteins associate with which capsule type. Independently: an experimental screen of recombinant candidate proteins tested against 118 K-types. Then we compared.
121
Rafal Mostowy @rafalmostowy.bsky.social · 29/04/2026
This raised 3 questions: ❓ What actually determines capsule tropism in prophages — are these really depolymerases? ❓ Why are depolymerases so hard to detect in prophage genomes — biology or annotation limits? ❓ When we do find candidates, are they even active?
100
Rafal Mostowy @rafalmostowy.bsky.social · 29/04/2026
Depolymerases match phages to specific capsules — different enzyme ↔ different sugar. But in prophages, they're surprisingly hard to find. A recent AI model by Concha-Eloko et al. tried to predict capsule specificity from prophage sequences, but struggled. We wanted to understand why.
120
Rafal Mostowy @rafalmostowy.bsky.social · 29/04/2026
🎉 New in @plosbiology.org: Temperate phages of Klebsiella pneumoniae show capsule-driven host tropism — but how? The working assumption was depolymerases: enzymes that chew through the bacterial sugar capsule. Turns out that's not the whole story. 🧵 journals.plos.org/plosbiology/...
64919
Reposted by Rafal Mostowy
Peter Fineran @peterfineran.bsky.social · 18/03/2026
Great fun presenting some of our research @otagomicroimmuno.bsky.social @universityofotago.bsky.social Including transposon mutagenesis of phages. www.biorxiv.org/content/10.6...
biorxiv.org
High-throughput transposon mutagenesis defines the essential genome of diverse phages
Phages are important drivers of bacterial evolution with therapeutic potential as antimicrobials. However, gaps in our understanding of phages and our inability to rapidly engineer them with new genetic cargo hinders progress towards phage-based therapies. To address the lack of unbiased, genome-wide mutational tools for phages, we developed transposon mutagenesis employing CRISPR-anti-CRISPR (Acr)-based selection and deep-sequencing (Phage Tn-seq). Transposon mutagenesis was effective for phages with unmodified or hypermodified genomes and a jumbo phage that protects its DNA within a nucleus. Phage Tn-seq enabled phage gene essentiality assignment consistent with structural proteomics and core gene conservation. Insertion biases allowed prediction of transcriptional direction and early injected phage DNA regions. We exploited the method to rapidly deliver new cargo to phage genomes in just a few days and used an AI-designed Acr to expand the phage transposon toolbox. Phage Tn-seq is a versatile tool to advance our understanding and applications of phages. ### Competing Interest Statement The authors have declared no competing interest. Royal Society Te Apārangi, https://ror.org/04tajb587 Alexander von Humboldt Foundation, https://ror.org/012kf4317 The L’Oréal Groupe Australia University of Otago, https://ror.org/01jmxt844
0184
Reposted by Rafal Mostowy
Eduardo Rocha @epcrocha.bsky.social · 17/03/2026
Horizontal gene transfer is often depicted as a process distributing pre-existing functions to novel genetic backgrounds. Yet HGT can also increase the rate of functional innovation after transfer. Here's a brief review on the topic: ecoevorxiv.org/repository/v... #evosky #microsky
ecoevorxiv.org
From Trading Genes to Crafting New Tricks: How Horizontal Gene Transfer Potentiates the Emergence of Novel Functions
19142
Reposted by Rafal Mostowy
Mart Krupovic @mkrupovic.bsky.social · 03/03/2026
With Eugene Koonin, we propose a concept of “the selfish ribosome”, under which evolution of life is viewed as a ribosomal takeover, where the ribosome evolved to consume most of the cell’s resources, while other cellular componentry ensures the propagation of the ribosome. arxiv.org/abs/2602.23268
5250106
Rafal Mostowy @rafalmostowy.bsky.social · 26/02/2026
Official Euraxess ad (deadline 25 March): www.euraxess.pl/jobs/412428 If interested, please contact me early — I’d be happy to discuss the project before the formal deadline.
euraxess.pl
Assistant Professor 27.2026
1227.1101.27.2026 RECTOR of the Jagiellonian University announces a selection procedure for the position of an ASSISTANT PROFESSOR (F/M)
000
Rafal Mostowy @rafalmostowy.bsky.social · 26/02/2026
Looking for a strong computational biologist with experience in genomic/protein sequence analysis and a publication track record. Based at MCB (UJ), collaborating with Zuzanna Drulis-Kawa’s group (Wrocław). Informal enquiries very welcome.
100
Rafal Mostowy @rafalmostowy.bsky.social · 26/02/2026
📢 Hiring a Postdoc in Computational Phage Biology (Kraków, Poland) We study the evolution & structural modularity of prophage-encoded glycan-degrading enzymes in Klebsiella pneumoniae — combining genomics and AlphaFold-based analyses. Details in attached PDF 👇
docs.google.com
Postdoc Ad
Postdoctoral Position in Computational Phage Biology Microbial Genomics Group, Malopolska Centre of Biotechnology (MCB), Jagiellonian University, Kraków PL Summary We are seeking a postdoctoral rese...
11012
Reposted by Rafal Mostowy
George Bouras @gbouras13.bsky.social · 14/01/2026
Phold's manuscript is now available @narjournal.bsky.social thanks to @susiegriggo.bsky.social @npbhavya.bsky.social @vijinim.bsky.social @linsalrob.bsky.social @martinsteinegger.bsky.social @milot.bsky.social @eunbelivable.bsky.social & others not on bsky #phagesky academic.oup.com/nar/article/...
academic.oup.com
Protein structure-informed bacteriophage genome annotation with Phold
Abstract. Bacteriophage (phage) genome annotation is essential for understanding their functional potential and suitability for use as therapeutic agents.
18644
Rafal Mostowy @rafalmostowy.bsky.social · 14/11/2025
Congratulations Zam! Very well deserved 🥳🍾🎉👏
100
Reposted by Rafal Mostowy
Beatriz Beamud @beatrizbeamud.bsky.social · 08/10/2025
Can we exploit past phage infection events (prophages) to decipher the specificity of phage receptor-binding proteins such as depolymerases?🔎 Happy to share our recent work at @natcomms.nature.com 🔽 #microsky #phagesky www.nature.com/articles/s41...
nature.com
Unlocking data in Klebsiella lysogens to predict capsular type-specificity of phage depolymerases - Nature Communications
Here, the authors exploit the genetic information encoded in Klebsiella prophages to model the interplay between bacteria, prophages, and their depolymerases, using a directed acyclic graph-model and a sequence clustering-based model.
24926