GAMA Miguel Angel

1/12 🧵 Do you want to learn how to design proteins using AI but don’t know anything about biology? I created a free 10-lesson course on YouTube. It’s now available in Spanish (original) and English (autodubbing w/Kokoro 82M). Here’s an overview of the topics covered in each lecture :)

GAMA Miguel Angel reposted

LoL-align is out in @natmethods.nature.com! It aligns protein structures without superposition. More sensitive than TM-align and DALI, producing better alignments 5–20× faster. Now available in Foldseek! 📄 www.nature.com/articles/s41... 💾 github.com/steineggerlab/foldseek 🌐 search.foldseek.com

GAMA Miguel Angel reposted

AlphaFold Database is expanding into pandemic preparedness. Together with NVIDIA, DeepMind, EBI et al. we exhaustively predicted ~1.7 million homo- & heterodimers across 2,812 viral proteomes, resulting in 8,028 high-confidence predictions. 📄 research.nvidia.com/labs/dbr/ass... 🌐 alphafold.ebi.ac.uk

GAMA Miguel Angel reposted

Milot Mirdita

@milot.bsky.social

16 Sept

ColabFold 1.6.3 is out! 2.5x faster, pip-installable, ipSAE+pDockQ2 scores. Thanks Choonghwan Lee, Marielle Russo, Gyuri Kim 🐍pip install colabfold[alphafold] CF2 Sneak Peak with AF3/Boltz/Protenix/ESMFold2… 🐍pip install "colabfold[alphafold3]@git+https://github.com/sokrypton/ColabFold@af3-preview"

GAMA Miguel Angel reposted

Sergey Ovchinnikov

@sokrypton.org

17 Sept

Introducing highly experimental localfold.org Building on @martinsteinegger.bsky.social af2 webgpu port, @milot.bsky.social optimizations & jax ports of af3-like models by @marielle.bsky.social, Choonghwan Lee, Julia Buhmann. WARNING: runs directly on your 💻, may drain 🪫 & eat data📱 & overheat 🔥💻

Martin Steinegger 🇺🇦@martinsteinegger.bsky.social

Run AlphaFold2 locally in your browser via WebGPU, no installation needed, it's just a static webpage using your own GPU or CPU. Short proteins run in seconds. Larger ones are still slow: a 291-aa protein takes ~7 minutes on my M4 Pro. ❗It's just a PoC. 🌐 martin-steinegger.github.io/alphafold2-w...

GAMA Miguel Angel reposted

Protein Structure Evolution (ProSE) Seminar

13 Sept

Fold Spacer lets you fly through protein structures (Weekend project #2). It’s my first game: I originally set out to build a racer with structures as the tracks, but was a little too crazy. So it became this instead. You can upload your own structures. 🌐 martin-steinegger.github.io/Fold-Spacer/

GAMA Miguel Angel reposted

Sergey Ovchinnikov

@sokrypton.org

14 Sept

Finally a more intuitaive way to learn pLDDT/pAE? 😎 sokrypton.github.io/protein_figh... (Character idea from @hannes-stark.bsky.social & Alex Waldherr)

GAMA Miguel Angel reposted

@proteinstructure.bsky.social

1 Sept

Join us next Tuesday, 5PM CET for our first ProSE after the summer! Betül Kaçar @kacarlab.bsky.social will talk about the origin and early evolution of ancient proteins! 📜🧬 tinyurl.com/prose-seminar2

GAMA Miguel Angel reposted

Lada Isakova

@ladaisa.bsky.social

3 Apr

Excited to share that my first paper is finally out in @pnas.org! We ask what limits the exploration of protein sequence space and find that shared ancestry and divergence time play a much larger role than selection or epistasis. www.pnas.org/doi/10.1073/...

pnas.orgDescent from a common ancestor restricts exploration of protein sequence space | PNASHow functional protein sequences are distributed in sequence space is fundamentally important for evolutionary theory and protein design, particula...

GAMA Miguel Angel reposted

Philip Romero

@philromero.bsky.social

18 Aug

What if AI could interact directly with biology? Congrats to Coban, who gave AI the ability to experiment and learn through feedback. Over 25 autonomous rounds, it uncovered the determinants of enzyme specificity. Give AI the ability to experiment, then get out of the way. doi.org/10.64898/202...

doi.orgLearning protein function through autonomous experimental interactionBiological AI learns primarily from existing observations, but many questions cannot be answered from available data alone. Here we show that AI can instead acquire knowledge by acting directly on biological systems and learning from the consequences. We developed a closed-loop framework in which autonomous agents design protein variants, construct and characterize them in a robotic laboratory, learn from the resulting experimental feedback, and decide what experiments to perform next. We then allowed the system to operate continuously and without human intervention for approximately one month, during which multiple agents independently explored protein sequence space while learning from shared experimental experience. Applied to glycoside hydrolases, the agents discovered enzymes with substantially altered substrate specificity toward non-native sugars and progressively learned the structure of the underlying sequence-function landscape. The resulting experimental experience also revealed determinants of substrate specificity and protein expression that were not specified as learning objectives. These results demonstrate that AI can autonomously interact with biology over extended periods to acquire knowledge through experience, establishing a framework for biological discovery driven by continuous experimental interaction. ### Competing Interest Statement The authors have declared no competing interest. National Institute of General Medical Sciences, 5R01GM150929

GAMA Miguel Angel reposted

Adam Phillippy

@aphillippy.bsky.social

6 Aug

For the past 30 years, “whole-genome sequencing” has been a misnomer. Today the T2T Consortium publishes a dozen papers heralding a future of truly complete genomes for humans and nearly any vertebrate 👨‍🔬🐒🐦🐀🦒🐎🫏🐹🐟 (sorry, no salamanders): www.cell.com/consortium/t... 🧵[1/15]

GAMA Miguel Angel reposted

1 Aug

Riboseek is a fast RNA/DNA search. More sensitive than nhmmer at 250x speed. Structure-aware realignment produces MSAs approaching rMSA quality. Plus 1.7M precomputed RNA MSAs, and an API to search your own 📄 www.biorxiv.org/content/10.6... 💾 github.com/steineggerla... 🌐 search.foldseek.com/riboseek

GAMA Miguel Angel reposted

Johanna von Wachsmann

@johannavw.bsky.social

16 June

🧬 New preprint! We clustered 5.6 million bacterial genomes into genomically cohesive units (GCUs) 500× faster than existing tools. (In just 14 hours, 16.5 GB RAM using 48 CPUs). 🦠🐙Meet gemsparcl 💎✨! www.biorxiv.org/content/10.6...

GAMA Miguel Angel reposted

6 June

Replying to @martinsteinegger.bsky.social

Meet the Folddisco Marv, designed by Hyunbin Kim, who also developed Folddisco.

GAMA Miguel Angel reposted

nature.comStructural motif search across the protein universe with Folddisco - Nature BiotechnologyFolddisco enables protein structural motif search in million scale databases.

6 June

Folddisco is now published @natbiotech.nature.com. It’s a fast motif search for similar 3D DISCOntinuous residues like catalytic sites or zinc fingers across the entire protein universe. 📄 www.nature.com/articles/s41... 💾 folddisco.foldseek.com 🌐 https://search.foldseek.com/folddisco

Martin Steinegger 🇺🇦@martinsteinegger.bsky.social

Folddisco finds similar (dis)continuous 3D motifs in large protein structure databases. Its efficient index enables fast uncharacterized active site annotation, protein conformational state analysis and PPI interface comparison. 1/9🧶🧬 📄 www.biorxiv.org/content/10.1... 🌐 search.foldseek.com/folddisco

20 May

Today, the AlphaFold2 paper reached the milestone of 50k citations according to Google Scholar! And AlphaFold3 will likely reach 15k citations tomorrow. Congratulations to the entire AlphaFold team, as well as to all the scientists who helped democratize protein structure prediction 🥳

Protein Structure Evolution (ProSE) Seminar

20 May

If you still do not know how AF2/AF3 work, as well as everything that has happened since their release, here I explain in detail (~8 hours) the “AlphaFoldmania” :) bsky.app/profile/mian...

GAMA Miguel Angel@miangoar.bsky.social

8/12 In the 8th lecture "AlphaFold" (7h 39mins), we review in detail the AF2 and AF3 architectures, how they revolutionized structural biology, their strengths and weaknesses and what the post-AlphaFold era looks like for protein design youtu.be/4K8SDxk85a0

GAMA Miguel Angel reposted

@proteinstructure.bsky.social

5 May

Join ProSe next week Tuesday, when Noelia Ferruz @noeliaferruz.bsky.social is talking about "Controllable Protein Design with Protein Language Models and Reinforcment Learning", TUE, May 12, 5PM CET! Sign-up here: tinyurl.com/prose-seminar2

GAMA Miguel Angel reposted

ace-gtdb.bsky.social

@ace-gtdb.bsky.social

15 Apr

GTDB release 11 based on RefSeq 232 (R11-RS232) is live at gtdb.ecogenomic.org. This release covers 901,341 genomes (23% increase) and has 199,923 species clusters (39% increase). Release notes at: forum.gtdb.ecogenomic.org/t/announcing.... Release statistics at: gtdb.ecogenomic.org/stats/r232.

gtdb.ecogenomic.orgGTDB - Genome Taxonomy DatabaseThe Genome Taxonomy Database (GTDB) is an initiative to establish a standardised microbial taxonomy based on genome phylogeny.

github.comGitHub - miangoar/AI-driven-protein-design: Resources for learning AI-driven protein designResources for learning AI-driven protein design. Contribute to miangoar/AI-driven-protein-design development by creating an account on GitHub.

11/12 I also created this GitHub repository with: +300 tools organized into 25 categories +70 databases (12 categories) +130 learning resources (9 categories) github.com/miangoar/AI-... Among them, I want to highlight this other free course by Kieran Didi structural-bioinformatics.netlify.app

11/12 All the course information, the link to the slides, and more can be found on my site in GithubPages. In addition, YouTube has automatically dubbed the course into 18 other languages to make learning more accessible. I hope you find it useful :) miangoaren.github.io/teaching/pro...

9/12 In the 9th lecture "AI-driven protein design" (8h 27min), we cover the design toolkit: from directed evolution and rational design, to protein language models (representation learning) and generative AI to create both protein sequences and structures youtu.be/PvMNlxZv_Bg

10/12 In the final lecture "Data & biases" (4h 16min), we discuss the main biological databases (eg. PDB, UniProt, NCBI), data cleaning strategies, data leakage, and the biases that can silently affect the generalization capabilities of your models youtu.be/bEt7tZKvfiI

7/12 In the 7th lecture "Protein evolution" (2h 38mins, and my personal favorite), we trace how proteins originated from simple peptides, how mutations shape the evolutionary paths and how epistasis drives the evolution of proteins youtu.be/rkmWSR8BUms

5/12 In the 5th lecture "Protein structure" (2h 55mins), we explore the principles of structural biology: from amino acids and secondary structure to fold classification schemes and the uneven shape of the protein universe youtu.be/7GmPNVhJhw0

6/12 In the 6th lecture "Protein function" (1h 58mins), we cover how proteins fold inside the cell, how enzymes work and how function is regulated through distinct mechanisms like allostery, post-translational modifications, and proteostasis youtu.be/Un6QaTM412A

3/12 In the 3rd lecture "Deep learning" (1h 54mins), we’ll review how neural networks work, from neurons and backpropagation to modern architectures. Then we explore the main DL frameworks used to build models youtu.be/YiEmCQuW-xc

4/12 In the 4th lecture "Transformers & language models" (3h 42mins), we break down how the original Transformers architecture work, the differences between BERT and GPT, scaling laws, modern LLMs and how to work with them youtu.be/tNAKnz_tDIc

2/12 In the first lecture "Basic computing concepts" (1h 58mins), we’ll review how CPUs and GPUs work, as well as essential software for data analysis like GNU/Linux and the python ecosystem for bioinformatics youtu.be/RddVvvYRpTc

3/12 In the 2nd lecture "Machine learning" (1h 39mins), we’ll review what artificial intelligence is and its subfields, the current capabilities of the algorithms, and how a model is trained in general youtu.be/9fEl5RsLKJs

GAMA Miguel Angel reposted

17 Mar

AlphaFold database has entered the era of complexes. Together with NVIDIA, DeepMind and EBI, we use ColabFold, OpenFold and MMseqs2-GPU to predict ~31 million complexes (homo & hetro-dimers) resulting in 1.8 million high-quality predictions 📄 research.nvidia.com/labs/dbr/ass... 🌐 alphafold.ebi.ac.uk

GAMA Miguel Angel reposted

Yun S. Song

@yun-s-song.bsky.social

21 Feb

Can we simulate realistic evolutionary trajectories and “replay the tape of life”? In this work, we propose a flexible, generalizable deep learning framework for modeling how the entire protein sequence evolves over time while capturing complex interactions across sites. 1/n doi.org/10.64898/202...

doi.orgdoi.org

GAMA Miguel Angel reposted

Richard McElreath 🐈‍⬛

@rmcelreath.bsky.social

12 Feb

Jeffrey Ross-Ibarra@jrossibarra.bsky.social

This isn't shortread data, but alignments of genome assemblies with N50 > 100Mb. The plot is the # of bp per 100Kb window that are dropped because of missing data -- because those bp physically are not there in one of the other 25 genomes. This is what normal plant genomes look like, folks.

bluesky-map.theo.ioBluesky MapInteractive map of 3.4 million Bluesky users, visualised by their follower pattern.

12 Feb

I am in the "Life Sciences Super Cluster" and quite far away from many colleagues in protein design who are in the cluster called "Computational Chemistry Nexus" 😭

Theo Sanderson@theo.io

I made a map of 3.4 million Bluesky users - see if you can find yourself! bluesky-map.theo.io I've seen some similar projects, but IMO this seems to better capture some of the fine-grained detail

6 Feb

Does anyone know the meaning of AlphaGenome and its impact? It’s not my area, so I don’t know how important it is. But I think is not equivalent to AlphaFold2, since no other area in biology has the high-quality data and structure provided by PDB, UniProt and CASP competition.

doi.orgFrom Binding to Catalysis: Emergence of a Rudimentary Enzyme Conferring Intrinsic Antibiotic ResistanceAbstract. How does enzymatic activity emerge? To shed light on this fundamental question, we study type B dihydrofolate reductases (DfrB), which were disco

6 Feb

Sometimes I see Nature papers as elegant $13k commercials from AI companies inviting you to subscribe to their chatbots

GAMA Miguel Angel reposted

Claudèle Lemay-St-Denis

@claudele.bsky.social

8 Oct 2025

How does catalysis emerge from non-catalytic domains? In our new paper, we show that catalytic activity can arise without conserved active-site residues — through multimerization and electrostatic features instead. A striking case of catalysis evolving from binding.

Molecular Biology and Evolution@molbioevol.bsky.social

@claudele.bsky.social et al. studied the fundamental question of how enzymatic activity emerges by focusing on type B dihydrofolate reductases, suggesting DfrB domains evolved rudimentary catalysis from a binding capacity. 🔗 doi.org/10.1093/molbev/msaf215 #evobio #molbio

GAMA Miguel Angel reposted

Klara Hlouchova lab

@hlouchova-lab.bsky.social

3 Nov 2025

Can proteins fold and function with half of the amino acid alphabet? Using only 10 residues, we designed stable, mutation-resilient structures—no aromatics or basics involved. A minimalist foundation for ancient biology and synthetic design. tinyurl.com/37t8br4v #ProteinDesign #OriginsOfLife

tinyurl.comAncient amino acid sets enable stable protein foldsEarly proteins likely arose from a chemically limited set of amino acids available through prebiotic chemistry, raising a central question in molecular evolution: could such primitive compositions yie...

4 Feb

I recorded ~4h where we cover the main bio databases, data processing methods, many sources of bias and topics like generalization and data leakage :) youtu.be/SKpHaHgvCKE Slides drive.google.com/file/d/1jpEwDBncJCRviG_DaWs2EpzCL_1BfB9t/view English is available only via auto-translated subtitles

GAMA Miguel Angel@miangoar.bsky.social

🧵1/3 I created this free 37-hour course, distributed across 10 lectures, to introduce AI-based protein design. For more information about the course and its specific topics, please visit the official course page:

3 Feb

I recorded ~8h introducing the main algorithms for protein design: from classical approaches to protein language models, AlphaFold, ESMFold, MPNN, diffusion models and more :) youtu.be/wKUYtAt87d4T... Slides drive.google.com/file/d/1EPLj... English is available only via auto-translated subtitles

GAMA Miguel Angel@miangoar.bsky.social

2 Feb

I’ve recorded ~8h explaining the architectures of AlphaFold, AF2 & AF3, as well as the context needed to understand their development, applications and limitations :) youtu.be/_jDRr5BcTaY Slides drive.google.com/file/d/1i4QE... English is available only via auto-translated subtitles

GAMA Miguel Angel@miangoar.bsky.social

1 Feb

The 7th lecture is available on YouTube :) We will review how proteins emerge and diversify throughout evolution, considering mutations and molecular interactions youtu.be/qaypRS8SX5M Slides drive.google.com/file/d/1BfQd... English is available only via auto-translated subtitles

GAMA Miguel Angel@miangoar.bsky.social

nature.comResurrected nitrogenases recapitulate canonical N-isotope biosignatures over two billion years - Nature CommunicationsThe study shows that nitrogenase enzymes have maintained stable isotope signatures over billions of years, revealing how ancient microbes shaped Earth’s nitrogen cycle and offering a new experimental ...

1 Feb

2/2 I’ve reviewed many courses, yet few give evolution the importance it deserves. They acknowledge it, but rarely go beyond algorithms like AlphaFold. Understanding evolution helps us understand how our models are biased and how to mitigate those biases.

GAMA Miguel Angel reposted

Kaçar Lab at UW-Madison

@kacarlab.bsky.social

30 Jan

Our new paper, out today! We resurrected ancient nitrogenases first used by life on Earth 3 billion years ago. We combined synthetic biology and geology & validated their chemical #biosignature in rocks that helps reveal ancient life on Earth!(and beyond!) Link: www.nature.com/articles/s41...

GAMA Miguel Angel reposted

Clockwork

@watchclockwork.com

31 Jan

💪 NEW VIDEO: Flying over the A-band of an atomic-scale model of a vertebrate muscle sarcomere. Let's explore the molecular mechanics that make your muscles work. Rendered using @bradyajohnston.bsky.social 's molecular nodes Model based on the incredible work of the @raunser-lab.bsky.social lab

31 Jan

The 6th lecture is now available on YouTube :) We’ll review how proteins adopt their 3D shape, how they perform their functions and how their activity is regulated youtu.be/cZs8XtVYa5A Slides drive.google.com/file/d/1TpPj... English is available only via auto-translated subtitles

GAMA Miguel Angel@miangoar.bsky.social

30 Jan

The fifth lecture of the course is now available on YouTube :) We’ll review amino acid chemistry and how we organize and classify proteins youtu.be/gE6qXwpBP_s Slides drive.google.com/file/d/1F99V... For now, the English version is only available through the automatic translation of the subtitles

GAMA Miguel Angel@miangoar.bsky.social