Sign in

Kelvin Lau @klausenhauser on 🙅‍♂️🐘🪡

@klausenhauser.bsky.social
2.2K followers 2.1K following 1.1K posts

🇨🇦🇭🇰 @EPFL_en protein production + structure facility 🇨🇭| ex @ubc @uwaterloo @unige_en | here for #proteins #structuralbiology #xray #cryoem #plants #mountains #politics #transit #urbanism 🐱🐈 staying for the #hottakes | My views here🧪🧬🌱🔬🏳️‍🌈

PostsRepliesMedia
Kelvin Lau @klausenhauser on 🙅‍♂️🐘🪡 @klausenhauser.bsky.social · 02/10/2026
too many possibilities for molecular replacement (20 domains all same fold) + aniso diff 5A / 2.5A + 60% solvent. not profoundly surprising but much smoother experience than any available single software. prompt. density. structure. with a side of dashboard.
110
Kelvin Lau @klausenhauser on 🙅‍♂️🐘🪡 @klausenhauser.bsky.social · 01/10/2026
Watermarking proteins, why? How is this even a nature paper? I agree with the many reviewers that pushed back. Also this is open peer review? Redacations www.nature.com/articles/s41...
nature.com
Function-preserving watermarking of AI-generated proteins - Nature
SynthIDBio watermarks AI-generated protein sequences and structures while preserving function, enabling provenance detection and offering a proof-of-concept tool for biosecurity and information integr...
3111
Kelvin Lau @klausenhauser on 🙅‍♂️🐘🪡 @klausenhauser.bsky.social · 26/09/2026
AI accelerating biology research for me.... not tonight!
240
Reposted by Kelvin Lau @klausenhauser on 🙅‍♂️🐘🪡
Martin Pacesa @martinpacesa.bsky.social · 21/09/2026
ʙɪɴᴅᴄʀᴀꜰᴛ2 is out, and we're not waiting for the paper. The full code drops today, free for academic and industry use. We're releasing it early so you can start designing right now, and bring its full power to the current Adaptyv competition. github.com/PacesaLab/Bi...
4292142
Kelvin Lau @klausenhauser on 🙅‍♂️🐘🪡 @klausenhauser.bsky.social · 16/09/2026
Did some reviews and…. Why do ppl show biophysical assays without the actual data? KD values with no curves or fits. An ITC table without a single raw thermogram. You wouldn’t publish a Western without the blot, but for binding assays we just say “trust me bro”. Free the curves!!
4233
Reposted by Kelvin Lau @klausenhauser on 🙅‍♂️🐘🪡
Sergey Ovchinnikov @sokrypton.org · 14/09/2026
Finally a more intuitaive way to learn pLDDT/pAE? 😎 sokrypton.github.io/protein_figh... (Character idea from @hannes-stark.bsky.social & Alex Waldherr)
727698
Reposted by Kelvin Lau @klausenhauser on 🙅‍♂️🐘🪡
Michael Hothorn @structplantbio.bsky.social · 11/09/2026
Nice work by @cellsensing.bsky.social Glycan recognition by a plant damage-sensing immune receptor www.science.org/doi/10.1126/...
science.org
Glycan recognition by a plant damage-sensing immune receptor
Pathogens target and degrade the extracellular matrix surrounding plant cells. A central question is how cell wall-derived damage-associated molecular patterns (DAMPs) are recognized and integrated to...
06734
Kelvin Lau @klausenhauser on 🙅‍♂️🐘🪡 @klausenhauser.bsky.social · 04/09/2026
you know youre out of it when I run a string of programs and its all out of date and expired. I guess its a total reinstall day
010
Reposted by Kelvin Lau @klausenhauser on 🙅‍♂️🐘🪡
Reese Richardson @reeserichardson.bsky.social · 25/08/2026
A massive update: At least ~15 companies~ are selling scientists antibodies using faked validation data. We've documented 18,000+ manipulated images on 17,000+ products sold by leading laboratory suppliers including Thermo Fisher, Abcam, Santa Cruz Biotechnology, Millipore Sigma and Bio-Techne. 1/🧵
18455329
Kelvin Lau @klausenhauser on 🙅‍♂️🐘🪡 @klausenhauser.bsky.social · 26/08/2026
And I wake up to my friend sending me this. I think this is the highlight of the probe. It’s just a cut and paste world doing WB…
1114
Kelvin Lau @klausenhauser on 🙅‍♂️🐘🪡 @klausenhauser.bsky.social · 25/08/2026
Pfttt so much manipulation. I wonder where these people went to school reeserichardson.blog/2026/08/25/a...
reeserichardson.blog
At least 15 companies are selling antibodies using faked validation data
Popular vendors’ catalogs feature more than 18,000 manipulated images
087
Kelvin Lau @klausenhauser on 🙅‍♂️🐘🪡 @klausenhauser.bsky.social · 21/08/2026
This was my first week back after a few weeks break. Reminder to everyone to put your brain to rest once in awhile. Only felt the impact after coming back! A virtual prize to the person that can guess the location
2101
Kelvin Lau @klausenhauser on 🙅‍♂️🐘🪡 @klausenhauser.bsky.social · 19/08/2026
@anthropic.com beat me to it. I wanted to orchestrate the same workflow. Super powerful and enabling. For those in the know this will greatly accelerate progress. I do still have my concerns for those who are beginners and juniors and then just implementing. www.anthropic.com/research/Cla...
anthropic.com
How Claude is accelerating protein design and analytical chemistry
In this post, we share two results that show how Claude can help life scientists increase the pace of their research. In the first, we tested Claude’s ability to design protein binders from scratch, a key step in creating protein-based drugs that has historically taken a specialist weeks or months per target. In the second example, we evaluated whether Claude can accelerate chemical analysis. Claude Opus 5, a generally available model, was given NMR and LC-MS data (the data that allows chemists to assess the identity and purity of the compounds they work with).
4216
Kelvin Lau @klausenhauser on 🙅‍♂️🐘🪡 @klausenhauser.bsky.social · 14/08/2026
Hmm quite some critical points hit already in 2026. Now there’s ai-2040.com?r=tw
ai-2040.com
AI 2040: Plan A
A detailed forecast and recommendation for how the US, China and the rest of the world should navigate superintelligence.
010
Kelvin Lau @klausenhauser on 🙅‍♂️🐘🪡 @klausenhauser.bsky.social · 17/07/2026
There are usually not reports or mentions during the protein design process that it does not match the original design or the oracle that filters it. What does that mean? Anyone have a word for that? Im sure some of you can appreciate that the model and data differ by -1 positions
240
Kelvin Lau @klausenhauser on 🙅‍♂️🐘🪡 @klausenhauser.bsky.social · 09/07/2026
Tried a workflow to increase productivity. Theres a lot of in silico cloning to do that is between both GUI and text. Give it whatever base vector, define where you want it. Give it the sequences (or have it find them for you) and then output back the final sequences and a table ready for a CRO
130
Kelvin Lau @klausenhauser on 🙅‍♂️🐘🪡 @klausenhauser.bsky.social · 01/07/2026
Excited to try out Claude Science today!!
280
Kelvin Lau @klausenhauser on 🙅‍♂️🐘🪡 @klausenhauser.bsky.social · 24/06/2026
finally tried it out with Coot 1.3.1 works great! just have to get used to the commands. but I wish there was a mutate binding.
1121
Kelvin Lau @klausenhauser on 🙅‍♂️🐘🪡 @klausenhauser.bsky.social · 23/06/2026
Thanks Claude Cowork! I just replaced 10 hours of searching on Cryosparc and Relion forums with just an afternoon of AI.
020
Reposted by Kelvin Lau @klausenhauser on 🙅‍♂️🐘🪡
WePlanet @weplanetint.bsky.social · 17/06/2026
A big win for science and sustainable farming in the EU! New Genomic Techniques (NGTs) have the potential to help farmers grow more resilient crops, reduce inputs such as pesticides and fertilisers, strengthen food security, and support a more sustainable agricultural system.
2126
Kelvin Lau @klausenhauser on 🙅‍♂️🐘🪡 @klausenhauser.bsky.social · 13/06/2026
First being a scientist is bad now you’re just banned because you’re a foreigner.. and I paid for the product!
140
Kelvin Lau @klausenhauser on 🙅‍♂️🐘🪡 @klausenhauser.bsky.social · 11/06/2026
Posted this in LinkedIn. Shameless advert for some new structures 👉 lnkd.in/gT9Be3tH. But more structural biologists here! Nice to have Claude control my ChimeraX. At least for visualizations and analysis I can see it really speeding up things. IF you know what you’re doing and what to ask for.
1274
Kelvin Lau @klausenhauser on 🙅‍♂️🐘🪡 @klausenhauser.bsky.social · 11/06/2026
Trying to help someone setup MCP for ChimeraX, wanted to see if Fable can do it. I cant even read a website on it anymore. Opus works fine.
210
Kelvin Lau @klausenhauser on 🙅‍♂️🐘🪡 @klausenhauser.bsky.social · 10/06/2026
Always good to remember that the PDB sequence of a structure is not always the sequence of the protein made nor the sequence of the protein analyzed, its the sequence of the protein a human remembered to type in.
2212
Reposted by Kelvin Lau @klausenhauser on 🙅‍♂️🐘🪡
Oli Clarke @olibclarke.bsky.social · 21/05/2026
CS tools is very handy! Just added another couple of CS tools scripts to my github - one for making adjustable templates for picking membrane proteins in nanodiscs & micelles: github.com/olibclarke/c... (cont. next post:)
Example nanodisc templates
1326
Kelvin Lau @klausenhauser on 🙅‍♂️🐘🪡 @klausenhauser.bsky.social · 20/05/2026
I think LinkedIn Games is the millienial digital crack cocaine…
040
Reposted by Kelvin Lau @klausenhauser on 🙅‍♂️🐘🪡
Nature Portfolio @natureportfolio.nature.com · 19/05/2026
Two papers in Nature present AI systems that can assist throughout multiple processes involved in scientific research. The systems are designed to assist researchers in accelerating scientific discovery, not to replace them. go.nature.com/4tFHspz go.nature.com/3RQzXyL #Academicsky 🧪
This is figure 1, which shows Co-Scientist design, multi-agent architecture, and experimental validation summary.
02810
Kelvin Lau @klausenhauser on 🙅‍♂️🐘🪡 @klausenhauser.bsky.social · 19/05/2026
Might enjoy using this tool and see if it’s the model building or innate to the dataset! CrystalBoltz soniaminseokim.github.io/crystalboltz...
soniaminseokim.github.io
CrystalBoltz: End-to-End Protein Structure Determination via Experiment-Guided Diffusion<br>for X-Ray Crystallography
Deformable Neural Radiance Fields creates free-viewpoint portraits (nerfies) from casually captured videos.
042
Kelvin Lau @klausenhauser on 🙅‍♂️🐘🪡 @klausenhauser.bsky.social · 07/05/2026
Phenix 2.1 ❤️Coot 1.2.x
161
Kelvin Lau @klausenhauser on 🙅‍♂️🐘🪡 @klausenhauser.bsky.social · 04/05/2026
Flower Dump! Lots of Alliums and Irises
030
Kelvin Lau @klausenhauser on 🙅‍♂️🐘🪡 @klausenhauser.bsky.social · 27/04/2026
I do enjoy this leaf
180
Kelvin Lau @klausenhauser on 🙅‍♂️🐘🪡 @klausenhauser.bsky.social · 24/04/2026
Another Friday thought. After I was cited unknowingly by some young scientists hundreds of km away 😅 (does this mean I went viral?) “AI: better make it your tool before it starts giving you the instructions”
161
Kelvin Lau @klausenhauser on 🙅‍♂️🐘🪡 @klausenhauser.bsky.social · 22/04/2026
Calling all xray structure folks that have worked with denovo proteins. Do you suffer from high rfactors? Myself, someone else I know and now multiple structures all seem to suffer from this. Assuming we all tried our best in model building, what could be some causes?
51512
Reposted by Kelvin Lau @klausenhauser on 🙅‍♂️🐘🪡
Torsten Schwede @torstenschwede.bsky.social · 13/03/2026
Is #AI hitting a plateau in structure prediction? Help us find out at CASP17! 🧪🧬 Calling for Targets: Immune Complexes, protein - ligand complexes, RNA/DNA, conformational ensembles, membrane proteins, viral origins, and large complexes. The Rule of Thumb: If AF3 can’t model it, we want it.
The Critical Assessment of Structure Prediction (CASP) experiment is calling for prediction targets: Immune Complexes, Organic Ligand-Protein Complexes, Nucleic Acids and Complexes, Conformational Ensembles, Difficult Protein Structures and Complexes. 
Rule of Thumb: If AlphaFold3 can generate a high-quality model, it is likely not a CASP-grade challenge. If it struggles, we want it.
24935
Kelvin Lau @klausenhauser on 🙅‍♂️🐘🪡 @klausenhauser.bsky.social · 21/04/2026
Ive always been taught verbatim for decades that Glycines are helix breakers, but why is this protein filled of alpha helices so rich in glycines everywhere!
000
Kelvin Lau @klausenhauser on 🙅‍♂️🐘🪡 @klausenhauser.bsky.social · 18/04/2026
Bye bye dogma? www.science.org/doi/10.1126/...
science.org
Protein-templated synthesis of dinucleotide repeat DNA by an antiphage reverse transcriptase
Defense-associated reverse transcriptases (DRTs) are widespread bacterial anti-phage systems that use unconventional mechanisms of polynucleotide synthesis. We show that DRT3, which comprises two dist...
0100
Kelvin Lau @klausenhauser on 🙅‍♂️🐘🪡 @klausenhauser.bsky.social · 07/04/2026
Some people cleave tags religiously. This person didnt and made me solve their perfectly structured Histag. Probably also helped in packing. Might have to do experiments without it to show its not an artifact though. Team cleave or not cleave?
an xray density map depicting blue meshed density where an atomic model has been built.
8293
Kelvin Lau @klausenhauser on 🙅‍♂️🐘🪡 @klausenhauser.bsky.social · 27/03/2026
I can finally stop cringing and sighing when I hear that sds-page separates only by size as most textbooks say
091
Kelvin Lau @klausenhauser on 🙅‍♂️🐘🪡 @klausenhauser.bsky.social · 11/03/2026
How much about thermodynamics of AI binders are known? I think this is the first time I’ve seen an endothermic interaction. (Kinetics is fast on fast off) #EMBLProteinQC
1100
Kelvin Lau @klausenhauser on 🙅‍♂️🐘🪡 @klausenhauser.bsky.social · 08/03/2026
Happy that when public services were put to a referendum (proposal to decrease of the annual household charge of 335 CHF (~$430) to 200 for the public broadcaster) that people understood and made the right choice.
130
Kelvin Lau @klausenhauser on 🙅‍♂️🐘🪡 @klausenhauser.bsky.social · 20/02/2026
Started this project years ago, but now everything is deposited on @addgene.bsky.social and a short @protocolsio.bsky.social is online. Try it out to make your own (cheap) homemade nuclease/Benzonase! dx.doi.org/10.17504/protocols.io.261ged7xov47/v2 Thank you @jcoker10.bsky.social and Michael Lim!
33212
Kelvin Lau @klausenhauser on 🙅‍♂️🐘🪡 @klausenhauser.bsky.social · 31/12/2025
First vacation that I spent more on museums and temples than on food. What a rich cultural history Egypt has spanning millennia
Inside the new Grand Egyptian MuseumView of the pyramids Cat in MosqueRamasseum
1190
Kelvin Lau @klausenhauser on 🙅‍♂️🐘🪡 @klausenhauser.bsky.social · 18/12/2025
Its that time of the year again OOO
050
Kelvin Lau @klausenhauser on 🙅‍♂️🐘🪡 @klausenhauser.bsky.social · 15/12/2025
My favourite place this weekend. And some new infrastructure sightings. New train and a new gondola mid level station (nicer than some airports!)
New BLM Bergman in front of the mountains, Mönch and Jungfrau A view of Eiger, Mönch and JungfrauInterior of the mid-mountain Birg station Ski slopes with Schilthorn in the distance
080
Kelvin Lau @klausenhauser on 🙅‍♂️🐘🪡 @klausenhauser.bsky.social · 12/12/2025
What’s your end of the year task? Cleaning all the reason we use in half a year. Of course on an indestructible Amersham era pump with its detector intact
2280
Kelvin Lau @klausenhauser on 🙅‍♂️🐘🪡 @klausenhauser.bsky.social · 11/12/2025
Seen elsewhere… and was triggered. Don’t think I would set up the void peak that has the lower melting temperature if I only had one chance. The small monomer and higher melting curve in the dotted example however.. don’t trust your AI infographics
071
Kelvin Lau @klausenhauser on 🙅‍♂️🐘🪡 @klausenhauser.bsky.social · 08/12/2025
Saw this nice short protocol paper on LI. Not totally HTP but totally feasible in an academic lab. Right in the intersection of drug discovery, automation, structural biology and protein expression. Ps the guys looking for a job downunder! www.biorxiv.org/content/10.6...
020
Kelvin Lau @klausenhauser on 🙅‍♂️🐘🪡 @klausenhauser.bsky.social · 04/12/2025
Most protein design binders look the same, bundles of alpha helices or clearly a VHH. But the strategies in getting there are getting more diverse. The pipelines that humans are deciding to use are super interesting. proteinbase.com/competitions...
proteinbase.com
Nipah Binder Competition
Design a protein capable of neutralizing the Nipah virus, a pathogen with up to 75% mortality rate and high pandemic potential, currently lacking effective treatments.
031
Reposted by Kelvin Lau @klausenhauser on 🙅‍♂️🐘🪡
Margot Riggi @margotriggi.bsky.social · 03/12/2025
Lots of fun working on this project! Looking forward to see more and more animations published in the future!
062
Reposted by Kelvin Lau @klausenhauser on 🙅‍♂️🐘🪡
Daniela Barillà @danielabarilla.bsky.social · 20/11/2025
Let’s wrap things up: my commentary on the Asgard hypernucleosomes. Congratulations to all the authors of the paper 🍾! kwnsfk27.r.eu-west-1.awstrack.me/L0/https:%2F... Cryo-EM reveals open and closed Asgard chromatin assemblies: Molecular Cell www.cell.com/molecular-ce...
cell.com
Cryo-EM reveals open and closed Asgard chromatin assemblies
Ranawat et al. show the cryo-EM structures of Asgard archaeal chromatin assemblies, revealing that the histone HHoB assembles into both compact closed and extended open hypernucleosomes. The closed co...
03913