Kelvin Lau @klausenhauser on 🙅♂️🐘🪡
@klausenhauser.bsky.social
🇨🇦🇭🇰 @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🧪🧬🌱🔬🏳️🌈
From a pathogenic bacteria…. Of flies
November 20, 2025 at 6:51 PM
From a pathogenic bacteria…. Of flies
Interesting SEC as well. Hint: putative lectin binder.
November 20, 2025 at 5:21 PM
Interesting SEC as well. Hint: putative lectin binder.
so good 🥮
November 18, 2025 at 4:04 PM
so good 🥮
always exceptions. but Im still not happy!
November 18, 2025 at 4:03 PM
always exceptions. but Im still not happy!
It looked better in 4K
November 13, 2025 at 6:16 PM
It looked better in 4K
Reposted by Kelvin Lau @klausenhauser on 🙅♂️🐘🪡
For-profit journals are supposed to improve research quality, yet they're perversely incentivized to churn out whatever they can monetize. This was happening before AI (see Strain: bit.ly/43gJPUM), and AI will make it worse.
It's insane that we volunteer our time to help them do so.
4/n
It's insane that we volunteer our time to help them do so.
4/n
November 11, 2025 at 11:52 AM
For-profit journals are supposed to improve research quality, yet they're perversely incentivized to churn out whatever they can monetize. This was happening before AI (see Strain: bit.ly/43gJPUM), and AI will make it worse.
It's insane that we volunteer our time to help them do so.
4/n
It's insane that we volunteer our time to help them do so.
4/n
Ever since this observation, still haven’t found a pattern
And here it is! Base1 DNA sequence was derived from a protein sequence derived from a deposited, but unpublished PDB. It was codon optimized for Ecoli from two vendors (well only 1 at the start). Clearly, the Base1 sequence or its mutants show little expression from either vendor. Same conditions.
October 30, 2025 at 7:16 AM
Ever since this observation, still haven’t found a pattern
No. I don’t feel like I need to do more than I did before. Nor do I do this on any other similar social media. It should just be easy and works, without being intrusive or nasty.
October 24, 2025 at 3:34 PM
No. I don’t feel like I need to do more than I did before. Nor do I do this on any other similar social media. It should just be easy and works, without being intrusive or nasty.
That sounds slightly less scientific :) hallucination has a balance of just random but also truth. Kind of like real dreams
October 24, 2025 at 7:15 AM
That sounds slightly less scientific :) hallucination has a balance of just random but also truth. Kind of like real dreams
This is most important when I’m scrolling and want to stumble on to interesting papers and findings.
I also understand the moment they make more of an algorithm it’ll be gamed
I also understand the moment they make more of an algorithm it’ll be gamed
October 23, 2025 at 10:36 PM
This is most important when I’m scrolling and want to stumble on to interesting papers and findings.
I also understand the moment they make more of an algorithm it’ll be gamed
I also understand the moment they make more of an algorithm it’ll be gamed
I think everyone has their reasons and brain sanity and safety is one of the most important. Personally I just don’t find the critical mass nor the type of ephemeral (yet algo forced) connections. Here you have to look to connect vs being fed potential connections.
October 23, 2025 at 10:35 PM
I think everyone has their reasons and brain sanity and safety is one of the most important. Personally I just don’t find the critical mass nor the type of ephemeral (yet algo forced) connections. Here you have to look to connect vs being fed potential connections.
I think the word itself already gives negative vibes, but red flag level?
October 23, 2025 at 10:26 PM
I think the word itself already gives negative vibes, but red flag level?
Most useful paper I read today
October 21, 2025 at 5:41 PM
Most useful paper I read today