1/6 One of the key features of functional proteins is their inherent structural flexibility. In our recent work at @icmlconf, we introduce flexibility to protein structure design. Our model FliPS generates protein structures with custom per-residue flexibilities as validated by

RT @xiaofei_lin: BindCraft is now published in @Nature! . The open-source pipeline for de novo protein binder design achieves significantly….

RT @MartinPacesa: Wait what now? De novo binder in human?.

RT @ferruz_noelia: I’ve thoroughly enjoyed reading two (VERY!) recent papers that model protein sequences by retrieving evolutionary inform….

RT @workshopmlsb: Stay tuned for details on the 6th edition of MLSB, officially happening this December in downtown San Diego, CA!.

@icmlconf 6/6 Amazing team work with great co-authors @LeifSeute, Nicolas Wolf, Simon Wagner, @arneelof, @JanStuehmer and @graeter_mbm. Try out FliPS and BackFlip yourself, code and Google Colab tutorials are available on GitHub!.

@icmlconf 5/6 We introduce a framework in which we generate candidate protein structures conditioned on flexibility with FliPS and use BackFlip to select the candidates whose predicted flexibility profile best matches the target before running expensive MD simulations.

@icmlconf 4/6 We also introduce BackFlip - an equivariant graph neural network that can accurately predict backbone flexibility as derived from MD simulations in less than a second per protein. Crucially, BackFlip infers flexibility solely from the backbone geometry without requiring

@icmlconf 3/6 In a series of experiments, we demonstrate that FliPS samples novel, realistic proteins with diverse secondary structure composition and a remarkable resemblance to custom target flexibility profiles, as verified in 300ns Molecular Dynamics (MD) simulations of the designed

@icmlconf 2/6 FliPS is a conditional flow matching model for protein structure generation. The model receives a flexibility profile as conditional input feature and learns how to generate realistic protein structures while respecting target flexibilities.

finally!! did the alnTM-score mistake myself. .

RT @TheYotg: The Geometric Deep Learning textbook: Chapter 5 - Graphs. In multiple branches of science, from sociology to particle physics,….

ICML travel grants for PhD students in the US whose funding was cut: .

I finally won a lottery and got the CA visa?! Something must be wrong. Guess that means me + ICML 2025!

RT @RouskinLab: At 15, I left all my family and friends and came to the US , ALONE, seeking a country where MERIT and HARD WORK mattered-no….

RT @owl_posting: What could Alphafold 4 look like? (Sergey Ovchinnikov, Ep #3). 2 hours listening time . (links below). To those in the (ma….

RT @sokrypton: Any billionaires with no idea what to do w/ their 💰 money? This could be your chance to stop the death of science, technical….

RT @wellingmax: When I open the news every morning I am getting the feeling I am watching a bad version of Star Wars. And there is are more….

What in the world have we been watching yesterday….

RT @gklambauer: Extending the RANGE of Graph Neural Networks: Relaying Attention Nodes for Global Encoding. Virtual nodes introduced that p….