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.

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.

plDDT scores are not reliably comparable between methods. You cannot assume a plDDT of 90 from, e.g., Boltz is identical in its trustworthiness to a plDDT of 90 from AlphaFold3.

Highlights of Model Quality Assessment in CASP16 Model quality assessment (MQA) remains a critical component of structural bioinformatics for both structure predictors and experimentalists seeking to use predictions for downstream applications. In ...

Our paper on the performance of ModFOLDdock2 at #CASP16 onlinelibrary.wiley.com/doi/10.1002/...

We used IntFOLD, and MultiFOLD to model interactions between Cx62 and calpain-1: "Inter-platelet communication driving thrombus formation is regulated by extracellular calpain-1 cleavage of connexin 62" | Haematologica haematologica.org/article/view...

Prediction and quality assessment of protein quaternary structure models using the MultiFOLD2 and ModFOLDdock2 servers Abstract. Understanding the structures of protein complexes is pivotal for breakthroughs in health, agriculture, bioengineering, and beyond. MultiFOLD2 and

Our new paper has been published: Prediction and quality assessment of protein quaternary structure models using the MultiFOLD2 and ModFOLDdock2 servers url: academic.oup.com/nar/article/...

Prediction and quality assessment of protein quaternary structure models using the MultiFOLD2 and ModFOLDdock2 servers Abstract. Understanding the structures of protein complexes is pivotal for breakthroughs in health, agriculture, bioengineering, and beyond. MultiFOLD2 and

Our new paper is out: Prediction and quality assessment of protein quaternary structure models using the MultiFOLD2 and ModFOLDdock2 servers
academic.oup.com/nar/advance-...

Inter-platelet communication driving thrombus formation is regulated by extracellular calpain-1 cleavage of connexin 62 - PubMed Connexin (Cx) gap junction proteins are expressed by a multitude of cells and function as plasma membrane hemichannels or dock to form intercellular communication tunnels. Whilst Cx43 has garnered con...

Pleased to share that our latest study has been published in Hematologica!

@rdgplateletlabs.bsky.social

pubmed.ncbi.nlm.nih.gov/40045892/

Have protein-ligand co-folding methods moved beyond memorisation? Deep learning has driven major breakthroughs in protein structure prediction, however the next critical advance is accurately predicting how proteins interact with other molecules, especially small mo...

Excited to share our latest preprint evaluating AlphaFold3, Boltz-1, Chai-1 and Protenix for predicting protein-ligand interactions, featuring our newly introduced benchmark dataset 🌹Runs N’ Poses🌹!

www.biorxiv.org/content/10.1...

🧵👇 (1/n)

CASP16 group picture - in Punta Cana

Models can be viewed interactively on desktop or mobile browsers in any orientation, similar to ModFOLDdock2.

Our MultiFOLD2 server is also now freely available to all. It's the top-ranked server on the hardest domain targets in #CASP16 by GDT_TS. With integrated stoichiometry prediction it outperforms AF3 on common multimer subsets in CAMEO. www.reading.ac.uk/bioinf/Multi...

Results can be viewed on desktop or mobile browsers in portrait or landscape.

The ModFOLDdock2 server is now online and freely available to all. The ModFOLDdock2 variants ranked within the top few groups at #CASP16 in QMODE1/2 metrics. ModFOLDdock2 is the best method for predicting the interface accuracy of modelled protein complexes.
www.reading.ac.uk/bioinf/ModFO...

ModFOLDdock2 is the top ranked method for predicting the interface accuracy of modelled protein complexes #CASP16. The server is available here: www.reading.ac.uk/bioinf/ModFO...

My group members, Shaima and @recep44.bsky.social, presented their posters on MultiFOLD2 for multimer modelling and FunFOLD5 for predicting protein-ligand interactions at #CASP16. Lots of interest and positive feedback!

Yesterday at #CASP16 I presented our new ModFOLDdock2 servers for the quality assessment of multimer models. It's great to see progress in the field but there are still many challenges ahead!